"2,000 customers" or "1,847 customers": precise vs round numbers on a landing page
Published on 25 August 2026 · 8 min read
"Joined by 2,000 customers" or "joined by 1,847 customers": same reality, opposite perceptions. The first reads like an estimate, maybe like a target displayed as if it had already been hit. The second reads like a record, a database export run this morning. That difference in processing has a name in consumer psychology — the numerical precision effect — and it matters enormously on a landing page, where almost every number on screen is doing argumentative work. The right question is never "precise or round?" in the abstract, but "precise or round, in which spot, for which job?"
Why a precise number reads like a calculated number
A round number signals an intermediate step: someone took a real result and smoothed it. A precise number signals the absence of that smoothing — it still carries the trace of where it came from. That trace, not the value itself, is what produces credibility. In 2008, Chris Janiszewski and Dan Uy published "Precision of the Anchor Influences the Amount of Adjustment" in Psychological Science (study on Google Scholar): when the anchor they were given was precise rather than round but of comparable magnitude, participants adjusted away from it less in their own estimates. The authors explain it through the resolution of the mental scale involved — a precise number summons a fine-grained scale where you adjust in small steps, a round number a coarse scale where you adjust in leaps. On a landing page, that means a precise number holds better in the visitor's head.
Precision as a competence signal — but not for everyone
Guang-Xin Xie and Ann Kronrod tested the effect directly in advertising in the Journal of Advertising in 2012, with "Is the Devil in the Details? The Signaling Effect of Numerical Precision in Environmental Advertising Claims" (study on Google Scholar). Across four experiments, a precise numerical claim made the advertiser look more competent: the precise figure was judged more informative, more "scientific". But the effect is moderated by the reader's advertising scepticism — the most sceptical consumers are far less affected, except when they know little about the topic. That nuance is decisive on a sales page: precision mostly convinces a visitor who already extends you a minimum of credit. With a battle-hardened audience, it doesn't replace verifiable proof, it accompanies it.
The flip side: round numbers are easier to remember, and sometimes more honest
Precision has a cost: it's hard to remember and hard to repeat. "Over 2,000 customers" survives a conversation; "1,847 customers" doesn't survive the first word-of-mouth retelling. Round numbers are cognitively fluent, therefore memorable, therefore shareable. They also carry an underrated honesty: if you don't measure down to the individual customer, "over 2,000" is more accurate than a fake unit-level count. So the useful dividing line is functional: a claim number, there to give an order of magnitude, is better round; a proof number, there to kill a specific doubt, must be precise, dated and verifiable.
The "fake precise" trap
There's a third category, the most dangerous one: the precise number invented to look precise. "97% satisfaction" with no survey behind it, "12 hours of content" with no one having timed it. These hold up right until someone asks, and collapse at the first question from a serious prospect. An owned-up round number costs you a few points of credibility; a precise one caught out costs you all of it and takes the rest of the page with it — the same mechanism described in our article on social proof on a landing page, where shaky proof does more damage than no proof at all.
Where to be precise, where to be round: the decision table
| Page element | Recommended format | Example | Why |
|---|---|---|---|
| Customer or subscriber counter | Precise | 1,847 customers helped | Signals a figure that was recorded, not an estimate rounded upwards |
| Measurable outcome promise | Precise if provable | Average reply time: 3h40 | Precision commits you: it must be sourceable if challenged |
| Customer review score | Precise, with the volume | 4.8/5 from 210 reviews | A perfect score looks filtered; the decimal makes the sample credible |
| Displayed price | Different logic (see below) | $89 rather than $90 | Here precision acts on perceived magnitude, not on credibility |
| Guarantee and commitment period | Round | Refunded within 30 days | A commitment must be memorable and free of arithmetic ambiguity |
| Number in the main headline | Depends on its job | In 7 days / 1,847 customers | Round if it structures the promise, precise if it acts as proof |
| Order of magnitude in B2B | Round or low threshold | Over 200 firms equipped | Protects confidentiality and won't go stale next quarter |
Behind that table sits a single rule: precision is a commitment. Every decimal you display promises that somewhere there's a spreadsheet row justifying it. Where that row exists, go precise. Where it doesn't — data that shifts daily, a measurement never taken, a client asking for discretion — the honest rounding wins, phrased as a threshold ("over 200") rather than a value ("200").
Price: an entirely different logic
On a price, precision doesn't act on credibility but on perceived magnitude, which changes everything. Manoj Thomas, Daniel Simon and Vrinda Kadiyali showed this in Marketing Science in 2010 with "The Price Precision Effect: Evidence from Laboratory and Market Data" (study on Google Scholar): buyers underestimate the magnitude of precise prices. In their first experiment, participants judged $395,425 to be smaller than $395,000 — a pure judgement error, but a systematic one. Their analysis of real residential real-estate transactions in two separate markets points the same way: properties with more precise list prices sold higher. In other words, a precise price doesn't say "this price is justified", it says "this price is smaller than it is". A distinct mechanism from counters, and one that combines with the left-digit effect covered in our article on prices ending in 9 and with the reference price built in our guide to price anchoring. The counter-intuitive consequence: social proof down to the individual customer and a round premium price sit together perfectly well on the same page.
Ratings, deadlines, guarantees: three zones to handle separately
The review score: 4.8/5 beats 5/5
A perfect 5/5 triggers instant suspicion of filtering, whereas a 4.8/5 shown with the review count describes a real distribution, disappointed customers included. The decimal does all the work: it proves an average was computed over a sample, therefore that a sample exists — the direct application of the imperfection effect in customer reviews. The score alone isn't enough either: without volume, "4.8/5" could rest on four reviews, which is why it matters to calibrate how many testimonials to show.
