<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Aedan Evangelista: Blog</title>
    <link>https://www.aedanevangelista.com/blog/</link>
    <atom:link href="https://www.aedanevangelista.com/blog/feed.xml" rel="self" type="application/rss+xml" />
    <description>Notes on UI/UX design for founders, plus travel, food and self improvement.</description>
    <language>en</language>
    <lastBuildDate>Wed, 18 Nov 2026 00:00:00 GMT</lastBuildDate>
    <item>
      <title>Best text for a call to action button (not Submit)</title>
      <link>https://www.aedanevangelista.com/blog/best-text-for-call-to-action-button/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/best-text-for-call-to-action-button/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>The best text for a call to action button names what happens after the click. See rewrites of Submit, Click here and Learn more.</description>
      <content:encoded><![CDATA[<p>Someone fills in your contact form. Name, email, a few lines about their project. Then they reach the button, and it says &quot;Submit&quot;.</p>
<p>Submit what, and to whom? Will they get a reply, a call, a newsletter they never asked for? The button doesn't say, so they have to guess. Some people guess and click. Some people leave.</p>
<p>Here's the fix. A button should name the result of clicking it. That's usually a verb plus an object in two to four words: &quot;Send message&quot;, &quot;Book a call&quot;, &quot;Create account&quot;. Labels like &quot;Submit&quot;, &quot;Click here&quot; and &quot;Learn more&quot; make people guess what happens next.</p>
<h2>What should a button say instead of &quot;Submit&quot;?</h2>
<p>Say what the person gets or what happens when they press it. Look at what the form does, then write that on the button.</p>
<ul>
<li>A contact form sends a message. The button says &quot;Send message&quot;.</li>
<li>A quote form asks for a price. The button says &quot;Get my quote&quot;.</li>
<li>A signup form makes an account. The button says &quot;Create account&quot;.</li>
<li>A booking form reserves a time. The button says &quot;Book a call&quot;.</li>
</ul>
<p>Here's a before and after for a contact form.</p>
<p><strong>Before:</strong> the heading says &quot;Contact us&quot;. The button says &quot;Submit&quot;. After the click, the page reloads and nothing else happens.</p>
<p><strong>After:</strong> the heading says &quot;Tell me about your project&quot;. The button says &quot;Send message&quot;. Under it, one small line: &quot;I reply within a day.&quot;</p>
<p>The form has the same fields. The second one tells you what you're doing and what comes next. If you're unsure about the fields themselves, the post on <a href="https://www.aedanevangelista.com/blog/placeholder-vs-label-in-forms/">placeholder vs label in forms</a> covers the boxes above the button.</p>
<h2>Why is &quot;Submit&quot; a problem?</h2>
<p>It describes what the computer does. It says nothing about what the person gets.</p>
<p>&quot;Submit&quot; is a developer's word. The form submits data to a server. That's true, and no visitor cares.</p>
<p>The visitor is at the last step. They've done the work of filling everything in, and now they have to commit. A vague button gives them a reason to stop and wonder.</p>
<p>There's a second problem. &quot;Submit&quot; fits every form on the internet, so it tells you nothing about this one. If you can paste a label onto any button on any site and it still works, it's too generic.</p>
<p>A quick test: cover everything on the page except the button. Can you still tell what will happen when you press it? &quot;Send message&quot; passes. &quot;Submit&quot; doesn't.</p>
<h2>What's wrong with &quot;Click here&quot; and &quot;Learn more&quot;?</h2>
<p>They describe the click or promise nothing specific, so the reader has to read around them to know where they go.</p>
<p>&quot;Click here&quot; tells people how to use a mouse. They know. The useful part is where the click leads, and that's the part you left out.</p>
<p>It's also a problem for people who use a screen reader. The W3C's guidance on <a href="https://www.w3.org/WAI/WCAG22/Understanding/link-purpose-in-context.html" target="_blank" rel="noopener">link purpose</a> notes that assistive technology can give users a list of the links on a page. Pulled out of the page like that, a row of &quot;Click here&quot; and &quot;Learn more&quot; sounds like the same link repeated ten times.</p>
<p>&quot;Learn more&quot; is a little better, but learn more about what? Name the thing.</p>
<ul>
<li>&quot;Click here&quot; becomes &quot;Download the price list&quot;.</li>
<li>&quot;Learn more&quot; becomes &quot;See pricing&quot; or &quot;Read the case study&quot;.</li>
<li>&quot;Get started&quot; becomes &quot;Start free trial&quot;.</li>
<li>&quot;Continue&quot; becomes &quot;Go to payment&quot;.</li>
</ul>
<h2>How long should button text be?</h2>
<p>Two to four words is the usual range. Go longer only when the extra words remove a doubt.</p>
<p>One word is rarely enough. &quot;Send&quot; is fine in a chat app where the context is obvious. On a form, &quot;Send message&quot; is clearer and costs you one word.</p>
<p>Four or five words can work when the short version would leave a question open. &quot;Send my quote request&quot; is longer than &quot;Submit&quot;, and it's better, because it tells you that you're asking for a quote and nobody is charging you yet.</p>
<p>Past five words, the button starts doing a sentence's job. Move the extra detail to a small line under the button instead. &quot;Create account&quot; on the button, &quot;Free, no card needed&quot; under it.</p>
<p>Start with the verb. &quot;Book a call&quot; reads faster than &quot;Call booking&quot;.</p>
<h2>How do I write good button text for my own site?</h2>
<p>Finish the sentence &quot;I want to...&quot; from the visitor's side, and put the ending on the button. This is the routine I'd use on any page.</p>
<ul>
<li><strong>Finish the sentence.</strong> &quot;I want to... book a call.&quot; &quot;I want to... see pricing.&quot; If the ending sounds odd, the label is wrong. Nobody says &quot;I want to submit&quot;.</li>
<li><strong>Match the page that comes next.</strong> If the button says &quot;See pricing&quot;, the next page should be the pricing page, with &quot;Pricing&quot; at the top. A label that promises one thing and delivers another costs you trust.</li>
<li><strong>Use the same words everywhere.</strong> If the header button says &quot;Book a call&quot;, the one at the bottom of the page shouldn't say &quot;Schedule a consultation&quot;. People will wonder if those are two different things.</li>
<li><strong>Name the scary ones clearly.</strong> A delete dialog with &quot;OK&quot; and &quot;Cancel&quot; makes people stop and reread. &quot;Delete project&quot; and &quot;Keep project&quot; are much harder to misread.</li>
<li><strong>Answer the worry under the button.</strong> One short line is enough: &quot;No card needed&quot; or &quot;Takes two minutes&quot;. Only write it if it's true.</li>
</ul>
<h2>What does this look like on a whole site?</h2>
<p>The same rule everywhere: the button names the result. Here's a made-up example.</p>
<p>Say you run a bookkeeping service. Your homepage has a header button that says &quot;Get started&quot;, a hero button that says &quot;Learn more&quot;, and a contact form at the bottom that ends in &quot;Submit&quot;. Three buttons, three vague promises.</p>
<p>Rewrite them around what a visitor actually wants. The header and hero buttons both become &quot;Book a free call&quot;, so people know there's one way in and what it costs them. The pricing link becomes &quot;See monthly prices&quot;. The form button becomes &quot;Send message&quot;, with &quot;I reply within a day&quot; under it.</p>
<p>Nothing about the design changed. The colours, the layout and the fields are the same. But every button now answers the question a visitor has right before they click: what happens next? If the page still isn't getting enquiries after that, the post on <a href="https://www.aedanevangelista.com/blog/why-is-my-landing-page-not-converting/">why your landing page is not converting</a> looks at the rest of it.</p>
<p>If you'd like a second pair of eyes on your landing page, buttons included, <a href="https://www.aedanevangelista.com/#contact">book a call</a>. Send me the link and I'll tell you what I'd change first.</p>
]]></content:encoded>
    </item>
    <item>
      <title>Consistent icon style in UI: why mixed sets look cheap</title>
      <link>https://www.aedanevangelista.com/blog/consistent-icon-style-in-ui-design/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/consistent-icon-style-in-ui-design/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>Mixed icon sets make a UI look stitched together. How to spot it, why emoji fail as icons, and how to pick one family and weight.</description>
      <content:encoded><![CDATA[<p>You've fixed the colors. You've fixed the spacing. The app still looks a bit cheap, and you can't say why.</p>
<p>Look at your icons. If they came from different sets, they have different stroke widths, corners and sizes, so the screen looks stitched together. The fix is one icon family, one weight for most of the UI and one grid size. Change weight only on purpose, like Fill for the active tab.</p>
<h2>Why do mixed icon sets make a UI look cheap?</h2>
<p><strong>Short answer:</strong> Every icon set is drawn with its own rules, and your eye catches the clash even when you can't name it.</p>
<p>An icon set is a small design system. Someone picked a line thickness, how round the corners are, how much empty space sits around each shape and how much detail each drawing gets. Every icon in the set follows those rules, so they look like siblings.</p>
<p>Take icons from three sets and you get three sets of rules on one screen.</p>
<p>Picture a common sidebar. The Home icon has a thin 1.5px line and rounded ends. The Settings gear below it has a 2px line and sharp corners. The Reports icon is a solid shape. The Billing icon came from a set drawn on a 20px grid, so it looks smaller than the others even inside the same box.</p>
<p>Nobody looks at that and thinks &quot;the stroke widths don't match&quot;. They think &quot;this feels like a side project&quot;. And they decide that before they read a word on the screen.</p>
<p>Now picture the same sidebar with four icons from one family. Same line, same corners, same size. Nothing else changed, and it looks like one team built it.</p>
<h2>How can I tell if my icons don't match?</h2>
<p><strong>Short answer:</strong> Put them in one row at the same size and check line thickness, corners, fill and how big each one looks.</p>
<p>Take a screenshot of every screen and crop out the icons. Line them up. Then check:</p>
<ul>
<li><strong>Line thickness.</strong> Are some lines thicker than others?</li>
<li><strong>Corners and line ends.</strong> Rounded on some, square on others?</li>
<li><strong>Filled or outlined.</strong> Is there a solid icon sitting next to outlined ones for no reason?</li>
<li><strong>Visual size.</strong> Do some look bigger or smaller inside the same box?</li>
<li><strong>Detail.</strong> Is one a simple shape while the next has tiny details you can't read at 20px?</li>
</ul>
<p>The usual suspects are the icons that got added later. The first ten came with the template or the UI kit. The eleventh wasn't in that set, so someone grabbed one from somewhere else. Repeat that for a year and you have four sets in one product.</p>
<h2>Can I just use emoji as icons?</h2>
