Writing proof a buyer can actually check

Every B2B site claims results. Almost none of them can be checked. That is why the proof section — the part built specifically to create trust — is the part buyers scroll past fastest.

The fix is not better claims. It is claims of a different kind: specific enough that someone could go and verify them, and unremarkable enough to be obviously true.

Why the usual proof stopped working

Four formats do most of the work on B2B sites, and all four have been devalued by ubiquity.

The logo wall. A row of client logos with no story attached. It tells a buyer you did something for those companies. It could have been a two-day project six years ago. Everyone has one, including firms with nothing behind it, so it carries almost no information.

The unattributed testimonial. "They transformed our business." — Director, Manufacturing Company. If it cannot be attributed, a reader cannot distinguish it from one you wrote yourself. Some of them are.

The percentage with no denominator. "Increased leads by 300%." From what to what? Three to twelve is 300%, and so is a hundred to four hundred. A number without its base is a shape, not a fact.

The adjective pile. Trusted, proven, results-driven, industry-leading. These are what a site says when it has nothing specific, and buyers have learned to read them that way.

What checkable proof looks like

The shift is from impressive to specific. These are not the same thing, and when they conflict, specific wins.

Name the situation precisely. Not "a manufacturing client". "A 14-person industrial valve distributor in the Mid-Atlantic whose entire pipeline came from one salesperson's contacts." A reader in that situation recognises themselves. A reader who is not can still tell it is real, because nobody invents that.

Say what you actually did. Not "developed a growth strategy". The mechanism: what was built, what changed, what it replaced. Specifics about method are hard to fake and easy to verify by asking one follow-up question.

Give numbers their denominator. "From four enquiries a month to eleven" beats "175% increase" — it is smaller-sounding and far more believable, and believable is the job.

Use the unremarkable numbers. This is the counterintuitive one. A modest, precise result reads as true. A spectacular one reads as marketing, even when it is true. If your best case is a good one and your typical case is a fair one, showing the typical one buys more trust.

Say what did not work. Almost nobody does this, and it is disproportionately effective. A case study that notes the first approach underperformed and what you changed is enormously more credible than one where everything went perfectly, because real projects do not.

When the client cannot be named

The common objection, and it is a real constraint rather than an excuse. Plenty of good work is under NDA or with clients who simply prefer not to appear.

You can still be specific. What you cannot do is retreat into adjectives.

• Describe the company by shape, not name: size, sector, region, structure.
• Describe the situation and the mechanism in full. Neither is usually confidential.
• Be specific about the numbers you are allowed to give, and say plainly that others are withheld.
• Say why they are not named. "This client does not do publicity" is a normal sentence that costs nothing.

Anonymous and specific works. Named and vague does not. If you have to choose, keep the specificity.

Proof that is not a case study

Case studies are one format and often not the strongest.

A working artefact. Something a prospect can use without talking to you demonstrates competence directly rather than describing it. Our two are the Lead Leak Audit and The Qualified 25. Neither says we are good at this; both let someone find out.

Written conditions. Any firm can promise results. Publishing the conditions under which the promise applies — including where it does not — is a claim with a cost attached, which is what makes it informative. Ours is on the homepage rather than buried in a contract nobody sees before signing, it says where it does not apply, and the specifics are set per client rather than asserted as a universal number. Applying this article to our own site is the least we can do with it.

Showing the method. Articles like this one are proof of a sort: it is hard to write specifically about how something fails without having watched it fail. That is the argument for a blog on a B2B site, and it is a better argument than keywords.

Third-party reviews. Not on your site and not under your control, which is exactly why they carry weight.

Where to put it

Two rules.

Next to the claim it supports, not in a proof section. A section called "Case Studies" is a place people do not go. The same content placed immediately after the relevant claim gets read, because it answers the doubt at the moment it forms.

Above the fold, at least once. Something specific and checkable should be visible before anyone scrolls. If the first screen is all assertion, the rest is read defensively.

The general failure this all serves: companies with genuinely good work describing it so cautiously that they become indistinguishable from companies with none. The cure is not louder claims. It is smaller, harder, checkable ones — and it usually means going back to a client and asking what actually changed, which is a conversation worth having anyway.

More in B2B web design.