Section 1
Key takeaways
• A case study is not a narrative about your competence; it is a tool the buyer uses to justify the purchase to themselves and their boss while you are absent from the decision. • Quantification is the moat, not the case study itself, 51% of B2B buyers say data-and-research-backed claims are what make content memorable enough to act on . • Case studies are the second-most effective B2B content format (53%, behind video at 58%) yet are used by 75% of marketers, meaning the format is crowded but the quantified version is not. • The SIR structure, Situation, Intervention, Result, works because it leads with the number, not the adjective, and lets the prospect reach their own conclusion. • If your case study still reads as persuasive when you delete every adjective, it's built right. If deleting the adjectives leaves nothing, you wrote a testimonial, not a proof.
Section 2
Why your best proof has to work while you sleep
Start with the timing problem, because it reframes everything else. The modern B2B purchase happens mostly in the dark: buyers self-educate, shortlist, and build internal consensus before a vendor knows they exist. The decisive moment, when a buyer decides whether you belong on the shortlist, happens during the anonymous, self-serve phase. No discovery call, no tailored pitch, no chance to "handle the objection." Your assets are unsupervised, and they either carry the argument or they don't. A practitioner framing of it is blunt but accurate: a good case study "does selling work while your reps sleep" . That is the job, persuade a stranger you'll never speak to, on a timeline you don't control. Here's the trap. Most founders respond to "we need proof that works without us" by producing more content, more stories, more logos, more testimonials. But the format is already saturated: three in four B2B marketers (75%) use case studies . The differentiator was never the existence of a case study; it's whether the case study contains evidence a skeptical buyer can act on. If your positioning narrative upstream is still fuzzy, fix that first, the case study amplifies a clear story and exposes a vague one (more on that in the positioning work that makes proof land).
Section 3
What metric makes a case study persuasive instead of pleasant?
There's a measurable answer. When buyers were asked what makes content memorable and shareable enough to act on, the single biggest driver was that "the content uses data and research to support claims", 51%, the top factor in the Demand Gen Report content preferences survey . Not the storytelling. Not the design. The data. Pair that with the effectiveness ranking. In CMI's 2025 B2B benchmark, a survey of 980 B2B respondents, case studies/customer stories came in as the second-most effective content type at 53%, behind only video at 58% . The format works, but so does everyone's. The 51% data point tells you where the leverage actually sits: inside the case study, in whether it's quantified. The implication for a service business is precise. A case study that says "the client was thrilled with our partnership" competes with every other case study that says the same thing, which is all of them. A case study that says "support tickets dropped from 1,400/month to 410/month in one quarter after we re-routed tier-1 triage" competes with almost nobody, because almost nobody published the number. This is what Flint McGlaughlin of MECLABS Institute means in one of the more useful lines on the subject: "The goal of marketing is not to make a claim; the goal of marketing is to foster a conclusion." A claim is "we improved their conversion rate." A fostered conclusion is showing the prospect "0.96% → 8.1% in 90 days, single change," and letting them think the word "improved." The second one survives a skeptic. The first one is just your opinion of your own work.
Section 4
The anatomy of an un-quantified case study
You've almost certainly written one. The typical un-quantified case study has four moves: introduce the impressive client, describe the challenge vaguely, narrate the collaborative process, end with a glowing quote. Every move is about texture; none of it is falsifiable. "Challenge: they needed to scale" tells a buyer nothing about whether it resembles their challenge. "We rolled out a comprehensive solution" tells them nothing about what you'd do for them. "They were delighted" is unmeasurable, so the skeptical brain discounts it to zero. What's missing is a number the prospect recognizes as their own, which is the framework.