<p><strong>Short answer:</strong> In a chat message, sure. In your product's navigation, no.</p>
<p>Emoji are tempting. They're free, colorful and already on your keyboard. But they cause problems.</p>
<p>You don't control how they look. Apple, Google and Microsoft each draw their own emoji. Your app will look different on an iPhone, an Android phone and a Windows laptop.</p>
<p>You can't style them the way you style an icon. An emoji keeps its own colors. You can't set it to your brand color for the active item and a quiet gray for the rest.</p>
<p>And each emoji is a small full-color picture. Eight of them in a sidebar compete with your actual content.</p>
<p>Before: a sidebar with a house emoji, a bar chart emoji, a credit card emoji and a gear emoji. After: the same four items with line icons in one gray, and the active one in your brand color. The second one looks like something people pay for.</p>
<h2>Which icon weight should I use, and when?</h2>
<p><strong>Short answer:</strong> Pick one weight for almost everything, and use a second weight only when it means something.</p>
<p>Many icon families come in more than one weight. <a href="https://github.com/phosphor-icons/homepage">Phosphor</a>, for example, has six: Thin, Light, Regular, Bold, Fill and Duotone. They aren't meant to be mixed freely. A sensible split looks like this:</p>
<ul>
<li><strong>Regular</strong> for most of the UI: navigation, buttons, table actions.</li>
<li><strong>Fill</strong> for the selected state. In a bottom nav, the active tab gets Fill and the other tabs stay Regular. The change tells people where they are.</li>
<li><strong>Bold</strong> when icons are small or sit next to bold text, so they don't look thin beside it.</li>
<li><strong>Thin and Light</strong> for large icons, like an empty state, where Regular starts to look heavy.</li>
<li><strong>Duotone</strong> for a few feature spots, like a pricing page. Not for navigation.</li>
</ul>
<p>The test is simple. Can you say why this icon has a different weight? If the answer is &quot;it looked nice&quot;, change it back.</p>
<p>Do the same with size. Pick one size for navigation and one for icons inside text and buttons, and stick to them.</p>
<h2>What should I look for in an icon family?</h2>
<p><strong>Short answer:</strong> One grid, a few strict drawing rules, enough icons to cover your product and a license that allows commercial use.</p>
<p>Good families publish their rules. <a href="https://lucide.dev/guide/design/icon-design-guide">Lucide's design guide</a> says every icon is drawn on a 24 by 24 pixel canvas with 2px strokes. Rules like that are why hundreds of icons from different contributors still look like one set.</p>
<p>Check four things before you commit:</p>
<ul>
<li><strong>Coverage.</strong> Does it have every icon on your list, including the odd ones like an invoice, a receipt or a specific file type?</li>
<li><strong>Weights.</strong> If you want a filled active state, does every icon come in both outline and fill, drawn from the same shape?</li>
<li><strong>Brand logos.</strong> If you show &quot;Sign in with Google&quot; or social links, check whether the family includes logos in the same style, or plan where they'll come from.</li>
<li><strong>License.</strong> Read it once. Make sure you can use the icons in a product you sell.</li>
</ul>
<p>When weights are made from the same drawing, swapping Regular for Fill on the active tab works. Nothing jumps or changes shape. Any well-made family does the job. The point is to pick one.</p>
<h2>How do I fix the icons in a product that's already live?</h2>
<p><strong>Short answer:</strong> List every icon, pick one family, swap them all in one pass and write the rule down.</p>
<ul>
<li><strong>List them.</strong> Go screen by screen and note every icon and what it stands for.</li>
<li><strong>Pick one family</strong> that covers the whole list, including brand logos if you show any. Check that the license allows commercial use.</li>
<li><strong>Swap them in one pass.</strong> A half-done swap gives you two styles on screen at once.</li>
<li><strong>Set the defaults.</strong> One weight, one size, one color for inactive icons.</li>
<li><strong>Write the rule down</strong> where your team will see it: &quot;Icons come from this family, Regular, 24px. Fill only for the active state.&quot;</li>
</ul>
<p>It's a small job next to a redesign, and it changes how the whole product feels.</p>
<p>If your product looks a bit off and you can't tell why, <a href="https://www.aedanevangelista.com/#contact">book a call</a>. Show me a few screens and we'll go through what to fix first. I reply within a day.</p>
]]></content:encoded>
    </item>
    <item>
      <title>How to design a website without content (no lorem ipsum)</title>
      <link>https://www.aedanevangelista.com/blog/design-a-website-without-content/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/design-a-website-without-content/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>Copy not written yet? You can still start designing. Use rough draft copy instead of lorem ipsum. Here&#39;s how, step by step.</description>
      <content:encoded><![CDATA[<p>The pages are agreed. The designer is ready. Then someone asks for the copy, and there isn't any. The About page is a blank doc and the headline is &quot;something about trust&quot;.</p>
<p>So can design start? Yes. You can start designing a website before the copy is written, as long as you design around rough draft words and never around lorem ipsum. Final copy can wait. Words of some kind can't.</p>
<p>Here's how to do that, whether you're the owner or the designer.</p>
<h2>Can you design a website before the copy is written?</h2>
<p><strong>Short answer:</strong> Yes. Start with rough draft copy. Don't start with empty boxes.</p>
<p>The usual advice is &quot;content first&quot;, and the idea is right: a layout holds words. But if content first means waiting for finished, approved copy, a small business site can sit for weeks before anyone opens a design file.</p>
<p>The other extreme is worse. The designer draws a hero with a tidy five-word fake headline and two lines of lorem ipsum. Everyone approves it. Then the real headline shows up at fourteen words, wraps to four lines on a phone and pushes the button off the first screen. That section gets designed a second time.</p>
<p>Here's the difference, using a made-up bookkeeping business:</p>
<ul>
<li><strong>Before:</strong> &quot;Lorem ipsum dolor sit amet&quot; in a box sized for five words.</li>
<li><strong>After:</strong> &quot;Bookkeeping for small clinics. Your books closed by the 5th of every month.&quot; It's rough and it may be wrong, but it's the right length and it's about the right thing.</li>
</ul>
<h2>What do you do first when there's no copy?</h2>
<p><strong>Short answer:</strong> Give each page one job, then list its sections in order. Do it in a plain doc before you open a design tool.</p>
<figure>
  <img src="https://www.aedanevangelista.com/assets/images/blog/design-a-website-without-content-one-job.webp" alt="A home page plan with one job, book a call, and five sections in order" width="1548" height="1040" loading="lazy" decoding="async" />
</figure>
<p>For each page, write one line: what should a visitor do here? Build each page around one action, like booking a call or signing up. &quot;Home: book a call. Pricing: start the free plan.&quot; That's enough.</p>
<p>Then outline the page as the questions a visitor asks, in the order they ask them:</p>
<ol>
<li>What is this, and is it for me?</li>
<li>How does it work?</li>
<li>Why should I believe you?</li>
<li>What does it cost?</li>
<li>What do I do next?</li>
</ol>
<p>No sentence is written yet, but this outline is already content. You know the hero needs a headline and one button. You know section three needs proof, so someone has to go find a real review. A designer can work from that today.</p>
<p>It also tells you what to leave out. If a section doesn't answer one of the visitor's questions, it doesn't need designing. Cutting a section at this stage costs a line in a doc. Cutting it after it's designed costs a day.</p>
<h2>What do you put in the design instead of lorem ipsum?</h2>
<p><strong>Short answer:</strong> Draft copy you write yourself, in plain words, at about the real length.</p>
<figure>
  <img src="https://www.aedanevangelista.com/assets/images/blog/design-a-website-without-content-lorem-ipsum.webp" alt="Lorem ipsum fits the hero, but the real 14-word headline pushes the button off the screen" width="1548" height="1040" loading="lazy" decoding="async" />
</figure>
<p>Some people call it proto-content. It doesn't have to be good writing. Write what each section has to say, the way you'd text it to a friend:</p>
<ul>
<li><strong>Headline:</strong> &quot;We do the books for small clinics so you don't have to.&quot;</li>
<li><strong>Line under it:</strong> &quot;Fixed monthly price. Books closed by the 5th.&quot;</li>
<li><strong>Button:</strong> &quot;Book a 15-minute call&quot;</li>
</ul>
<p>That takes ten minutes. And it does something lorem ipsum can't do: people react to it.</p>
<figure>
  <img src="https://www.aedanevangelista.com/assets/images/blog/design-a-website-without-content-ugly-draft.webp" alt="A rough draft headline with the owner's handwritten correction" width="1548" height="1040" loading="lazy" decoding="async" />
</figure>
<p>The owner reads the draft and says &quot;we don't only do clinics&quot;, or &quot;nobody cares about the 5th, they care about tax season&quot;. That's the real copy starting to appear, in the very first review.</p>
<p>Lorem ipsum gets a nod. A wrong draft gets corrected, and the correction is the point.</p>
<h2>Where do the draft words come from if you're not a writer?</h2>
<p><strong>Short answer:</strong> From the owner's mouth, and from words the business has already written somewhere else.</p>
<figure>
  <img src="https://www.aedanevangelista.com/assets/images/blog/design-a-website-without-content-owners-words.webp" alt="A line the owner said on a call becomes the hero headline" width="1548" height="1040" loading="lazy" decoding="async" />
</figure>
<p>Start with a 20-minute call. Ask if you can record it. Then ask:</p>
<ul>
<li>What do you say when someone asks what you do?</li>
<li>Who is your best customer, and what were they stuck on before they found you?</li>
<li>What do people ask before they buy?</li>
<li>Why do people pick you over the other option?</li>
</ul>
<p>Write the headline from their words. Owners often say it better out loud than they write it. On a call they'll say &quot;we close your books by the 5th so you stop dreading month end&quot;. In a doc, the same person writes &quot;comprehensive financial solutions&quot;.</p>
<p>Then go looking for words that already exist:</p>
<ul>
<li><strong>Customer reviews.</strong> The phrases customers repeat are your proof section, and sometimes your headline.</li>
<li><strong>Sales emails and proposals.</strong> The pitch is already written in there.</li>
<li><strong>Replies to customer questions.</strong> That's your FAQ.</li>
<li><strong>Competitor sites.</strong> Look at which sections they have and how long each one runs. Don't copy their sentences.</li>
</ul>
<p>If you're the owner and there's no designer yet, do this part alone. Record a voice memo answering the four questions, then paste the best sentences into the outline.</p>
<h2>How do you stop the layout breaking when the real copy arrives?</h2>
<p><strong>Short answer:</strong> Design for the longest version from the start.</p>
