"Honesty over hype" is the second value on our About page, right under "answer the obvious questions automatically." It reads simply on a slide. In practice, it's the value that costs us the most, on a weekly basis, in ways I can actually point to. This is an attempt to explain the mechanics of that cost — and why we keep paying it.
The market rewards confidence, not accuracy
AI software has an unusual property: most buyers can't verify a vendor's claims before they buy. You can't test-drive "understands your customers." You can't kick the tires on "99% accurate." So in the moment of decision, the claim that sounds most confident tends to win the click, independent of whether it's true. This isn't a flaw in any individual buyer's judgment — it's a structural feature of a category where verification happens after the purchase, not before.
Which means the marketing that wins short-term attention is, almost by construction, the marketing least constrained by what's actually true. "Replaces your entire front desk." "100% accuracy, guaranteed." "Understands your customers like a human would." We don't write lines like these. Competitors do, regularly, and some of them out-convert us on the exact same search terms.
Claims like these sit in an interesting spot: they're the kind of language India's Advertising Standards Council exists to scrutinize once it crosses from ordinary puffery into a specific, checkable factual claim — a guarantee, a percentage, a head-to-head comparison. Most AI marketing copy is written carefully enough to stay on the puffery side of that line, which is exactly what makes it hard to challenge and easy to keep writing.
What we actually say instead
We say AIVA resolves roughly 96% of voice calls without a human. Not "handles everything" — a specific number, with an implied 4% it doesn't get right or shouldn't try to. We say it answers questions and books appointments, not that it "replaces" your team. We say it needs two to four weeks of calibration before it performs well, not that it works perfectly on day one.
Put the two versions of each claim side by side and the pattern is obvious. "Replaces your front desk" versus "handles the repetitive share of your calls and hands off the rest." "100% accurate" versus "resolves about 96% without a human, and we'll show you exactly what falls in the other 4%." "Works out of the box" versus "plan on two to four weeks of watching it and adjusting." Every one of those is a weaker sentence than the hype version, in the narrow sense of "less exciting to read." I think they're stronger sentences in the sense that matters, which is that they're still true three months after a customer signs the contract.
The case for rounding up anyway
I want to steelman the other side here, because the argument for a little exaggeration isn't purely cynical. A reasonable theory of how buyers actually read marketing copy goes like this: everyone knows vendors round up, so buyers apply a mental discount to every claim before deciding — "100% accurate" gets silently downgraded to something more like 85% in the buyer's head, and a vendor who states 96% plainly just gets read as slightly worse than the vendor claiming perfection, discount and all. Under that theory, our honesty doesn't read as honesty. It reads as a weaker product.
We think that theory used to be closer to true than it is now, specifically in this category. A buyer who's been burned once by a vendor's "100% accurate" claim doesn't apply a mental discount the second time — they stop trusting the number at all, discount or no discount, and start looking for a vendor willing to name the number that isn't 100. We're betting our positioning on that shift being real and continuing, not on buyers never having learned to discount in the first place. It's a bet, not a certainty, and I'd rather say that plainly than pretend we've proven it.
The bill we've actually seen
I want to be specific about the cost rather than wave at it. We've lost deals — I don't have an exact count, but it's not zero — to competitors whose pitch decks promised things ours doesn't. We've had prospects tell us directly that a competitor's claim of "100% accuracy" was more compelling than our honest description of an escalation system built around the fact that AIVA sometimes needs to hand off to a person. Confident and wrong beat careful and right, in that room, on that day.
One pattern we've seen more than once: a prospect brings a competitor's comparison sheet into a sales call, sits it next to our numbers, and asks why ours look smaller. The honest answer — that theirs are aspirational and ours are measured — doesn't always land in a thirty-minute call. Sometimes the deal goes to the bigger-looking sheet. We've made peace with that particular kind of loss, because it's usually a preview of a worse conversation that customer would have had with us three months later, just moved earlier and onto someone else's desk instead of ours.
Hype is a loan against future trust. It always comes due — usually in week three, when the product meets the promise.
What we do instead of a bigger pitch deck
If we're not going to out-promise a competitor, the honest alternative is to make our claims checkable instead of just statable. We publish what our resolution-rate number actually measures — including where a single average can hide a badly performing category underneath it — rather than just repeating "96%" and hoping nobody asks what it means. We let business owners listen to real calls during onboarding instead of asking them to trust a dashboard summary. Neither of those is a growth hack. Both are slower than a confident claim. They're also the only things that hold up when a skeptical buyer — and by now, most of them are skeptical, for good reason, a fear we hear about directly and have written about honestly — decides to actually check.
What we've gotten in exchange: customers who arrive with accurate expectations churn less. Our earliest cohort of customers — the ones who signed up when we had the least polished pitch and the most honest one — are disproportionately still with us. That's not a controlled experiment, but it's a pattern I trust, because it shows up in every retrospective we've done on churn.
Why we can afford this, and why that matters
We've never raised outside capital, which means our growth has never been evaluated on a quarter-over-quarter curve by anyone who wasn't also a customer. That's not a moral advantage — plenty of honest companies are funded, and plenty of bootstrapped companies hype shamelessly. But it does remove one specific pressure: the temptation to inflate a claim to hit a growth number for someone else's slide. Our only growth number that matters is renewals, and renewals don't respond well to promises that don't hold up. I've written in more detail about what staying bootstrapped has actually cost us elsewhere — this is one of the quieter benefits on the other side of that ledger, one that doesn't show up until you watch what a funded competitor's marketing starts to sound like in month eighteen, when the growth curve needs defending.
Where this is going
My honest guess is that AI marketing hype has a shelf life, and it's shortening. Buyers who got burned by an overpromising chatbot in 2023 or an IVR vendor in 2019 are more skeptical of confident AI claims than they were two years ago — several of our customers told us as much, unprompted, in sales calls. If that trend continues, honesty stops being a cost and starts being a filter that works in our favor. I don't know how fast that shift happens, or whether it happens fast enough to matter for us specifically. We're betting on it anyway, mostly because the alternative — hyping now and hoping the bill never comes — isn't a bet we're willing to make with other people's trust. It's the same reasoning, at bottom, that keeps "AI-powered" off our own homepage: a category that's been oversold this often doesn't need one more confident voice added to it. It needs a few vendors willing to be checkable instead.