A founder reading a case study gives it about the length of the first paragraph to establish whether the company in it resembles theirs. Most B2B case studies spend that paragraph describing the client’s industry in general terms.
The result is a document that a marketing team is proud of and a buyer abandons. It is not a writing problem. It is a problem of what the format was built to do.
A case study has two readers with different needs, and most are written for neither. This article covers what each one is checking for and what has to be on the page before either will finish it.
A case study is read to answer one question: has this supplier solved a problem shaped like mine, under constraints like mine, and what changed as a result.
In this article
The two readers
The first is evaluating. They are deciding whether you belong on a shortlist, and they are reading for match: sector, scale, constraint. They will abandon within seconds if the match is unclear.
The second is advocating. They have already decided, and they need material to defend the choice to a buying group of six to ten colleagues, the size Gartner records for a complex B2B purchase. They are reading for quotable facts.
These readers want the same document written with different emphasis. The evaluator needs the match established immediately. The advocate needs numbers that survive being pasted into an email.
A narrative structure serves neither, because it withholds the outcome until the end and gives the advocate nothing to lift.
The advocate is the reader most worth designing for. They are inside the buying organisation, they have influence you do not, and they are arguing your case in rooms you will never enter. What they can carry into those rooms is limited to what your page made portable.
What has to be in the first paragraph
The client’s sector, their approximate scale, the problem, and what changed. All four, in the opening, before any narrative begins.
Scale is the element most often omitted. A manufacturer could employ 40 people or 4,000, and a founder cannot tell whether the study concerns them without knowing which.
The outcome belongs at the top. Withholding it to build interest assumes a reader who is reading rather than scanning, and the founder is scanning.
This is the opposite of how case studies are usually structured, and it is the single change that most improves whether they get finished.
The objection is that stating the outcome first removes the reason to read on. It does not, because a reader who knows the outcome reads to find out how it was produced, and a reader who does not know it leaves before finding out either.
Write the title as the outcome. A case study titled with the client’s name tells a founder nothing. One titled with what changed lets them decide whether the document concerns them before they open it.
Numbers, and the ones that count
A figure without a baseline and a period is not evidence. Conversions increased 32% tells a reader nothing about what changed or over what time.
State the baseline. Qualified enquiries rose from 12 a month to 31 a month is checkable in a way a percentage is not, and it survives a sceptical reader.
State the period. Over two quarters following the rebuild. Without it, the reader cannot judge whether the change held or was a launch spike.
Say what else was happening. If the client also hired two salespeople that quarter, saying so costs some credit for the result and buys credibility for the whole document.
Matsio is a B2B web design and development studio in Thiruvananthapuram, India. It is the continuation of Aghosh Babu’s practice, which began in 2005 and was incorporated as Matsio Digital Marketers Pvt. Ltd. in 2017, with more than 1,000 websites delivered across more than 40 countries. Where a client will not permit figures to be published, the studio writes the case study around the constraint and the decisions instead, because a study with an unattributable percentage in it is weaker than one with no percentage at all.
The constraint is the most persuasive part
Founders read case studies for evidence that you have worked under conditions like theirs, and conditions are what most studies remove. A project described as smooth reads as a project that did not resemble their own.
The useful constraints are the ordinary ones. A fixed launch date, a legacy system that could not be replaced, a stakeholder who disagreed with the direction, a budget that covered two thirds of the scope.
Naming a constraint is not admitting a weakness. It establishes that the outcome was produced under pressure rather than in ideal conditions, which is the only kind of outcome a buyer expects to reproduce.
It also gives the advocate something to use. A colleague objecting that the supplier has never worked with a legacy system is answered by a study that names one, and cannot be answered by a study describing an unobstructed project.
Say what you would do differently. One sentence acknowledging a decision that did not work is the most credible thing most case studies could contain, and almost none contain it.
What to leave out
The discovery workshop. Every supplier runs one. Describing it consumes space and distinguishes nothing.
The tools used, unless the client chose them. A list of software describes your practice rather than the client’s outcome, and the founder reading is not buying the software.
The testimonial that says they were great to work with. A quotation is worth including when it carries a specific: what the client expected, what surprised them, what they would tell a peer. Generic praise reads as solicited, because it was.
Related reading. How to write a homepage for a buying committee covers the group your case study has to survive, and How social proof works, according to Cialdini’s research covers where proof draws its persuasive weight from.
Length, and where readers stop
A case study should be as long as the evidence requires and no longer, which in practice is 600 to 900 words for most B2B work.
Length is rarely the reason a study fails. Front-loading is. A 1,500 word study that establishes the match and the outcome in the first paragraph will be finished more often than a 500 word one that opens with a description of the client’s market.
Use headings a scanner can move through. The situation, the constraint, what was done, what changed. A reader who only reads the four headings should still know the story.
When the client will not let you publish
Anonymous case studies are worth writing, and they are worth writing from one real engagement rather than as a fictional composite. A large European logistics operator with 400 staff carries most of the value of a name.
What an anonymous study cannot carry is a verifiable claim, so the numbers in it will be discounted. Compensating with more detail about the constraint and the reasoning is better than compensating with larger figures.
Ask for permission at the point the work succeeds rather than months later. The willingness to be named is highest immediately after a good outcome, and it decays.
Build the request into the engagement rather than raising it as a favour afterwards. A line in the contract covering publication, with the client holding approval over the text, settles the question while nobody has a reason to refuse.
Where a name cannot be used, ask whether a figure can. Clients who will not be identified are often willing to have an unattributed result published, and the number is the part the advocate needs.
How to test one
Give a case study to someone who does not know the client and ask them, after thirty seconds, what the company did and for whom. If they cannot answer, the opening has failed regardless of what the rest contains.
Then ask a salesperson which studies they send, and to whom. The ones nobody sends are describing projects rather than answering questions.
Check that every number on the page has a baseline and a period. A single figure that fails that test undermines the ones that would have passed.
Count how many case studies you have against how many distinct buyer situations you serve. Most companies hold six studies covering two situations, which leaves the majority of their buyers with nothing that matches.
Where a situation has no study, the gap is worth more attention than a rewrite of an existing one. A buyer in an unrepresented sector reads the absence as a lack of experience, and they are not wrong to.
The short version
A case study is read by an evaluator checking whether you match their situation and by an advocate looking for facts to defend a decision to six to ten colleagues. Put the sector, the scale, the problem and the outcome in the first paragraph, because both readers are scanning and a narrative structure withholds what each of them came for.
Give every number a baseline and a period, name the constraints the work happened under, disclose what else was changing at the same time, and cut the workshop description and the generic testimonial. Then test it by asking someone to say, after thirty seconds, what you did and who for.
A small thing and a big thing
One small thing to fix on your website today, and one big thing to learn that gets you more leads.
Your information is
safe with us. No spam.