<p>Real copy tends to run longer than the neat version in your head. So test every text block with a short version and a long one:</p>
<ul>
<li>A headline at 5 words and at 14 words.</li>
<li>A button that says &quot;Start&quot; and one that says &quot;Book a 15-minute call&quot;.</li>
<li>A testimonial of one line and one of five lines.</li>
<li>A name like &quot;Jo Li&quot; and a name like &quot;Maria Cristina de los Santos&quot;.</li>
</ul>
<figure>
  <img src="https://www.aedanevangelista.com/assets/images/blog/design-a-website-without-content-long-version.webp" alt="The same hero on two phones, with a 5-word and a 14-word headline, both fitting" width="1548" height="1040" loading="lazy" decoding="async" />
</figure>
<p>Check all of them at phone width. If the layout survives the long version, a copy change later is a text edit. If it only works with the short version, every copy change is a redesign.</p>
<p>The hero is where this bites hardest. For what that first screen needs once the words exist, see <a href="https://www.aedanevangelista.com/blog/what-to-put-in-a-landing-page-hero-section/">what to put in a landing page hero section</a>.</p>
<h2>What can be finished before any copy exists?</h2>
<p><strong>Short answer:</strong> Everything that stays the same no matter what the words say.</p>
<figure>
  <img src="https://www.aedanevangelista.com/assets/images/blog/design-a-website-without-content-finish-first.webp" alt="Type, colour, buttons, grid and nav, all done before any copy" width="1548" height="1040" loading="lazy" decoding="async" />
</figure>
<ul>
<li><strong>Type.</strong> Fonts, sizes and line height.</li>
<li><strong>Colour.</strong> The palette, and whether text passes contrast on each background.</li>
<li><strong>Grid and spacing.</strong> Column widths and the gaps between sections.</li>
<li><strong>Components.</strong> Buttons, form fields, cards and the footer.</li>
<li><strong>Navigation.</strong> The page list and the menu labels. These are short and you know them early.</li>
</ul>
<p>None of it needs an approved sentence. Settle these while the owner works on the words, and the two jobs run side by side instead of one after the other.</p>
<h2>How do you keep draft copy from going live?</h2>
<p><strong>Short answer:</strong> Mark every draft line, so nobody mistakes it for final.</p>
<p>Good draft copy has one risk. It looks real. Someone approves the page, and a headline written in ten minutes ends up on the live site.</p>
<figure>
  <img src="https://www.aedanevangelista.com/assets/images/blog/design-a-website-without-content-draft-tags.webp" alt="Draft lines tagged DRAFT, and a search before launch that finds none left" width="1548" height="1040" loading="lazy" decoding="async" />
</figure>
<p>A few habits stop that:</p>
<ul>
<li>Keep one copy doc as the source, with a status on every section: draft, owner checked or final.</li>
<li>Tag draft text in the design file. A small &quot;DRAFT&quot; label or a highlight colour works.</li>
<li>Never make up numbers or customer quotes, even in a mockup. Write &quot;[number here]&quot; or &quot;[real quote from a customer]&quot; instead. A fake testimonial looks real very fast.</li>
<li>Before launch, search the file for the tag. There should be none left.</li>
</ul>
<p>Copy not written and the site can't wait? <a href="https://www.aedanevangelista.com/#contact">Book a call</a>. Bring whatever you have, even if it's only notes. I reply within a day.</p>
]]></content:encoded>
    </item>
    <item>
      <title>How many pricing tiers should a SaaS have? Fewer</title>
      <link>https://www.aedanevangelista.com/blog/how-many-pricing-tiers-should-a-saas-have/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/how-many-pricing-tiers-should-a-saas-have/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>How many pricing tiers should a SaaS have? Fewer than you think. When two plans are enough, when three make sense, and when to add one.</description>
      <content:encoded><![CDATA[<p>You're building a pricing page and it keeps growing. There's a Starter plan, a Pro plan, a Team plan, and an Enterprise column with a &quot;Contact us&quot; button because every other SaaS seems to have one. Now you have four columns, a table with twenty rows, and you can't tell which plan you would pick yourself.</p>
<p>Here's the short version. Fewer plans are clearer. Three plans with one marked as recommended is a good ceiling, and it's also the most common setup: close to half of the companies in <a href="https://chartmogul.com/blog/saas-pricing-report/">ChartMogul's analysis of over 600 pricing pages</a> offer three tiers. A small tool with one kind of customer can do fine with two.</p>
<h2>How many pricing tiers should a small SaaS have?</h2>
<p><strong>Short answer:</strong> Two or three. Pick the smallest number that covers the customers you have today.</p>
<p>I think of every plan as a question you ask the visitor: &quot;Are you this kind of customer?&quot; One plan, one question. Four plans, four questions, and the visitor has to answer all of them before they can press a button.</p>
<p>If you sell to one kind of customer, you need one paid plan. Add a free one if people need to try the product before they trust it. That's two.</p>
<p>Three makes sense when you have clearly different groups. A solo user, a small team and a company do not need the same things, so each gets a column.</p>
<p>Here's a before and after.</p>
<p><strong>Before:</strong> four plans called Starter, Plus, Pro and Business. The only difference is a limit: 5 projects, 15 projects, 50 projects, unlimited. To choose, the visitor has to guess how many projects they'll have next year.</p>
<p><strong>After:</strong> two plans. Free does the core job. Pro adds the things you want once the work gets serious. The visitor answers one question: &quot;Do I need those extras yet?&quot;</p>
<h2>Why do fewer plans make the decision faster?</h2>
<p><strong>Short answer:</strong> Every plan you add means more comparisons for the visitor, and the number grows faster than the columns do.</p>
<p>With two plans there is one comparison: this one or that one. With three plans there are three pairs to compare. With four there are six. You doubled the columns and got six times the comparisons.</p>
<p>This is choice overload, one of the <a href="https://www.aedanevangelista.com/blog/5-laws-of-ux/">5 Laws of UX</a>. When people face too many choices, they often pick none.</p>
<p>On a pricing page, picking none looks like this. The visitor doesn't choose the wrong plan. They close the tab to &quot;think about it&quot;.</p>
<p>So the goal of the page is a fast, confident choice. Cutting a column is the cheapest way I know to get there.</p>
<h2>When are two plans enough?</h2>
<p><strong>Short answer:</strong> When you have one kind of customer, and one kind of customer needs one paid plan.</p>
<p>Say you sell a scheduling tool for solo hair stylists. Every customer is one person with one chair. They all need the same thing: a booking page and reminders.</p>
<p>Two plans cover that:</p>
<ul>
<li><strong>Free:</strong> a booking page and up to a set number of bookings a month. Enough to see if it works.</li>
<li><strong>Pro:</strong> a flat monthly price. It adds unlimited bookings, text reminders and your own logo on the page.</li>
</ul>
<p>Free does the core job. Pro is for when the business is busy enough to need more. There is one comparison to make, and the visitor can make it in a few seconds.</p>
<p>The same thinking applies to the product itself. A tool for one kind of customer usually does fewer things well. A short feature list makes a short pricing page.</p>
<h2>When does a third plan earn its place?</h2>
<p><strong>Short answer:</strong> When a real group of customers needs something different. More of the same thing is only a limit, and a limit can live inside a plan you already have.</p>
<p>Good reasons to add a plan:</p>
<ul>
<li>A group keeps asking for something the others don't need, like several people sharing one account.</li>
<li>The buyer is different. A manager paying for five people decides differently from a freelancer paying for one.</li>
<li>That group costs you more to serve, so the price has to change with it.</li>
</ul>
<p>Weak reasons to add a plan:</p>
<ul>
<li>Your competitor has three.</li>
<li>The page looks empty with two columns.</li>
<li>You want a plan nobody is meant to buy, so the middle one looks cheap. Everyone still has to read that column.</li>
</ul>
<p>In the stylist example, a Team plan would earn its place the day salons with five chairs start asking for one shared calendar. Until then, it's a column nobody needs.</p>
<h2>Should I mark one plan as recommended?</h2>
<p><strong>Short answer:</strong> Yes when you have three. With two, the layout usually does it for you.</p>
<p>With three plans, pick the one that fits most of your customers and make it stand out. Give it a border, a &quot;Recommended&quot; label and the only filled button. The other two get outlined buttons. Now the visitor has a default, and they only read the other columns if the default feels wrong.</p>
<p>Only use a label like &quot;Most popular&quot; if it's true.</p>
<p>With two plans you don't need a badge. Decide which action you want more, and give that plan the filled button. If you want signups first, make Free the loud one. If you want paid users, make Pro the loud one.</p>
<h2>What should the plans be called and how should I show them?</h2>
<p><strong>Short answer:</strong> Plain names, a visible price, and only the differences.</p>
<p>Call the plans what they are. Free, Pro and Team are clearer than Spark, Boost and Orbit. Nobody should have to learn your vocabulary to see the price.</p>
<p>Show the price as a number. If you have monthly and yearly billing, show both, like &quot;$12 a month or $120 a year&quot;.</p>
<p>Then list only what changes between plans.</p>
<p><strong>Before:</strong> two columns with 14 rows of checkmarks each. Eleven of the rows are the same in both.</p>
<p><strong>After:</strong> Free lists what you get. Pro says &quot;Everything in Free, plus&quot; and then three lines.</p>
<p>The second version is shorter, and the visitor can see what the money buys without playing spot the difference.</p>
<p>If your pricing page has grown a column too many, <a href="https://www.aedanevangelista.com/#contact">book a call</a> and send me the link to your current page. I'll help you cut it down to one clear choice. I reply within a day.</p>
]]></content:encoded>
    </item>
    <item>
      <title>How to build with AI without losing quality</title>
      <link>https://www.aedanevangelista.com/blog/how-to-build-with-ai-without-losing-quality/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/how-to-build-with-ai-without-losing-quality/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>AI output drifts once you ask for more than a few things. Set plain rules first, check every output against them, and keep the final yes.</description>
      <content:encoded><![CDATA[<p>You tried AI on your product, and the first few results looked great. Then you asked for fifty more. Now the buttons don't match, the icons look like they came from five different sets, and you spend your evenings fixing things one by one.</p>
<p>The fix is to set the rules before the AI makes anything, then check every output against them. AI handles the volume. A person sets the rules and decides what ships.</p>
<p>Here's how that split works, and how to use it on your own product.</p>
<h2>Why does AI work get messy after the first few outputs?</h2>
