ReplayAutonomous app testing for teams who ship at AI-speed
Promoted
StoryVoice
Customer case studies from 5-minute voice interviews
27 followers
Customer case studies from 5-minute voice interviews
27 followers
Send a link, your customer talks for 5 minutes, and AI generates a publish-ready case study with real quotes and metrics. No writer. No follow-ups.
Hey hunters, maker here.
I kept seeing the same gap at company after company. Hundreds of happy customers, but only a handful of case studies. Not because the stories aren't there but producing them is brutal. Chase the customer, schedule a 45-minute call, transcribe it, stare down a blank doc, write, edit, chase approvals. Three to six weeks per story.
The thing that unlocked it for me was that almost nobody fills in a written testimonial form or answer questions for 45 mins, but almost everybody will happily talk for 5 minutes async with an AI.
So StoryVoice works like this: you send your customer a link. They talk to our voice AI for about 5 minutes in an engaging conversation, whenever suits them. StoryVoice turns that conversation into a high-quality publish-ready case study in their own voice, plus a LinkedIn post, an X post, a pull quote, and a sales email.
Two things I care about that I'd love you to poke at. It computes headline metrics only from numbers your customer actually said, and if an interview doesn't contain a real story, it declines to write one instead of faking it.
The goal is simple: a case study should cost 5 minutes of your customer's time, not weeks of yours.
You can try the AI interview yourself at storyvoice.io/demo (no signup), and there's 20% off any plan for launch week. Would love your feedback. Happy to answer anything.
— Venkat
Said I'd come read this properly, so here it is. The five minutes isn't where our case studies died, approvals were. Every one I've tried to ship got stuck at the customer's manager or their legal person, so a fast draft turns a six week problem into a five week one. The thing I'd pay for is the approval step: send them the piece with their own quotes highlighted and a single approve link, so nobody has to read a whole doc to say yes. Does the customer see the draft before the maker does?
@asadmalik901 Straight answer to your question: the maker sees it first. Customer talks, the draft lands in the maker's dashboard to edit and publish, and the customer never sees it unless the maker sends it over themselves. So you've found a real gap, not a nuance.
You're right about where the time goes, too. We compress the writing, and the writing was never your six weeks.
One thing that dents it slightly, though less than an approve link would: every quote is verbatim from what they actually said and nothing gets invented, so there's less for a legal reviewer to argue with than a ghostwritten piece where they're approving words they never spoke. Smaller effect than what you're describing, and I won't pretend otherwise.
Quotes highlighted, one approve link, nobody reads the whole doc. That's a good spec. Noting it properly rather than politely.
@asadmalik901 To answer yours first: right now the maker sees it first and the customer never sees it at all. Reading your comment I think that's backwards.
Two things I genuinely can't decide, if you'll indulge me. First, approve is the easy half. What happens when they want a change? A free-text "here's my edit" box just turns it back into email. Something narrower, flag this line / soften this number, is faster but I don't know if it covers what legal actually asks for. What did your blockers actually push back on, wording, numbers, or the fact that it existed?
Second, who clicks approve at your accounts? The person who was interviewed, or their manager who never spoke to us? Changes everything about whether the link goes direct or gets forwarded with context attached.
@venkat_a2 Numbers, every time. Nobody ever argued with a quote about how the product felt, they argued about whether we could publish a churn figure, and twice the fix was just dropping the number. So a soften this number control would have unblocked more of mine than a free text box ever would. And it was never the person I interviewed who approved it, it went to their manager and then legal, so the link gets forwarded and lands with someone who has no idea what they're looking at.
Dial
@venkat_a2 @asadmalik901 the forwarded-with-no-context part is probably doing more damage than the number itself. legal isn't equipped to fact check a churn figure, so the reflex when something unfamiliar lands in their inbox is to reject it, not evaluate it. one line of context riding along with the approve link - what the number is and where it came from - might unblock more of those than a soften-this-number control, since some of that pushback is unfamiliarity, not an actual factual objection.
Dial
the "decline to write one instead of faking it" bit is the detail that'd actually make me trust the output. what happens on the customer's end when that happens though - do they know their interview didn't produce anything, or does it just quietly vanish from the maker's dashboard? if a customer took the 5 minutes and nothing came of it with no explanation, that seems like the one way this could burn the exact goodwill it's built on.
@galdayan Straight answer: they don't know. The interview ends the same warm way either way, they get a proper sign-off, and then nothing further ever reaches them. The maker sees a panel saying there's no publishable story, plus the full transcript so they can see why. The customer never finds out.
So you've found the uncomfortable version of it. Declining to fake it is the honest half. The silence afterwards is the dishonest half, and right now we ship both.
The blocker is dull: the customer is the one person we currently can't reach. We don't capture their email anywhere in the flow, so there's nothing to send the polite version down.
Since you've clearly thought about it, what would you actually want that message to say? "Thanks, nothing publishable this time" reads honest to me but slightly insulting to them, and I can't tell if that's better or worse than silence.
Dial
@venkat_a2 the insulting read only happens because it's silent otherwise. one line of why would flip it, something like "didn't land on anything we could stand behind publishing" reads as an editorial call, not a rejection. and honestly if you're going to capture their email for the polite version anyway, that's the same email you need for a real testimonial pipeline later, so it's not even a one-off cost.
@galdayan That line is better than anything I had, and it works because it puts the judgement on us rather than on them. Which is also just true, the gate is about what we'll stand behind publishing, not about them being uninteresting. Taking it.
And you're right that the email isn't a one-off cost. Three things I want are all blocked on the same missing field, so it's one bit of plumbing, not three features.
The bit I'm still turning over: do you warn them up front that this might happen? Saying it at the start is honest and takes the sting out later, but it also plants "this might be a waste of your time" right before you ask someone for five minutes.
Dial
@venkat_a2 don't warn, reframe. tell them upfront what the actual process is - quick interview, some of these become full case studies, some don't - as a plain description, not a caveat. said as a maybe it plants doubt, said as a step it's just how the thing works. same information, but one reads like a risk and the other reads like a pipeline they're stepping into.
@asadmalik901 @galdayan You two specced a feature in this thread and it went live yesterday. What it does now:
When the story is written the person who spoke gets a link to it, either the moment the maker presses send or automatically a day later if they never get round to it. Same page the maker sees, except their link unlocks editing, approving and refusing. Refuse gets the same billing as approve, because an approve button on its own is a leading question.
Asad, no soften-this-number control. They can just edit the number. What I did build for your forwarded-link problem is that nobody can touch it or sign it without entering their name and role first, so the maker learns it was legal who approved it and exactly what legal changed.
Gal, the context riding along with the link is the part I'm gladdest you pushed. Every number now carries the customer's own words it came from and links back to the question that produced it. And since a made-up source is worse than none, the quote gets located in the real transcript, and any number whose source isn't in there gets dropped rather than published with a citation nobody checked. The no-story note went out with your sentence in it, near enough word for word.
Two things I still can't decide. Asad, does an editable page raise the bar for legal instead of lowering it, since signing something you could have changed feels heavier than ticking yes on something you couldn't? Gal, approval warns the maker before publishing but doesn't block them. Deliberate, and possibly the wrong side of that line.
If either of you fancies trying to break it, say the word.
@galdayan @venkat_a2 Editable is heavier, but heavy in the right way. Legal doesn't stall because a doc is long, they stall because they can't change it. Give a lawyer redline rights and they'll spend twenty minutes and sign. Give them approve or reject and they forward it up a level and you lose a week. The name and role on the signature is the bit I didn't ask for and will probably use most.