Section 5
The BGA framework: The SIR Proof Block
SIR stands for Situation → Intervention → Result. The discipline is that each of the three is a single quantified anchor, not a paragraph of color. Build it in this order, every time. 1. Situation, one baseline metric the prospect recognizes as their own pain State the starting condition as a number, not an adjective. Not "they were struggling with lead conversion." Instead: "Inbound demo requests were converting to closed deals at 0.96%, about 1 in 100." Rules of thumb: • One primary baseline. Pick the single metric that maps to your buyer's felt pain. Resist listing five. • Make it recognizable. The prospect should read the baseline and think "that's roughly my number." If it's unrecognizable to your ideal buyer, you picked the wrong metric or the wrong client to feature. • Include the unit and the timeframe ("per month," "per quarter"). A naked number is unverifiable; a number with a denominator is evidence. A quick service-business example. An accounting firm doesn't write "the client had messy books." It writes: "Month-end close took 11 business days, and 3 of every 10 reconciliations required manual rework." Now a prospect with a slow close sees themselves. 2. Intervention, exactly one named, attributable lever This is where most case studies collapse. They list everything the team did, "we audited, restructured, retrained, automated, and supported", which makes the result un-attributable. If you did ten things, the prospect can't tell which one caused the outcome, and neither can you. Name one lever and make it the hero: • Single attributable change. "We replaced manual tier-1 triage with a routing rule that sent the top 3 ticket categories straight to specialists." One sentence. One mechanism. • Make it portable. The prospect should be able to imagine you doing this specific thing for them. Vague interventions ("strategic optimization") aren't portable; specific ones are. • Acknowledge the supporting work honestly, then subordinate it. "(We also cleaned the knowledge base, but the routing rule drove the bulk of the change.)" Honesty about what else happened raises your credibility; pretending one lever did everything lowers it. 3. Result, a measured delta with a timeframe, and ideally a dollar figure Close the loop on the exact metric you opened with. Same metric, measured after. • Same unit, before and after. If the Situation was "0.96% conversion," the Result is "8.1% conversion", not "much better engagement." Never switch metrics between Situation and Result; that's the oldest tell of a result that didn't actually move. • Bound it with time. "8.1% in 90 days." A delta with no timeframe is a wish. • Translate to money where you can. "8.1% conversion on the same 600 monthly demo requests, at a $4,000 average deal, is roughly $170K/month in newly closed pipeline." The buyer's boss thinks in dollars, hand them the sentence so they can paste it into their justification. • Note the failure mode honestly if there is one. "It took three iterations on the routing logic before it stabilized" makes the 8.1% more believable, not less. A result with zero friction reads as fiction. Put together, one SIR Proof Block reads in under sixty seconds: Situation: Inbound demos converted to closed deals at 0.96% (≈1 in 100), ~600 requests/month. Intervention: We rebuilt the first-touch follow-up into a 3-step sequence triggered within 5 minutes of request, replacing the next-day manual email. Result: Conversion rose to 8.1% in 90 days, roughly $170K/month in newly closed pipeline at a $4K average deal. It took two sequence rewrites to get there. That block does the work of three pages of narrative, and it does it while you're asleep. The follow-up mechanics aren't incidental either, the speed-to-first-touch lever is its own discipline, covered in the systems that make follow-up automatic. Assembling the full asset One SIR Proof Block is the atom. The full case study is a thin wrapper around it: 1. A metric-first headline. Lead with the delta: "0.96% → 8.1% demo-to-close in 90 days." Not "How We Helped Acme Win." The number is the hook; the name is a footnote. 2. A one-line context tag so the prospect can pattern-match: "B2B SaaS, 12-person sales team, inbound-led." If your buyer doesn't look like this client, you want them to know fast. 3. The SIR Proof Block as above. 4. One verbatim client quote that names the number, not the vibe. "Our close rate went from 1% to 8%, we stopped hiring around the problem." A quote that repeats the metric reinforces it; one about how "lovely" you were dilutes it. 5. A single, low-friction next step. Don't end on applause. End on "Here's the 3-step sequence we used, see if your follow-up has the same gap," and link to a self-check, so the proof's momentum has somewhere to go. For a fill-in-the-blank version plus headline formulas, the Template Pack has the SIR skeleton ready to populate. You cannot quantify what you never measured. If your past projects have no baselines, start measuring on the current ones now, capture the Situation metric at kickoff so the Result is provable in 90 days. Building proof and delivering well turn out to be the same discipline. For the full positioning-and-proof sequence this fits inside, the StoryOS playbook is the map.
Section 6
A worked example: the un-quantified version vs. the SIR version
Take a real-shaped scenario, a fractional CFO service. Watch the same engagement told two ways. The victory lap: "When a fast-growing e-commerce brand came to us overwhelmed by their finances, our experienced team dove in. We brought clarity, structure, and confidence to their numbers. The founder told us we were a game-changer and couldn't imagine operating without us." A prospect reads that and learns nothing transferable. Which finance problem? What did you actually do? The skeptic's verdict: "Everyone says this." The SIR version: "Situation: A $6M e-commerce brand was making inventory buys off a spreadsheet reconciled monthly, carrying 140 days of inventory and out of cash twice in a year. Intervention: We installed a 13-week rolling cash-flow forecast updated weekly, tied to a single reorder trigger. Result: Inventory days dropped from 140 to 95 in two quarters, freeing roughly $480K of cash, and the brand hasn't hit a cash crunch since. (The forecast was wrong for the first month while we calibrated demand seasonality.)" Same engagement. The second is portable, falsifiable, and dollar-denominated. A founder with a cash-flow scare reads it and concludes, on their own, with no call, that you might free their cash too. That's the fostered conclusion: you didn't claim to be good at cash flow; you let $480K do the arguing. Once a prospect reaches that conclusion, the job shifts to making the next conversation easy, the demo-and-objection work downstream.
Section 7
You're running The SIR Proof Block right when…
You're running it right when you can delete every adjective and it still persuades, because the persuasion lives in the metrics, not the modifiers. When a prospect could forward the asset to their CFO with zero edits and the CFO would understand the bet in under a minute. When each case study has exactly one baseline, one named lever, and one measured delta with a date attached, and you'd be comfortable if the featured client fact-checked every number. And when you've stopped writing "they were thrilled" entirely, because you noticed the buyer never once acted on someone else's thrill. If, instead, your case studies switch metrics between start and end, list five interventions, or lead with the client's logo rather than the result, you wrote a testimonial. Testimonials feel nice. Quantified proof gets you onto the shortlist while you sleep.