<p><strong>Short answer:</strong> each output is made on its own, so small differences pile up until nothing matches.</p>
<p>Ask for one icon and you get a good icon. Ask for a second and that one is good too. Put them side by side and the lines are a bit thicker on one and the corners a bit rounder on the other. Neither is wrong. They just weren't made to the same rules.</p>
<p>With ten outputs you can fix that by eye. With hundreds you can't. And it's the same story with landing page sections, product screens and emails. Each one looks fine alone. Together they look like five people made them.</p>
<p>Here's the difference between two ways of asking:</p>
<ul>
<li><strong>Before:</strong> &quot;Draw me a settings icon.&quot; You get a settings icon in whatever style came out that time.</li>
<li><strong>After:</strong> &quot;Draw a settings icon on a 24px grid with a 1.5px stroke, round corners and 2px of padding inside the grid.&quot; You get an icon that can sit next to the others.</li>
</ul>
<p>The second request is longer and less fun to write. It's also the one that still works at number 500.</p>
<p>There's a second reason it drifts. When you leave a gap, the tool fills it with its most common default. That's why so many AI-built sites look alike. I cover the visible signs in <a href="https://www.aedanevangelista.com/blog/why-does-my-ai-app-look-generic/">why does my AI app look generic</a>.</p>
<h2>What rules should I set before the AI starts?</h2>
<p><strong>Short answer:</strong> the few things that must be the same in every output, written so plainly that a script could check them.</p>
<p>Say you run a small online shop and want AI to help build product pages and an icon set. A first rule list might be:</p>
<ul>
<li><strong>One grid.</strong> Every icon sits on the same 24px grid.</li>
<li><strong>One stroke.</strong> Every line is 1.5px, with round ends.</li>
<li><strong>One set of colours.</strong> Text, background and one accent. Nothing else.</li>
<li><strong>One type scale.</strong> Five heading and body sizes, used everywhere.</li>
<li><strong>One spacing scale.</strong> Gaps of 4, 8, 16, 24, 32 or 48px, and nothing in between.</li>
</ul>
<p>A good test for a rule is whether you can answer it with yes or no. &quot;Make it look clean&quot; fails that test. You and the AI will both read it differently every time. &quot;Every icon sits on the same 24px grid&quot; passes. It's either true or it isn't.</p>
<p>Keep the list short. Five rules that are always followed do more than thirty that are followed sometimes.</p>
<p>Write the rules down in a file the tool reads every time, not in a chat message you'll forget to paste. Design teams call these values design tokens: named colours, sizes and spacing that every screen pulls from. Once the values live in one place, the AI picks from your list instead of inventing a new shade of grey.</p>
<h2>What should AI do, and what should a person decide?</h2>
<p><strong>Short answer:</strong> let AI help with the volume. Keep the rules and the final yes for a person.</p>
<p>AI is good at the repetitive part. Drawing the fortieth variation, filling in names and search tags, turning one approved layout into ten pages, writing the code for a component you've already designed.</p>
<p>The calls a person should make:</p>
<ul>
<li><strong>The rules.</strong> Grid, colours, type, spacing. AI can suggest options, but someone picks.</li>
<li><strong>The naming.</strong> What each thing is called, so people can find it. An icon you can't find is an icon you can't use.</li>
<li><strong>Exceptions.</strong> Brand logos, special pages and anything that is allowed to break a rule, and why.</li>
<li><strong>What's missing.</strong> Look at the whole set and ask what a user will need that isn't there yet.</li>
<li><strong>The final yes.</strong> Someone reviews what goes out.</li>
</ul>
<p>My default split is simple: the tool proposes and a person decides. If nobody owns the final yes, the tool's defaults end up owning it.</p>
<h2>How do I check hundreds of outputs without losing my weekends?</h2>
<p><strong>Short answer:</strong> let a script catch the rule breaks, and save your eyes for judgment.</p>
<p>A lint step is an automatic check. It runs on every file and flags anything that breaks a written rule. The name comes from <a href="https://en.wikipedia.org/wiki/Lint_(software)">lint</a>, a code checker Stephen C. Johnson wrote at Bell Labs in 1978. For design work, a check can look for colours outside your palette, spacing values outside your scale, or icon strokes that aren't 1.5px. <a href="https://stylelint.io/">Stylelint</a> can do some of this for CSS.</p>
<p>This changes what review is for.</p>
<ul>
<li><strong>Without the check:</strong> review means hunting for mistakes. Is this line the right thickness? Does this one sit on the grid? Multiply that by hundreds of files and it never ends.</li>
<li><strong>With the check:</strong> the script answers those questions. You only look at what a script can't judge. Does this icon read as the thing it's named after? Would a customer understand this page?</li>
</ul>
<p>Review catches problems one at a time. A rule with a check stops the same problem from showing up again.</p>
<p>If you can't write a script, a checklist works. Run it on every output, every time, and answer each line yes or no.</p>
<h2>Does this work if I'm not making icons?</h2>
<p><strong>Short answer:</strong> yes. It works for anything you make in volume.</p>
<p>Say you're using AI to build pages for your site. Rules you could set first:</p>
<ul>
<li>One main button per page, and it always looks the same.</li>
<li>One set of heading sizes, used everywhere.</li>
<li>One set of spacing steps, and nothing in between.</li>
</ul>
<p>Then check each new page against that list before it goes live. A checklist you run by hand is fine to start with. The point is that the check happens every time, and the answer is yes or no.</p>
<p>The same goes for emails, social graphics and product screens. Pick the few things that must never change, write them so a stranger could check them, and check them.</p>
<p>And if you're hiring someone who builds with AI, ask two questions. What are your rules? How do you check them? If the answer is &quot;I look it over&quot;, expect the fifty-icon problem from the top of this post.</p>
<p>Want a site or app built fast that still looks like one person made it? <a href="https://www.aedanevangelista.com/#contact">Book a call</a> and tell me what you're making. I reply within a day.</p>
]]></content:encoded>
    </item>
    <item>
      <title>How to design a SaaS dashboard people actually use</title>
      <link>https://www.aedanevangelista.com/blog/how-to-design-a-good-saas-dashboard-ui/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/how-to-design-a-good-saas-dashboard-ui/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>Start a SaaS dashboard from the one question people open it to answer. What to put first, what to move a click away, and which widgets to cut.</description>
      <content:encoded><![CDATA[<p>You shipped a dashboard with nine charts. It looked great in the demo. Now people log in, scroll past all of it, click the same menu item they always click, and get on with their day.</p>
<p>That's the usual sign that the dashboard was built to fill a screen. People open it to answer one question. So start from that question, put the answer first, and move everything else a click away. Borrow patterns people already know, like folders, and don't fill space with charts nobody reads.</p>
<h2>What should go at the top of a SaaS dashboard?</h2>
<p>The answer to the one question people log in to ask, in words or one number, with the next action right beside it.</p>
<p>Every product has that question. For a booking tool it's &quot;who is coming in today?&quot; For a project tool it's &quot;what's late?&quot; For an invoicing tool it's &quot;who do I need to bill?&quot;</p>
<p>Write the question down before you draw anything. Then design the top of the page as its answer.</p>
<p>Before: a project tool opens on a donut chart of tasks by status, a line chart of hours logged and a bar chart of tasks per person. To find out what's late, you read three charts and work it out yourself.</p>
<p>After: the page opens with &quot;3 tasks are overdue&quot;. Under it are those three tasks, each with the owner and the due date. One click opens the task. The question is answered before you scroll.</p>
<p>If you can't agree on the one question, ask the people who use it. Five short calls with real customers usually settle it faster than a week of internal debate.</p>
<h2>How many charts does a dashboard need?</h2>
<p>As many as answer a question someone actually asks. For most small products that's one or two, and sometimes none.</p>
<p>A chart earns its place when the shape matters. Is revenue going up or down? Is this week unusual? A line can show that faster than a number.</p>
<p>A chart is filler when a sentence says the same thing. A donut with six slices that tells you &quot;most tasks are in progress&quot; could be one short line of text.</p>
<p>A quick test for each chart: what would someone do differently after looking at it? If the answer is nothing, it's decoration.</p>
<h2>Which dashboard widgets can I remove?</h2>
<p>Start with the five below. Each one takes space near the top and answers no question.</p>
<ul>
<li><strong>The big welcome banner.</strong> &quot;Welcome back, Sam&quot; with an illustration, taking up a third of the screen. A small greeting is fine. A banner pushes the real content down the page.</li>
<li><strong>The all-time total.</strong> A number that only goes up, like total invoices ever sent. It feels good for a week. After that it tells you nothing about today.</li>
<li><strong>The quick-links grid.</strong> Six tiles that repeat what's already in the sidebar. Now there are two menus to learn.</li>
<li><strong>The &quot;recent activity&quot; feed of everything.</strong> Every edit, login and comment in one stream. It's too long to read. If some events matter, show only those and call them what they are, like &quot;needs your reply&quot;.</li>
<li><strong>The donut chart with many slices.</strong> Hard to compare, easy to replace with a short list sorted from biggest to smallest.</li>
</ul>
<p>The one to keep at the top is the answer to the main question, with its next action.</p>
<p>You don't have to delete the rest. Move it to a Reports page. The people who want the numbers will find them there, and everyone else gets a faster start.</p>
<h2>How should I organise everything else?</h2>
<p>Use a structure people already know. Folders are the safest one, because almost everyone has used a file explorer.</p>
<p>Most of a product is lists of things: clients, projects, orders, files. The temptation is to invent a clever way to browse them. Tags, boards, a custom view with its own rules.</p>
<p>Each new pattern is something the user has to learn before they can do any work. <a href="https://lawsofux.com/jakobs-law/" target="_blank" rel="noopener">Jakob's Law</a> puts it simply: people spend most of their time in other products, so they expect yours to work the same way. A folder needs no lesson. You open it and the related things are inside.</p>
<p>Before: one long table of every invoice for every client, with five filters above it. To see one client's history, you set two filters and hope you picked the right ones.</p>
<p>After: a list of clients. Open a client and everything for that client is there.</p>
