In this article
- First: no traffic, or traffic that doesn’t convert?
- The nine friction points, in the order readers hit them
- 1. Category ambiguity
- 2. Value opacity
- 3. Unproven superlative
- 4. Proof mismatch
- 5. Missing metric
- 6. Price opacity
- 7. Gated tier
- 8. CTA hesitation
- 9. Commitment cliff
- Why the order matters more than the list
- Diagnostic summary
- Getting an outside read
- What this diagnosis can’t tell you
- Frequently asked questions
- Why is nobody signing up from my landing page?
- How do I know if it’s my page or my traffic?
- How many people do I need to test my landing page?
- Should I rewrite the whole page or change one thing?
- Does a low conversion rate always mean the page is broken?
If nobody signs up, the cause is almost always one of nine specific friction points, and they occur in a predictable order as a reader moves down the page: category ambiguity, value opacity, unproven superlative, proof mismatch, missing metric, price opacity, gated tier, CTA hesitation and commitment cliff. The earlier one appears, the more it costs you, because everything below it is read by fewer people. Diagnose in page order and fix the first one you find before touching anything else.
Before any of that, one question decides which article you should be reading.
First: no traffic, or traffic that doesn’t convert?
These feel identical from the inside and have completely different fixes.
Traffic but no conversions. Qualified people are arriving, moving through the page, and leaving. Your analytics show sessions with real dwell time and no signups. This is a friction problem on a live page, and you have behavioural evidence to work with: scroll depth, exit points, recordings. Start with Getting Traffic but No Conversions, which covers the measurement side properly.
No traffic. You’ve had a few dozen visitors, or none, because you haven’t launched or haven’t spent on acquisition yet. Nothing in your analytics means anything at this volume. You can’t diagnose from behaviour because there isn’t any, so you have to diagnose from the page itself. That’s covered in How to Test a Landing Page When You Have No Traffic Yet.
The rest of this article works for both, because the nine friction points are readable from the copy regardless of how many people have seen it. They’re the vocabulary we use in the Friction Index, which is the scoring system behind our conversion briefs.
The nine friction points, in the order readers hit them
1. Category ambiguity
The reader cannot tell what kind of thing this is. Not whether it’s good, but what it is: software, a service, a marketplace, an agency. This is the most expensive failure on the list because it happens in the first few seconds and everything downstream is wasted on someone who has already mentally filed you under “unclear”.
How to spot it: read your headline and subhead only. Could a competent stranger name the category out loud? Beware of headlines that describe an outcome without naming the thing (“Ship faster. Sleep better.”). Outcome-only headlines feel bold and leave the reader guessing.
The fix: name the category explicitly somewhere in the first two lines, then be as ambitious as you like. “Conversion testing for pre-launch pages” costs you six words and removes the ambiguity entirely.
2. Value opacity
The reader knows what category you’re in and still can’t tell what changes for them. Usually caused by copy written at the level of capabilities rather than consequences.
How to spot it: take each of your first three body sentences and ask “so what happens to me?”. If the answer requires you to add a sentence that isn’t on the page, the value is opaque.
The fix: write the consequence, then the mechanism. “Find out where buyers stall before you spend on ads, because a panel of persona agents reads the page first” beats leading with the architecture.
3. Unproven superlative
A strong claim arrives with nothing beside it. “The most powerful”, “the fastest way”, “the only platform that”. Every reader has seen a thousand of these, and the sceptic in your audience treats an unsupported superlative as evidence you have nothing better.
How to spot it: search your page for the words best, fastest, most, leading, only, revolutionary. For each hit, look at the screen space immediately to its right and below it. Proof within one screen, or it’s friction.
The fix: either attach evidence to the claim or downgrade the claim. A specific, modest, provable statement outperforms an unprovable grand one with most buyer types.
4. Proof mismatch
You have social proof and it belongs to someone else’s buyer. Enterprise logos on a page selling to solo founders. A testimonial from a marketing agency on a page aimed at engineers. The proof is real; it just doesn’t reassure the person reading.
