Your website isn't converting because of positioning, not design.

The contrarian take

If your website is converting below 2%, the problem is almost certainly not design. It's positioning. A redesign will give you a prettier version of the same problem and you'll be confused why the numbers didn't move.

The order of operations matters. Fix positioning first. Then write the copy. Then design the page. Skipping the first step is why so many redesigns disappoint.

The diagnostic question

April Dunford, in Obviously Awesome, gives the cleanest test for whether you have a positioning problem:

"Your current customers love you, but new prospects can't figure out what you're selling."

If that sentence describes your business, no amount of design polish will fix it. The site is faithfully communicating confused positioning. Fix the positioning and the design starts working.

Three other diagnostic signals:

• Your sales reps explain your product differently than your homepage does. The sales pitch that closes deals doesn't match the headline on the site.
• Your best customers describe what you do in words you don't use anywhere on the site. They've translated your fuzzy positioning into something concrete. You haven't caught up.
• Your competitors are growing faster despite having a worse product. They're winning on positioning, not product. You're losing the same way.

Any of these point to a positioning problem. Design can't fix any of them.

Why the design assumption is so common

The reason founders default to "we need a redesign" is structural.

1. Design is visible. A bad headline is invisible. A bad hero illustration is right there. The eye finds the visible problem, not the structural one.

2. Designers are easy to hire. Positioning consultants are rare and expensive. A designer can be hired on Upwork in an afternoon. So the available tool gets used, even when it's the wrong tool.

3. Redesigns produce a satisfying artifact. Positioning work produces a sentence. A redesign produces a beautiful site you can show people. The dopamine math favors the redesign.

4. Founders confuse their expression problem with their strategy problem. When the homepage isn't working, the assumption is that the words are wrong. Usually the thinking behind the words is what's wrong.

The result: founders spend $25,000 on a new homepage that converts at the same 1.4% the old one did. Then they blame the agency.

The four positioning failures that show up as design problems

Specific patterns I see over and over.

Failure 1: No specific audience

The headline says "For modern teams" or "For growth-stage companies." This isn't an audience, it's the absence of one.

A real audience is specific enough that the wrong person reads the headline and self-selects out. "For in-house legal teams at companies between 50 and 500 employees" is real. "For modern teams" is wallpaper.

The design symptom: the hero feels generic no matter what design treatment you apply.

The fix: name the audience in the hero. Lose the customers who aren't your real ICP. Your conversion rate will go up because the ones who stay are higher-intent.

Failure 2: Category confusion

Your category positioning doesn't match how customers think. Dunford's example: a company that called itself an "email automation platform" was losing deals because the real frame was "customer messaging", a different category with different buying criteria.

When you're in the wrong category, every feature comparison goes against you and you don't understand why. Customers are comparing you to competitors you didn't expect, because they slotted you into a different category than you intended.

The design symptom: the comparison table on your pricing page doesn't seem to land. The features that matter to you don't matter to buyers.

The fix: name your category deliberately. Pick the frame that makes your strengths look obvious. If you're "AI-powered project management" but your unique value is async coordination, you might really be in the "async collaboration" category against Loom and Slack, not in PM software against Asana and Linear.

Failure 3: Value confused with features

You list capabilities. "AI-powered. Real-time. Integrates with everything." What you don't list is what those capabilities do for the buyer's business.

Dunford: "We overestimate the buyer's ability to translate features to value for the business. You have advanced machine learning capabilities, so what? What does that capability enable for the customer's business?"

The design symptom: the feature grid on the homepage is dense and visually fine but visitors bounce. The features list isn't the problem. The "so what" is missing.

The fix: for every feature, write the outcome it produces. Replace "AI-powered insights" with "Find the 3 reasons your churn rate is climbing, in 30 seconds, not a quarter." Outcomes convert. Features don't.

Failure 4: Missing competitive frame

Your homepage acts like you have no competitors. No mention of alternatives. No comparison. No explicit positioning against the obvious incumbent.

The visitor's first question is "How is this different from [the thing I'm using now]?" If your homepage doesn't answer that, the visitor leaves to go answer it themselves, and most of them don't come back.

The design symptom: your bounce rate from the homepage to the pricing page is high. People are leaving to compare you to competitors before they even understand you.

The fix: name the alternatives. A comparison page. A "Why [your company] vs [competitor]" page. Even better, a comparison table on the homepage itself. Linear does this against Jira. Fathom does it against Google Analytics. Cursor does it against VS Code + Copilot. Be brave enough to name the fight you're picking.

The Dunford diagnostic

If your homepage could be a competitor's homepage with just the logo swapped, you have a positioning problem. The fix isn't visual. The fix is to be specific about who you're for, what category you're in, what value you deliver, and who you're competing against.

What a positioning fix actually looks like

The work isn't long. It's painful but it's not long.

Step 1: List five customers who love you

Five named customers. Real ones. Not theoretical buyers.

Step 2: For each, write down what they used before you

What were they doing about this problem before? A spreadsheet? A competitor? A consultant? Doing nothing? This is your real competitive set, not the one in your deck.

Step 3: For each, write the one specific outcome they get from your product

Not capability. Outcome. "They cut their close time from 14 days to 3." Not "They use our pipeline automation."

Step 4: Find the pattern

Look at the five customers. What's common about who they are, what they used before, and what outcome they get? That pattern is your positioning.

Step 5: Test the positioning against your homepage

Read your hero. Does it describe the audience you found? The competitive frame? The outcome? If not, the homepage is wrong. Rewrite the hero.

Step 6: Then redesign

Now design can do its job. You've given the designer something to express. They're no longer guessing at positioning while pushing pixels.

A real positioning shift can take a week. The website rebuild that follows takes 3–6 weeks. Founders consistently invert this, they redesign for two months and never do the positioning week.

When the problem really is design

To be fair, sometimes the issue is genuinely design. Three signals:

1. Your homepage looks like 2017. Old fonts, gradients from 2015, hero illustrations of diverse people at laptops. Visual age signals competence age. Real fix.
2. Your mobile experience is broken. 79–82% of SaaS traffic is mobile. If your mobile site is a busted desktop site, you're losing half your conversions to layout.
3. Page speed is terrible. Sub-2-second LCP is table stakes in 2026. Anything over 3 seconds tanks conversion mechanically, regardless of positioning.

If your positioning is sharp and one of these three is true, then yes, design fix. But this is usually 20–30% of cases. The other 70% are positioning, and the design fix won't move the needle.

The honest test

Try this. Write your one-sentence positioning. The sentence should answer:

• Who is this for, specifically?
• What category is it in?
• What outcome does it produce?
• What's the alternative they'd use otherwise?

If you can't write that sentence in 15 minutes, you have a positioning problem. If you can write it, but it doesn't appear anywhere on your homepage, you have a positioning problem that hasn't yet hit the site.

In either case, fixing the website starts with fixing the sentence. The pretty version of the sentence comes later.

The reading list

Three books that handle the positioning question better than any blog post can:

1. April Dunford, Obviously Awesome. The five components of positioning, the 10-step process. Required reading.
2. April Dunford, Sales Pitch. How positioning flows into the sales conversation. The link between homepage and pipeline.
3. Geoffrey Moore, Crossing the Chasm. The classic on early-market positioning, still relevant 35 years later.

Read these before you commission a redesign. Or even better: read the first one this weekend, redo your one-sentence positioning, and only then talk to a designer.