<p>The same goes for smaller patterns. A search box at the top. Settings behind a gear or your profile picture. The newest items first. People bring those habits from other apps they use, so you get them for free. There's more on this in <a href="https://www.aedanevangelista.com/blog/5-laws-of-ux/">the 5 Laws of UX</a>.</p>
<h2>What should I leave out of the product?</h2>
<p>Anything that would bring its own screens and statuses without serving the main question. Here's a made-up example.</p>
<p>Say you're building a small tool for freelance photographers to send invoices. The main question is &quot;I need to bill this client.&quot; Every client is a folder. Pick one, and the business details, bank details and client address are already filled in because they were saved once.</p>
<p>Then come the feature requests: drafts, estimates, payment reminders, expense tracking. Each one is reasonable. Each one also adds a list, a status and a reason to claim a spot on the first screen. Add all four and your clean dashboard turns back into the nine-chart wall.</p>
<p>Saying no is a trade. Some people will want those features. But a smaller product with a clear first screen is easier to use than a bigger one where you have to hunt. If you do add a feature later, give it its own page and keep the top of the dashboard for the main question.</p>
<h2>How do I check if my dashboard works?</h2>
<p>Show it to someone for a few seconds, hide it, and ask what they would do next. If they can't say, the top of the page is wrong.</p>
<p>You can run this with a teammate or a customer on a call. Don't explain anything first. If they describe the charts (&quot;there was a blue graph&quot;) but can't name an action, the page is showing data without an answer.</p>
<p>Then watch where people click first after logging in. If most of them leave the dashboard right away for the same page, that page is your real dashboard. Bring its key content to the front.</p>
<p>If your dashboard has grown into a wall of charts, <a href="https://www.aedanevangelista.com/#contact">book a call</a>. Show me the screen and tell me what your users log in to do. I'll tell you what I'd put first and what I'd move out of the way.</p>
]]></content:encoded>
    </item>
    <item>
      <title>Placeholder vs label in forms: which should you use?</title>
      <link>https://www.aedanevangelista.com/blog/placeholder-vs-label-in-forms/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/placeholder-vs-label-in-forms/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>Placeholders vanish when people type, so they can&#39;t check their answers. Why every form field needs a visible label, and when a hint is fine.</description>
      <content:encoded><![CDATA[<p>Someone is halfway through your form. They click the next box, start typing, and the gray hint is gone. Was this one for their name or the company name? They delete what they typed to see the hint again. Or they guess. Or they close the tab.</p>
<p>The fix is simple. Use a visible label above every field, and keep placeholders for a short format hint at most. Placeholder text vanishes when people type, it often fails contrast and it is not a reliable label for screen readers.</p>
<h2>What's the difference between a label and a placeholder?</h2>
<p><strong>Short answer:</strong> A label sits outside the field and stays. A placeholder sits inside the field and disappears when you type.</p>
<p>The label says what the field is: &quot;Email&quot;. The placeholder shows an example of what goes in it: &quot;name@company.com&quot;.</p>
<p>The trouble starts when the placeholder does the label's job. It looks clean in a mockup, because in a mockup every field is empty. Your visitors only see the form that way at the very start. After that they're looking at a form with their own typing in it and nothing else.</p>
<h2>Why do placeholder-only forms cause mistakes?</h2>
<p><strong>Short answer:</strong> Once someone types, the form stops telling them what each box is for.</p>
<p>Walk through it as the person filling it in.</p>
<p>Before: a form with four boxes. Inside them, in gray: &quot;Your name&quot;, &quot;Business name&quot;, &quot;Name on bank account&quot;, &quot;Email&quot;. No labels. It looks tidy.</p>
<p>They type their name in the first box. The hint is gone. They type the business name. That hint is gone too. Then their phone buzzes.</p>
<p>They look back. The third box has the cursor in it and no hint. Was this the business name again, or the name on the account? The only way to find out is to delete what they typed.</p>
<p>It gets worse at the end. Before sending, careful people check their answers. In this form there are four answers and no questions. &quot;Maria Santos&quot; and &quot;Maria L. Santos&quot; sit in two boxes, and nothing says which is which.</p>
<p>Browser autofill adds to it. It can fill all four boxes at once, sometimes in the wrong places, and the person never sees a single hint.</p>
<p>After: the same four fields with a label above each one. The labels stay, empty or filled. There's nothing to remember and nothing to guess.</p>
<h2>Aren't placeholders cleaner?</h2>
<p><strong>Short answer:</strong> They look emptier, but they aren't easier. A label adds one line per field and removes the guessing.</p>
<p>Placeholder-only forms are popular because the form looks shorter. It isn't. It asks the same number of questions. If your form feels long, cut a field you don't need. Don't hide the labels.</p>
<p>There's also a contrast problem. Placeholder text is light gray on purpose, so it doesn't look like a real answer. <a href="https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html">WCAG</a>, the standard for web accessibility, asks for a contrast ratio of at least 4.5:1 for normal text. Many placeholder grays are lighter than that, so they're hard to read with weaker eyesight or on a phone in sunlight. Make the gray darker and the field can look filled in already, so people skip it. You lose either way.</p>
<p>Then screen readers. A placeholder is not a dependable label for people who use them, and the <a href="https://html.spec.whatwg.org/multipage/input.html#the-placeholder-attribute">HTML standard</a> itself says the placeholder attribute should not be used as an alternative to a label. A real label also gives you something for free: tap the label and the cursor jumps into the field, which helps on a phone.</p>
<h2>When is a placeholder OK?</h2>
<p><strong>Short answer:</strong> As a short format example next to a real label, and only if the form still works without it.</p>
<p>Good uses:</p>
<ul>
<li>Label &quot;Email&quot;, placeholder &quot;name@company.com&quot;.</li>
<li>Label &quot;Website&quot;, placeholder &quot;https://&quot;.</li>
<li>A single search box with a search icon or button beside it. It still needs a label in the code for screen readers.</li>
</ul>
<p>Bad use: rules. &quot;At least 8 characters with one number&quot; inside the password box vanishes at the exact moment someone needs it. Put rules under the field as help text that stays.</p>
<p>Here's a quick test. Delete every placeholder in your form. Can someone still fill it in? If yes, your placeholders are hints and you're fine. If no, they're doing a label's job.</p>
<h2>What about floating labels?</h2>
<p><strong>Short answer:</strong> Better than placeholder-only, but still not as clear as a label that sits above the field from the start.</p>
<p>A floating label starts inside the box and shrinks to the top edge when you click in. It stays visible, which solves the main problem.</p>
<p>It has costs. The shrunken label is often very small. The empty field can still look filled in. And it leaves little room for a format hint. My default is the plain version: label above, box below.</p>
<h2>How do I lay out labels in a real form?</h2>
<p><strong>Short answer:</strong> Put the label above the field and close to it, with a bigger gap before the next field.</p>
<p>This is the Law of Proximity, one of the <a href="https://www.aedanevangelista.com/blog/5-laws-of-ux/">5 Laws of UX</a>. Keep related things close together. A common beginner mistake is the same gap everywhere, so a label floats halfway between two boxes and you can't tell which one it belongs to. For example, 6px between a label and its box and 20px between fields makes each pair read as one unit.</p>
<p>Above the field also works well on a phone. The eye runs straight down one line, and long labels have room.</p>
<p>Take most care with the forms people fill in once and rely on for a long time. Say you run a payroll tool. The bank details form is filled in once, then used for every payment after that. A number typed into the wrong box doesn't cause one mistake. It causes the same mistake every month. Those are the forms where a clear label above each field matters most, along with help text that stays and a review step before saving.</p>
<p>If people start your form and don't finish it, <a href="https://www.aedanevangelista.com/#contact">book a call</a>. Send me the link and we'll go through it field by field. I reply within a day.</p>
]]></content:encoded>
    </item>
    <item>
      <title>Web design trends 2027: which to use, which to skip</title>
      <link>https://www.aedanevangelista.com/blog/web-design-trends-2027/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/web-design-trends-2027/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>Which 2027 web design trends make a small business site clearer or faster, and which will date it. Each gets a verdict: use, careful or skip.</description>
      <content:encoded><![CDATA[<p>You're planning a new website for 2027. You open a few trend roundups, and now you want a glass navbar, a 3D hero and text that slides in as you scroll.</p>
<p>The worry is fair. You pay for all of it, and a year later the site looks like last year. So here's the short version: use a trend when it makes your site clearer or faster, and skip it when it's only decoration. Below, each trend gets a verdict: use it, use it carefully, or skip it.</p>
<h2>Which 2027 web design trends are worth using?</h2>
<p>The ones that help a visitor read and understand faster: big clear type, light pages, and real screenshots or photos.</p>
<p><strong>Big, clear type. Verdict: use it.</strong> A large headline is the cheapest way to make a page clear. It adds almost nothing to load time and it works on every screen.</p>
<p>Before: a small grey headline sitting on top of a busy photo. You have to squint to read it. After: one large headline in plain words that says what you sell and who it's for, on a calm background.</p>
<p>One warning. Big type makes a weak headline louder. &quot;Reimagine your workflow&quot; at 96 pixels is still vague. Fix the words first, then make them big. The post on <a href="https://www.aedanevangelista.com/blog/what-to-put-in-a-landing-page-hero-section/">what to put in a landing page hero section</a> goes through those words.</p>
<p><strong>Light, fast pages. Verdict: use it.</strong> Fewer scripts, fewer font files, smaller images. This is the quiet trend. Nobody will ever compliment it, and it will never look dated, because nobody sees it. They only notice that the page opened right away.</p>
<p><strong>Real screenshots and real photos. Verdict: use it.</strong> Show the actual product, the actual shop, the actual person. It looks plain next to a glossy render, and it answers the question a visitor came with: what do I get?</p>