How to spot it: list your testimonials and logos, then list your target buyer. If a reader would look at your proof and think “those aren’t companies like mine”, it’s doing nothing. Different buyer types need different proof, and a page with one flavour of evidence serves one of them.
The fix: segment the proof, or state the fit explicitly. A single line of context (“used by 40-person product teams and solo founders alike”) repairs a lot of mismatch cheaply.
5. Missing metric
The page makes quantitative promises in qualitative language. “Save time”, “boost conversions”, “significantly faster”. Readers who are evaluating seriously convert these into numbers in their heads, and if you don’t supply the number they supply a pessimistic one.
How to spot it: count the numerals on your page that aren’t prices or years. If there are fewer than three, you’re asking readers to take everything on faith.
The fix: put a real figure next to your central promise, even a conservative one. “Under 10 minutes” is a claim someone can check and hold you to, which is exactly why it’s persuasive.
6. Price opacity
There’s no pricing, or the pricing is a link, or every tier says “contact us”. A meaningful share of readers treat missing prices as a signal that the answer is “expensive” and disqualify themselves before you ever speak to them.
How to spot it: time how long it takes to find a number from the top of the page. More than one click and one scroll is opacity.
The fix: show the numbers, on the page, including the free or cheapest option. Ours: Free at $0, Starter $19/mo, Growth $49/mo, Pro $149/mo. That took one line and answered the question the reader was going to leave over.
7. Gated tier
Pricing is visible but the thing the reader actually wants sits behind “talk to sales”. This converts an interested self-serve buyer into a lead they didn’t agree to become, and many of them simply close the tab.
How to spot it: look at the feature the page spends most of its copy on, then check which tier it belongs to. If your hero sells something only available on the enterprise plan, you’re advertising a product most readers can’t buy.
The fix: make the headline promise available on a purchasable tier, or be explicit early about who each tier is for.
8. CTA hesitation
The reader reaches the button, understands the offer, and doesn’t press it. Usually because the button’s language implies more commitment than the page has earned, or because it’s vague enough to feel like a trap. “Get started” tells nobody what happens next.
How to spot it: read the button text on its own, out of context. Can you say precisely what occurs after the click? If not, the reader can’t either, and unresolved uncertainty at a decision point reads as risk.
The fix: describe the immediate next action rather than the eventual outcome. “See a sample report” is a smaller, clearer ask than “Get started free”, and it’s honest about what the click does.
9. Commitment cliff
They clicked, and the next screen asks for a credit card, a work email, a company size and a phone number. The page built enough trust for one small step and the form demanded a large one.
How to spot it: count the required fields between the click and the value. Every field is a place to stop.
The fix: shrink the first step until it costs almost nothing, and move the qualifying questions to after the user has seen something worth qualifying for.
Why the order matters more than the list
Here’s the arithmetic that makes the sequencing point, and it’s the reason to fix top-down rather than fixing whatever annoys you most.
Assume, purely as a modelling exercise rather than a measured finding, that a page has five of these nine problems and each one loses 20% of the readers who reach it. The share who make it through all five is:
0.8 × 0.8 × 0.8 × 0.8 × 0.8 = 0.8⁵ = 0.328
So about 33% of readers survive the page, and 67% are lost without a single catastrophic problem anywhere on it. No individual failure looks fatal. Compounding does the damage.
Now remove the two earliest problems and leave the other three:
0.8³ = 0.512
Throughput goes from 0.328 to 0.512, which is 0.512 ÷ 0.328 = 1.56, a 56% relative improvement from fixing two things. The two you removed were worth more than the three you left precisely because they sat earlier in the sequence and gated everything below them.
The numbers are illustrative. The structure isn’t. Friction multiplies rather than adds, and early friction is worth disproportionately more than late friction, which is why “fix the first thing a reader hits” beats “fix the thing that bothers you most”.