Deadlines and guarantees: round almost always wins
"Reply within 48h", "refunded within 30 days": on a commitment, the visitor isn't looking for proof, they're looking for a limit they can hold you to. A precise deadline ("within 47h") weakens the promise, because it suggests a calculation where a commitment is expected. One exception: the observed delay presented as a statistic rather than a guarantee ("average reply time over the last 30 days: 3h40"). That's back in proof-number territory, and precision becomes an asset again — provided you name the measurement period.
The headline number: round if it structures, precise if it proves
A number in a headline has two possible jobs. If it structures the promise ("Your landing page live in 7 days"), round wins: it has to be absorbed in one read, as our guide to writing a landing page headline that grabs attention points out. If it acts as immediate proof ("Used by 1,847 freelancers"), precise takes the lead. What to avoid: a precise number in a headline that nothing further down explains or sources.
The unverifiable number: the legal risk, and how to defuse it
A number on a sales page isn't decoration: it's a commercial claim. Under consumer protection law, a false or misleading figure about a product's characteristics, its expected results or the seller's standing counts as a deceptive commercial practice — and "it was just an estimate" is no defence when the figure was given down to the unit. The paradox is sharp: the more precise a number, the more credible it is, and the more you can be held to it. Precision is paid for in traceability:
- A named source: third-party review platform, analytics tool, customer database, published study — say where the data comes from, even in one line under the block.
- A date or a period: "as of 1 August 2026", "over the last 12 months". A number with no date goes false on its own.
- An explicit scope: 1,847 customers of what? Accounts created, paid purchases, active accounts? Vague scope is the most common weak point.
- A clickable link: a score pulled from an external platform is worth ten times the same score retyped on your own page, precisely because it's outside your control.
- An update routine: who refreshes the counter, how often — or replace it with a rounded threshold that stays true for months.
- A deliberate retreat: if nothing is sourceable, drop the number. A sentence without a figure convinces less than proof, but far more than proof that gets dismantled.
The checklist before publishing your numbers
- List every number on the page, headline to footer, badges and testimonials included.
- For each one, decide: claim number (order of magnitude to remember) or proof number (doubt to remove)? The format follows from the answer.
- Make precise every proof number backed by a real source, with its date and its scope.
- Round into a threshold ("over…") any number you don't measure at the level of detail displayed.
- Treat prices separately: perceived magnitude, not credibility — precise proof and a round price coexist without dissonance.
- Keep deadlines and guarantees round, unless it's an observed statistic over a named period.
- Delete any number you couldn't justify in under a minute to a demanding prospect.
Across the 10 LanderKit templates, the blocks that host these numbers — social proof strip, review score, counter, guarantee box — are already in place and waiting for your real data. The coach & consultant template shows the trade-off in action: a precise review score with its volume under the headline, a round guarantee next to the button, and a price that follows its own logic. Three numbers, three formats, one page — which is exactly right.
FAQ
Frequently asked questions
Should you round a customer counter while it's still small?
Don't round it up, and don't show it raw if it works against you. "1,847 customers" works; "23 customers" draws attention to how new you are rather than to your legitimacy. In that case, rephrase without a figure ("join the first users") or shift the proof to a detailed testimonial, which doesn't depend on volume.
Does a precise number have to be updated constantly?
Yes — that's the price of precision: a counter frozen for eight months gets spotted and cancels out the credibility gain. If you can't automate the refresh, prefer a rounded threshold ("over 1,800 customers") that stays true for months and that you raise occasionally.
Why does price follow a different rule from the other numbers on the page?
Because the mental mechanism isn't the same. On proof, precision acts on the credibility of the source; on price, it acts on perceived magnitude — Thomas, Simon and Kadiyali show that buyers underestimate precise prices. That trade-off is covered in detail in our article on prices ending in 9.
Is a percentage better proof than an absolute number?
It depends on the denominator. "92% of our customers renew" is strong if you know how many customers that is; over twelve customers it's both true and misleading. The rule: only show a percentage alongside its base, or use the absolute number when the base is too small to display.
Can you A/B test precise against round on your own page?
Yes, provided you test one number at a time and have enough traffic. The precision effect is real but modest compared with the clarity of the offer: if your value proposition isn't sharp, changing a counter's format won't make up for it.
Read next
Related articles
- The false consensus effect: why your landing page looks crystal clear to you and to nobody elseYou've reread your page ten times: the headline looks obvious, the promise self-evident, the arguments unbeatable. The problem is that you are the least qualified person on earth to judge. The false consensus effect — documented since 1977 — explains why a page written on instinct speaks to the founder and to no one else, and how to protect yourself from it without a research budget.
- The imperfection effect: why a perfect 5/5 rating can hurt trust on a landing pageA too-perfect number raises suspicion more than it reassures. "5/5 from 340 reviews" almost always sounds less true than "4.8/5 from 340 reviews" — and that's not just a hunch: research on online reviews documents precisely why a slight flaw makes a rating more credible, not less.
- The availability heuristic: why one concrete example convinces more than a statistic on your landing pageTwo proof blocks, only one survives in the visitor's memory: "92% of customers satisfied" or "Claire doubled her sign-ups in three weeks with this page." The statistic is more accurate, but the story is what convinces — a cognitive bias documented since the 1970s explains why, and more importantly how to put it to honest use.