<p><a href="https://linear.app/" target="_blank" rel="noopener">Linear's homepage</a> is a good example of the first and last of these together. A big plain headline says what the product is, and right under it you see the actual app: an issue, a backlog, an activity feed. No abstract shapes standing in for the product.</p>
<h2>Which trends should I use carefully?</h2>
<p>Bento grids, scroll animation and dark mode. Each one helps in the right place and hurts when it covers the whole site.</p>
<p><strong>Bento grids. Verdict: careful.</strong> A bento grid is a set of boxes in different sizes, like a lunch box. It works when you have a few features of similar weight and each one fits in a sentence.</p>
<p>It stops working when everything is a box. If the price, the main feature and a small detail all sit in boxes of the same weight, nothing looks more important than anything else. Check it on a phone too. There the grid turns into one long stack, and the order of that stack is the only layout left.</p>
<p><strong>Scroll animation. Verdict: careful.</strong> Small motion is useful. A button that reacts when you press it, or a panel that slides open, shows you what just happened.</p>
<p>The risky kind takes over the scroll. You move your finger and the page decides how far you go. Before: five full-screen scenes you have to sit through to reach the price. After: a normal page where one or two things move, and the price is a quick scroll away.</p>
<p>Also respect the &quot;reduce motion&quot; setting on people's devices. The W3C's guidance on <a href="https://www.w3.org/WAI/WCAG22/Understanding/animation-from-interactions.html" target="_blank" rel="noopener">animation from interactions</a> explains that motion can cause dizziness, nausea and headaches for people with inner ear disorders, and that the <code>prefers-reduced-motion</code> setting is how a site can tell they've asked for less.</p>
<p><strong>Dark mode. Verdict: careful.</strong> Dark is a fair choice if it suits your brand and the people you sell to. It's a weak choice if the only reason is that other sites in your space went dark this year. If you can offer both light and dark, do that. If you can only have one, pick the one your content reads best in.</p>
<h2>Which trends should I skip?</h2>
<p>Text on glass, AI-generated hero images, 3D scenes and custom cursors. They cost a lot and they put a date on the site.</p>
<p><strong>Text on blurred glass. Verdict: skip.</strong> The frosted panel looks good in a still mockup. On a live page, the background behind it changes as you scroll, so the contrast changes too. WCAG asks for a <a href="https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html" target="_blank" rel="noopener">contrast ratio of at least 4.5 to 1</a> for normal-size text. Text on glass can pass over one part of a photo and fail over the next. If you love the effect, keep it for a decorative panel with no text people need to read.</p>
<p><strong>AI-generated hero images. Verdict: skip.</strong> A visitor wants to see your product, your team or your work. A generated picture shows none of those. The style of these images also changes fast, so this year's look is easy to date next year. The post on <a href="https://www.aedanevangelista.com/blog/why-does-my-ai-app-look-generic/">why your AI app looks generic</a> covers the same problem across a whole interface.</p>
<p><strong>3D scenes and custom cursors. Verdict: skip for most small business sites.</strong> A 3D scene is usually a heavy file. On a mid-range phone on mobile data, it can be the reason the page is still blank. A custom cursor does nothing on a touch screen, and many of your visitors are on a phone.</p>
<h2>How do I tell if a trend will date my website?</h2>
<p>Take the trend away in your head. If the page still makes sense without it, it was decoration, and decoration is what dates. Run these checks before you add anything trendy:</p>
<ul>
<li><strong>Remove it.</strong> Does the page still say what you sell and what to do next? Then the trend isn't doing a job.</li>
<li><strong>Ask what it helps with.</strong> Can a visitor read, compare or buy faster because of it? If you can't name the task, skip it.</li>
<li><strong>Open it on an ordinary phone.</strong> Use mobile data, away from the office wifi. If the page waits on the effect, the effect goes.</li>
</ul>
<p>A trend is fine until it's the only idea on the page. A site with a clear offer and one trendy detail ages slowly. A site that is five trends stacked together looks old as soon as the next roundup comes out.</p>
<h2>Where should trends go on a site?</h2>
<p>In the parts that are cheap to change. Keep them out of the parts that are expensive to change.</p>
<p>Cheap parts are things like a hero image, an accent colour, or the style of an illustration. You can swap any of them in an afternoon without touching anything else.</p>
<p>Expensive parts are your navigation, your page structure and how your content is organised. If a trend shapes those, removing it later means rebuilding pages.</p>
<p>My default is a plain base with one small detail that feels current. Say you run a dental clinic. Keep the menu, the booking flow and the page layout boring and familiar. Then let a warmer accent colour or a new photo style carry the 2027 feel. When it starts to look old, you change one colour, not the whole site.</p>
<p>Planning a website for 2027 and not sure which ideas to keep? <a href="https://www.aedanevangelista.com/#contact">Book a call</a> and send me your list. I'll tell you which ones I'd use, which I'd tone down and which I'd skip.</p>
]]></content:encoded>
    </item>
    <item>
      <title>What to put in a landing page hero section: 5-second test</title>
      <link>https://www.aedanevangelista.com/blog/what-to-put-in-a-landing-page-hero-section/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/what-to-put-in-a-landing-page-hero-section/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>What goes in a landing page hero: an outcome headline, who it&#39;s for, the product, one button and real proof. Plus the five-second test.</description>
      <content:encoded><![CDATA[<p>Your landing page gets visits, but hardly anyone clicks. You know the product is good. The trouble is the first screen. A stranger lands, reads something like &quot;The future of work, simplified&quot;, can't tell what you sell, and leaves.</p>
<p>The first screen needs five things: one outcome headline, one line on who it's for, a look at the product, one main button you can see on a phone without scrolling, and real proof close by. If a stranger can't say what you do in five seconds, rewrite it.</p>
<p>Here's each part, with a before and after.</p>
<h2>What should the headline in my hero say?</h2>
<p><strong>Short answer:</strong> the outcome your buyer gets, in words they would use, followed by one line that says who it's for.</p>
<p>Most weak heroes open with a slogan. It sounds good in a meeting and tells a visitor nothing.</p>
<p>Take a made-up invoicing tool:</p>
<ul>
<li><strong>Before:</strong> &quot;Invoicing, reimagined.&quot;</li>
<li><strong>After:</strong> &quot;Send an invoice in two minutes. Your client pays from a link.&quot;</li>
</ul>
<p>The first could sit on any product in the category. The second tells you what you get.</p>
<p>Then add one line under it that names the buyer. &quot;For freelancers who bill a few clients a month.&quot; Now the right person knows they're in the right place, and the wrong person leaves without wasting your time or theirs.</p>
<p>If the words aren't written yet, don't stall the design. Draft a rough headline at the real length and fix it later. I explain how in <a href="https://www.aedanevangelista.com/blog/design-a-website-without-content/">how to design a website without content</a>.</p>
<h2>What should the hero image show?</h2>
<p><strong>Short answer:</strong> the product, or the result of using it.</p>
<p>The usual stock photo of people smiling at a laptop takes up half the screen and explains nothing. The image is the biggest thing on the first screen, so it should do some of the explaining.</p>
<ul>
<li><strong>Before:</strong> a stock photo of a team in a meeting room.</li>
<li><strong>After:</strong> a screenshot of the actual invoice your client will receive.</li>
</ul>
<p>If you sell a service, show the thing you deliver. A finished page, a report, a before and after. If the product isn't finished, a clean mockup of the main screen still beats a stock photo.</p>
<p>One check: cover the headline with your hand. Can you still guess what the business does from the image alone? If yes, the image is working.</p>
<h2>How many buttons should the hero have, and where do they go?</h2>
<p><strong>Short answer:</strong> one main button, visible on a phone without scrolling, with a label that says what happens next.</p>
<p>Build every page around one action, like booking a call or signing up. The hero is where that action first shows up.</p>
<ul>
<li><strong>Before:</strong> three buttons of equal weight. &quot;Learn more&quot;, &quot;Watch demo&quot;, &quot;Contact sales&quot;.</li>
<li><strong>After:</strong> one solid button, &quot;Start free&quot;, and at most one quiet text link beside it.</li>
</ul>
<p>Then open the page on your phone. On many sites the headline and image fill the screen and the button sits just below the edge. Shorten the headline or shrink the image until the button is in view.</p>
<p>Make it easy to tap too. The strictest accessibility guideline, <a href="https://www.w3.org/WAI/WCAG22/Understanding/target-size-enhanced.html">WCAG 2.2 Target Size (Enhanced)</a>, asks for targets of at least 44 by 44 CSS pixels. Your main button should clear that easily.</p>
<p>The label matters as well. &quot;Submit&quot; and &quot;Learn more&quot; don't say what you get. &quot;Book a 15-minute call&quot; does.</p>
<h2>What counts as real proof, and where does it go?</h2>
<p><strong>Short answer:</strong> something a visitor could check, placed right next to the button.</p>
<p>Proof answers the question &quot;why should I believe you?&quot; at the moment someone is deciding whether to click. So it belongs beside the button, not four screens down.</p>
<p>Real proof can be:</p>
<ul>
<li>A quote from a customer, with their name and company.</li>
<li>Logos of businesses you've actually worked with.</li>
<li>A number you can back up.</li>
</ul>
<p>If you're new and have none of that yet, don't make it up. Say what's true and lowers the risk. Say you run a new invoicing tool. &quot;Free plan. No card needed.&quot; and &quot;Your client pays without making an account&quot; are both checkable, and both tell a visitor that trying it costs nothing.</p>
<ul>
<li><strong>Before:</strong> &quot;Trusted by thousands of happy customers.&quot;</li>
<li><strong>After:</strong> one real quote with a name, or &quot;Free plan. No card.&quot;</li>
</ul>
<h2>How do I run the five-second test?</h2>
<p><strong>Short answer:</strong> show someone only your first screen for five seconds, hide it, and ask what you sell.</p>
<p>Pick a friend who doesn't know your business well. Open your site on your phone, hand it over, count to five, and take it back. Then ask three things:</p>
<ul>
<li>What does this company sell?</li>
<li>Who is it for?</li>
<li>What would you click?</li>
</ul>
<p>Don't explain or defend. Just listen. If they say &quot;some kind of software?&quot;, the headline isn't doing its job. If they can't say what they'd click, you have too many buttons or the main one is hidden.</p>
<p>It costs nothing and takes a minute. Run it again after every rewrite.</p>
<h2>What does a good hero look like in practice?</h2>