Diagnostic summary
| # | Friction | The tell in your copy | The fix |
|---|---|---|---|
| 1 | Category ambiguity | Headline names an outcome, never the thing | Name the category in the first two lines |
| 2 | Value opacity | Capabilities described, consequences implied | Lead with what changes for the reader |
| 3 | Unproven superlative | Best / fastest / only with no evidence nearby | Attach proof or downgrade the claim |
| 4 | Proof mismatch | Testimonials from a different buyer segment | Segment the proof or state the fit |
| 5 | Missing metric | Fewer than three numerals on the page | Put a real, checkable figure on the main promise |
| 6 | Price opacity | Prices need more than one click to reach | Show the numbers, including the cheapest tier |
| 7 | Gated tier | Hero feature only exists on “contact us” | Make the headline promise purchasable |
| 8 | CTA hesitation | Button text is vague out of context | Describe the immediate next action |
| 9 | Commitment cliff | Many required fields before any value | Shrink the first step, qualify later |
Work down that table in order. Stop at the first row that’s clearly true of your page and fix it before reading further, because fixing item 8 while item 1 is broken is rearranging furniture in a house nobody has found.
Getting an outside read
The hard part isn’t the list. It’s that you can’t apply it to your own page reliably, because you already know what your product does, so your headline reads as perfectly clear to the one person on earth who wrote it.
Getting round that means importing a perspective. Five strangers who match your buyer will do it if you can find them and get them to narrate rather than compliment. A paid research panel will do it well and slowly. Buyer Clone does it in under ten minutes: paste a URL and a panel of AI buyer-persona agents moves through the page as different customer types, reporting where each one stalls, doubts, bounces or converts, scored against the friction points above and returned as a ranked conversion brief. No snippet, no traffic, nobody to recruit. There’s a live sample report at /sample if you want to see the format before trying it on your own page.
What this diagnosis can’t tell you
Nine friction points cover the structural failures, which is most of what goes wrong on most pages. They don’t cover everything.
They can’t tell you the market doesn’t want your product. A perfectly clear page for something nobody needs will convert badly and every item above will read as fine. Nor do they cover traffic quality: if your ads are attracting the wrong audience, the page can be flawless and still fail, and no amount of copy work will repair a targeting problem.
Our own reads have specific limits. We’re strong on the structural side: unclear copy, missing proof, weak CTAs, pricing opacity, claims that arrive without evidence. We’re directional only on emotional nuance and on exact pricing sensitivity, so treat “this price feels high to the sceptic” as a flag to investigate rather than a number to act on. And once you have real traffic, this doesn’t replace user research or measurement. It replaces guessing while you wait for enough visitors to measure anything, and the maths on how many that takes is more sobering than most teams expect.
Frequently asked questions
Why is nobody signing up from my landing page?
The most common cause is that readers can’t quickly tell what your product is or what changes for them, which stops them before any of your persuasive copy is read. Work down the nine friction points in page order, starting with category ambiguity and value opacity, and fix the earliest one that applies. Early friction gates everything below it, so it’s worth disproportionately more than anything near the CTA.
How do I know if it’s my page or my traffic?
Check whether the visitors arriving match your target buyer by source, keyword and intent. If qualified people are arriving with real dwell time and leaving without acting, it’s a page problem. If your traffic is mismatched or largely bots, no copy change will save it and you should fix targeting first.
How many people do I need to test my landing page?
Far fewer than you think for finding problems, and far more than you think for measuring improvements. Comprehension failures are visible in a handful of readers, so five honest strangers will surface most structural friction. Detecting a small conversion difference between two versions is a separate job that requires thousands of visitors per variant.
Should I rewrite the whole page or change one thing?
Change the earliest broken thing first, then look again. Rewriting everything at once means you learn nothing about which change mattered, and it’s easy to fix a symptom while leaving the cause in place. One targeted pass, re-read, repeat is faster in practice than a full rebuild.
Does a low conversion rate always mean the page is broken?
No. Traffic quality, offer strength, price and market timing all move the number independently of the copy. The page is the cheapest thing to fix and usually the first place to look, but if the friction points above all come back clean, the problem is likely upstream in who you’re attracting or what you’re offering them.