<p><strong>Short answer:</strong> a plain headline that says what the product is, and the real product right under it.</p>
<p>Look at <a href="https://linear.app/">Linear</a>, the project tool for software teams. The headline reads &quot;The product development system for teams and agents&quot;. The line under it says it is built for planning and building products. There's no slogan to decode.</p>
<p>Then, right under the words, you see the product itself: an issue view with its activity feed and settings panel. You don't have to read a feature list to understand what you'd be using. The screenshot does that job.</p>
<p>Two things to copy from it:</p>
<ul>
<li><strong>Name the category plainly.</strong> &quot;Product development system&quot; tells you what shelf it sits on.</li>
<li><strong>Let the product be the picture.</strong> A real screen beats any illustration of one.</li>
</ul>
<p>If your first screen fails the five-second test, start with those two. Then check the button and the proof. For pages that get visits but no clicks for other reasons, see <a href="https://www.aedanevangelista.com/blog/why-is-my-landing-page-not-converting/">why is my landing page not converting</a>.</p>
<p>Want a second pair of eyes on your first screen? <a href="https://www.aedanevangelista.com/#contact">Book a call</a> and send me your link. I reply within a day.</p>
]]></content:encoded>
    </item>
    <item>
      <title>Why does my AI app look generic? 5 fixes that work</title>
      <link>https://www.aedanevangelista.com/blog/why-does-my-ai-app-look-generic/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/why-does-my-ai-app-look-generic/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>AI tools reuse the same defaults, so sites look alike. Here are the tells that give yours away and five small fixes that make it yours.</description>
      <content:encoded><![CDATA[<p>You built your site with an AI tool over a weekend. It works. But something about it feels off, and you can't say what. Your visitors can't say it either. They only know it looks like a lot of other sites they opened this month.</p>
<p>The reason is simple. AI tools reuse the same defaults, so sites start to look alike. Adrian Krebs scored 1,590 Show HN pages against 16 AI design patterns, and 22% of them triggered four or more (<a href="https://adriankrebs.ch/blog/design-slop/">his design-slop study</a>). You can fix it with five small decisions: a written design file, one accent colour, a deliberate font, real copy and one icon set.</p>
<h2>Why does my AI-built site look like everyone else's?</h2>
<p><strong>Short answer:</strong> The tool fills every gap you leave with its safest default, and everyone gets the same defaults.</p>
<p>When you type &quot;build me a landing page for my booking app&quot;, you haven't said which colour, which font, how much spacing or what tone. The tool has to pick something. It picks what it has seen most. The next founder types a similar prompt and gets the same picks.</p>
<p>The tool can follow design rules well. It only needs to be given some.</p>
<ul>
<li><strong>Before:</strong> &quot;Build a landing page for my booking app.&quot;</li>
<li><strong>After:</strong> &quot;Build a landing page for my booking app. Follow the rules in my design file: one accent colour, this font, no gradients, icons from one set.&quot;</li>
</ul>
<p>The second prompt takes ten more seconds to write. The result looks like it belongs to someone.</p>
<h2>How can you tell a site was vibe coded?</h2>
<p><strong>Short answer:</strong> Look for the same handful of defaults stacked on one page. One on its own is fine. Four or more is the pattern.</p>
<p>These are on <a href="https://adriankrebs.ch/blog/design-slop/">Krebs's list of patterns</a>:</p>
<ul>
<li><strong>Inter in a centered hero.</strong> Fix: pick a font on purpose (more on that below).</li>
<li><strong>Purple as the main colour.</strong> Fix: choose one accent that means something for your product.</li>
<li><strong>Gradients on backgrounds, buttons and text.</strong> Fix: use flat colour. Keep a gradient only if you can say why it's there.</li>
<li><strong>A small badge above the headline.</strong> Fix: delete it. Let the headline be the first thing people read.</li>
<li><strong>Identical feature cards with an icon on top.</strong> Fix: show one feature with a real screenshot.</li>
<li><strong>Emoji used as icons in the menu or sidebar.</strong> Fix: use a real icon set.</li>
<li><strong>Untouched shadcn components.</strong> Fix: change the corner radius, the spacing and the colours so they match your rules.</li>
</ul>
<p>And three more I look for myself. These are my own additions, not from the study:</p>
<ul>
<li><strong>Copy that could sit on any site.</strong> &quot;Build faster. Scale smarter.&quot; Fix: say what the product does.</li>
<li><strong>Emoji sprinkled through headings and bullet points.</strong> Fix: remove them and let the words carry it.</li>
<li><strong>Icons from two or three different sets.</strong> Fix: one family, one weight.</li>
</ul>
<h2>What should I fix first?</h2>
<p><strong>Short answer:</strong> Write the design file first, because every other fix lives inside it.</p>
<p>A design file is a plain text file in your project that the AI tool reads before it builds anything. Most AI coding tools can load a rules file at the start of every session. It doesn't need to be long. A good one covers:</p>
<ul>
<li>The one accent colour and where it's allowed to appear</li>
<li>The font, and the sizes for headings and body text</li>
<li>The spacing steps, so gaps stay consistent</li>
<li>How a button looks</li>
<li>Which icon set to use, and at which weight</li>
<li>Words and patterns to avoid, like gradients or a badge above the headline</li>
</ul>
<p>Then fix things in this order: write the design file, cut down to one accent colour, choose the font, rewrite the copy, and swap the icons for one set.</p>
<p>Without the file, you fix the purple today and it comes back the next time you ask for a new page. For more on setting rules and checking AI output against them, see <a href="https://www.aedanevangelista.com/blog/how-to-build-with-ai-without-losing-quality/">how to build with AI without losing quality</a>.</p>
<h2>Do I really need just one accent colour?</h2>
<p><strong>Short answer:</strong> For most small sites, yes. One accent shows people where to click.</p>
<p>If five things on the page are bright, nothing stands out. If only the main button is, the eye goes straight to it.</p>
<p>Say you run a made-up invoicing app. You pick green, because green reads as money and the app is about getting paid. Green goes on the main button, the active tab and the &quot;paid&quot; status. Everything else is neutral text on a neutral background. That is the whole rule. It's short enough for a tool to follow and short enough for you to check on every screen.</p>
<ul>
<li><strong>Before:</strong> purple buttons, a blue link, a pink gradient headline and a green success badge on one screen.</li>
<li><strong>After:</strong> neutral text and backgrounds, and one colour that only shows up on the thing you want clicked.</li>
</ul>
<p>Check the accent against your background for contrast too. A pale accent on white might look calm and still be hard to read.</p>
<h2>Which font and icons should I pick?</h2>
<p><strong>Short answer:</strong> Any font you chose on purpose, and one icon family in one weight.</p>
<p>A common move is to swap Inter for whichever font is popular this month and call it done. It doesn't help much. Krebs's list flags trendy picks like Space Grotesk, Instrument Serif and Geist too. If the tool picked it for you, or you picked it because everyone else did, it's still a default. The name of the font matters less than the reason you chose it.</p>
<p>Pick for your brand's voice. A law firm and a kids' app shouldn't sound the same, and they shouldn't be set in the same type either. Write the choice into your design file so the tool stops reaching for its default.</p>
<p>Icons work the same way. Mixed sets are easy to spot because the line thickness and corner shapes change from one icon to the next. Pick any single, well-made set and stick to one weight. What matters is that every icon comes from the same family.</p>
<h2>How do I make the copy sound like my business?</h2>
<p><strong>Short answer:</strong> Replace every line that could be pasted onto a competitor's site.</p>
<p>Read your headline and ask if another company could use it word for word. If yes, it's a placeholder, even if it sounds polished.</p>
<ul>
<li><strong>Before:</strong> &quot;Build faster. Scale smarter.&quot;</li>
<li><strong>After:</strong> &quot;Book a dentist appointment without calling the clinic.&quot;</li>
</ul>
<p>The second one names the customer, the task and the annoying part it removes. A tool can't write that for you without being told, because it comes from knowing your customers. The same goes for the first screen as a whole. See <a href="https://www.aedanevangelista.com/blog/what-to-put-in-a-landing-page-hero-section/">what to put in a landing page hero section</a> for the rest of it.</p>
<p>If your site works but looks like everyone else's, <a href="https://www.aedanevangelista.com/#contact">book a call</a> and send me the link. I'll tell you which of these to fix first. I reply within a day.</p>
]]></content:encoded>
    </item>
    <item>
      <title>Why is my landing page not converting? 4 checks first</title>
      <link>https://www.aedanevangelista.com/blog/why-is-my-landing-page-not-converting/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/why-is-my-landing-page-not-converting/</guid>
      <pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>Before you redesign, run four checks in order: a clear message, one button, a small first step and a page that matches the ad.</description>
      <content:encoded><![CDATA[<p>People land on your page and leave. Nobody signs up, nobody books a call. The first thought is usually &quot;the design must be bad, I need a new one&quot;.</p>
<p>Maybe. But a redesign is the slowest and most expensive fix, so check four cheaper things first, in this order. Is the message clear in five seconds? Is there one obvious button? How much does the first step ask for? Does the page match the ad or link that brought people there? Each one only matters once the one before it passes.</p>
<h2>Can a stranger tell what you do in five seconds?</h2>
<p><strong>Short answer:</strong> If not, fix the headline before you touch anything else.</p>
<p>This is the five-second test. Show your page to someone who doesn't know your business. Count to five and hide the screen. Then ask two questions: what does this company sell, and who is it for?</p>
<p>If they can't answer, the page is unclear. That's a different problem from ugly, and a new color palette won't fix it.</p>
<ul>
<li><strong>Before:</strong> &quot;The future of smarter workflows.&quot;</li>
<li><strong>After:</strong> &quot;Scheduling software for dental clinics. Patients book online, your front desk stops answering the phone.&quot;</li>
</ul>
<p>The first headline could be about anything. The second tells you the product, the customer and the result. I go deeper on this in the post on <a href="https://www.aedanevangelista.com/blog/what-to-put-in-a-landing-page-hero-section/">what a landing page needs above the fold</a>.</p>
<p>Ask a friend outside your industry, not a colleague. Colleagues already know what you do, so they fill the gaps without noticing. A stranger can't, and that's the point.</p>
<h2>Is there one obvious button?</h2>
<p><strong>Short answer:</strong> Pick one action for the page and make its button the loudest thing on the screen.</p>
<p>A lot of pages ask for four things at once: Start free trial, Book a demo, Watch the video, Join the newsletter. All four in the same size and color. When every button shouts, people don't pick one. They scroll, and then they leave.</p>
<ul>
<li><strong>Before:</strong> four buttons in the hero, all filled, all the same color.</li>
<li><strong>After:</strong> one filled button that says &quot;Start free trial&quot;, and one quiet text link under it for &quot;Watch a 2-minute demo&quot;.</li>
</ul>
<p>A quick check: squint at your page until the text blurs. If you can't tell which button is the main one, your visitors can't either. This is visual hierarchy, the first of the <a href="https://www.aedanevangelista.com/blog/5-laws-of-ux/">5 Laws of UX</a> I check on any page.</p>
<p>Also look at the label. &quot;Submit&quot; and &quot;Learn more&quot; don't say what happens next. &quot;Get my free quote&quot; does. There's more on this in the post on the <a href="https://www.aedanevangelista.com/blog/best-text-for-call-to-action-button/">best text for a call to action button</a>.</p>
<h2>How much does your first step ask for?</h2>
<p><strong>Short answer:</strong> Ask only for what you need to start. Collect the rest later.</p>
<p>Someone clicked your button. That's a small yes. Now look at what you ask for in return. If the next screen wants a phone number, a company size, a job title and a budget before they've seen anything, the trade is no longer worth it to them.</p>
<ul>
<li><strong>Before:</strong> a 9-field signup form: full name, work email, phone, company, job title, company size, country, password, how did you hear about us.</li>
<li><strong>After:</strong> 3 fields: name, email, password. The rest is asked inside the product, when there's a reason for it.</li>
</ul>
<p>Go through each field and ask: do I use this answer in the first five minutes? If not, it can wait.</p>
<h2>Does the page match the ad or link that brought people there?</h2>
<p><strong>Short answer:</strong> The headline should repeat the promise people clicked on.</p>
<p>People arrive with a sentence already in their head. It's the ad, the search result or the post they clicked. If your page opens with a different sentence, they think they're in the wrong place and go back.</p>
<ul>
<li><strong>Before:</strong> the ad says &quot;Free invoice templates for freelancers&quot;. The page headline says &quot;The all-in-one finance platform for modern teams&quot;.</li>
<li><strong>After:</strong> the ad says &quot;Free invoice templates for freelancers&quot;. The page headline says &quot;Free invoice templates for freelancers. Pick one and send it today.&quot;</li>
</ul>
<p>If you run several ads with different promises, one page can't match them all. Make a version of the page for each promise. Often only the headline and the first image need to change.</p>
<h2>What does a small first step look like in practice?</h2>
<p><strong>Short answer:</strong> Let people get the value before you ask them for anything.</p>
<p>Say you run a bookkeeping service and your page offers a free tax checklist. One version asks for name, email, phone, company size and annual revenue before the download starts. Another shows the first page of the checklist right there, then asks for an email to send the full file.</p>
<p>The second one asks for less, and it asks after the visitor has seen that the checklist is worth having. That's the idea behind check three.</p>
<p>The same works for software. Let people try a demo or a sample project without an account. If there's a free plan, don't ask for a card to start it. Every wall you put before the value is a place where people turn back.</p>
<p>When you do need details, ask at the moment they make sense. A shipping address belongs in checkout, not in signup. A team size belongs in the step where people invite their team.</p>
<h2>When is a redesign the right call?</h2>
<p><strong>Short answer:</strong> When all four checks pass and the page still looks untrustworthy or breaks on a phone.</p>
<p>Sometimes the design is the problem. Text too small to read on mobile, a layout that falls apart on narrow screens, or a look so dated that people doubt the business is still running. Those are real reasons.</p>
<p>But run the four checks first. If the headline is vague, a new design will just be a nicer-looking page with a vague headline. And if you do end up redesigning, you'll hand your designer a clear message, one action and a short form. That gives the redesign a much better starting point.</p>
<p>If you've run the four checks and aren't sure what you're looking at, <a href="https://www.aedanevangelista.com/#contact">book a call</a>. Send me the page and I'll tell you what I'd fix first. I reply within a day.</p>
]]></content:encoded>
    </item>
    <item>
      <title>The 5 Laws of UX I use on every project</title>
      <link>https://www.aedanevangelista.com/blog/5-laws-of-ux/</link>
      <guid isPermaLink="true">https://www.aedanevangelista.com/blog/5-laws-of-ux/</guid>
      <pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate>
      <category>UI/UX Design</category>
      <description>The 5 Laws of UX I use on every project, each with an interactive demo you can switch from newbie to expert, plus where I used them in my own work.</description>
      <content:encoded><![CDATA[<p>The five UX rules I use on every project are visual hierarchy, choice overload, chunking, Jakob's Law and the Law of Proximity. They matter because people decide fast: if a page makes them stop and think, they leave.</p>
<p>Most of them come from the <a href="https://lawsofux.com/" target="_blank" rel="noopener">Laws of UX</a> collection by Jon Yablonski. Visual hierarchy isn't one of its named laws (it's a design principle), but it's the one I check first, so it goes first here.</p>
<p>Each law below comes with a small screen you can try for yourself. It starts the way a newbie usually builds it. Switch it to Expert to see what an expert changes.</p>
<h2>Visual hierarchy: what should people see first?</h2>
<p>Size, weight, colour and space tell the eye where to go. When everything is the same size, nothing stands out, and people don't know where to start.</p>
<figure><img src="https://www.aedanevangelista.com/assets/images/blog/laws-of-ux-v1-hierarchy.webp" alt="Newbie: a hero where every line and button is the same size and grey. Expert: one big headline, a blue line under it and one dark button." width="1548" height="1040" /></figure>
<p><strong>How I use it:</strong> I decide the one thing each screen is for before I design it. On my own homepage, my name and what I make come first, in the largest type, and each screen has one black button. Everything else is smaller and quieter.</p>
<p><strong>Check your site:</strong> squint at your homepage. If you can't tell what the biggest thing is, your visitors can't either.</p>
<h2>Choice overload: why fewer options sell more</h2>
<p>When people face too many choices, they often pick none. Six options that all look equal feel like work. Three options with one suggested feel easy.</p>
<figure><img src="https://www.aedanevangelista.com/assets/images/blog/laws-of-ux-v1-choice.webp" alt="Newbie: six identical pricing cards in a grid. Expert: three cards, with the middle one marked Most picked and given a blue button." width="1548" height="1040" /></figure>
<p><strong>How I use it:</strong> in <a href="https://www.aedanevangelista.com/projects/invoic/">Invoic</a>, the invoice tool I built, you pick a template by stepping through the designs as real previews of your invoice. You see one at a time instead of a long list of settings, so choosing feels quick.</p>
<p><strong>Check your site:</strong> count the buttons, plans or menu links on your first screen. If there are more than three or four, cut some, or mark the one you'd pick.</p>
<h2>Chunking: how to make long content easy to read</h2>
<p>People can only hold a few things in mind at once. Break long forms and pages into small groups, and the same amount of content feels shorter.</p>
<figure><img src="https://www.aedanevangelista.com/assets/images/blog/laws-of-ux-v1-chunking.webp" alt="Newbie: one long form with eight fields stacked together. Expert: a three-step progress bar and only three fields on the first step." width="1548" height="1040" /></figure>
<p><strong>How I use it:</strong> <a href="https://www.aedanevangelista.com/projects/aecon/">Aecon</a>, my icon family, has 3,318 icons. Shown as one long grid, that would be a wall. So the icons are sorted into groups, with search on top, and you only ever look at one small set at a time.</p>
<p><strong>Check your site:</strong> find your longest form or page. Can you split it into steps or sections with clear headings?</p>
<h2>Jakob's Law: why your site should work like others</h2>
<p>People spend most of their time on other sites. So they expect yours to work the same way: logo at the top left, menu at the top right, cart in the corner. Moving those around doesn't feel creative to visitors. It feels broken.</p>
<figure><img src="https://www.aedanevangelista.com/assets/images/blog/laws-of-ux-v1-jakob.webp" alt="Newbie: cart top left, logo top right, search floating in the middle and the menu at the bottom. Expert: logo top left, menu and cart top right, content below." width="1548" height="1040" /></figure>
<p><strong>How I use it:</strong> I save surprises for the content, not the controls. On this site, the logo sits at the top left and takes you home, and the menu is at the top right, like almost every site you know.</p>
<p><strong>Check your site:</strong> open it on your phone and try to get home, find the menu and contact you. If any of that takes a second look, move it to where people expect it.</p>
<h2>Law of Proximity: why spacing is a design decision</h2>
<p>Things that sit close together read as one group. When every gap on a page is the same, nothing looks related, and people have to work out the structure themselves.</p>
<figure><img src="https://www.aedanevangelista.com/assets/images/blog/laws-of-ux-v1-proximity.webp" alt="Newbie: a form where labels and boxes all have the same gap. Expert: each label sits close to its box, and the fields are grouped into Contact and Address." width="1548" height="1040" /></figure>
<p><strong>How I use it:</strong> in my case studies, the facts (role, year, live site) sit tight together as one block, with much more space before the next section. Small gaps inside a group, big gaps between groups.</p>
<p><strong>Check your site:</strong> look at your forms. Each label should be closer to its own box than to the box above it.</p>
<h2>A quick check for your homepage</h2>
<p>Run through these five questions on your own site. Each one takes less than a minute.</p>
<ol>
<li><strong>Hierarchy:</strong> is one thing clearly the biggest on the first screen?</li>
<li><strong>Choice:</strong> are there three or four options at most, with one suggested?</li>
<li><strong>Chunking:</strong> are long forms and pages split into small steps or sections?</li>
<li><strong>Jakob's Law:</strong> are the logo, menu and contact where people expect them?</li>
<li><strong>Proximity:</strong> do labels sit close to what they describe, with bigger gaps between groups?</li>
</ol>
<p>If you answered no to two or more, your site is probably making visitors think harder than they should. That's usually fixable without a full redesign.</p>
]]></content:encoded>
    </item>
  </channel>
</rss>
