Six months ago, someone on the team — not naming names, but it was a genuinely good pitch — wanted "AI-Powered Business Assistant" as the homepage hero line. It tested well in the mockups. It's what half our competitors lead with. It's what people type into Google when they're looking for exactly what we sell. We didn't ship it. The homepage says AIVA answers calls, books appointments, and answers customer questions — in twelve languages, at any hour. No "AI-powered" anywhere above the fold. I want to explain why, because it wasn't an obvious call, and I don't think it's the choice every company in our position should make.
The phrase that stopped meaning anything
"AI-powered" is on toasters now. It's on to-do list apps, CRM plugins, spreadsheet add-ons, and at least one water bottle I saw at a trade show that claimed to use AI to remind you to drink water. When a phrase gets attached to that much unrelated inventory, it stops carrying information. It used to signal "this does something genuinely new." Now it signals "someone in marketing had a deadline." Regulators have started noticing the same overuse: the US FTC's own guidance on the subject, "Keep your AI claims in check", warns advertisers directly that "artificial intelligence" is, above all, a marketing term that gets abused — not a reason to distrust the underlying phrase ourselves, but a useful outside confirmation that we weren't the only ones who'd noticed it had stopped meaning much.
The specific problem for us: "AI-powered" describes the technology, not the outcome. It tells you what's under the hood. It doesn't tell you that when a customer calls your clinic at 9 PM on a Sunday asking about tomorrow's availability, someone — something — answers, checks your actual calendar, and books the slot. That's the thing worth knowing. The technology is how we did it, not what we did.
What "it tested well" actually meant
I should be more specific about that mockup test, because "it tested well" is exactly the kind of vague claim this whole essay is arguing against. We showed five people outside the company two versions of the homepage hero — one saying "AI-Powered Business Assistant," one describing the job — and asked what they thought the company did. Every one of them correctly identified the AI version as "some kind of AI tool for business," faster than the plain version. Three of the five, asked a follow-up — "okay, but what does it actually do when your phone rings?" — couldn't answer from the headline alone. They'd correctly identified the category and completely missed the product. That gap is the whole argument. A category label is fast to parse and slow to inform.
The tension we didn't pretend away
I'll be honest about the part that makes this a harder call than it sounds: people do search "AI-powered receptionist" and variations of it. That's a real, measurable slice of search intent, and a few of our narrower landing pages use the phrase because that's literally the term someone typed to find us. Refusing to ever say it anywhere would be its own kind of dishonesty — pretending we don't know how people search.
The distinction we landed on: we'll meet you where you search, but we won't lead with it where we're not responding to a specific query. On the page that's making the first impression, the homepage, we describe the job, not the category. That's the compromise. It's not purity. It's just where we decided the line was worth drawing.
"AI-powered" tells you what's inside the box. We'd rather tell you what happens when your phone rings.
Where we do say it, and why that's not a contradiction
To make the distinction concrete: our AI receptionist cost page and our best AI receptionist for small business roundup both use "AI receptionist" and "AI-powered" language throughout, plainly and without embarrassment. Someone landing there typed a query with those exact words in it, looking to compare options. Answering them in different language than the one they searched with would be its own kind of unhelpful — a semantic word game that makes them work harder to confirm we're relevant. The homepage doesn't have that excuse. Nobody arrives there mid-search for a category. They arrive curious about a company, and the first sentence is the one chance to tell them something true and specific instead of something true and generic.
The steelman for the buzzword
I don't want to make the other side sound dumber than it is, because there's a real argument for the categorical headline, and it's not just laziness. A meaningful share of our first-time visitors don't yet know that "AI receptionist" is a category that exists at all — they've never searched for it, never heard the term from a competitor, and are arriving via a referral or an ad with no context. For that visitor, a fast category signal does real work: it tells them, in half a second, roughly what kind of company this is, before they've committed any attention to reading further. A plain job description asks for slightly more of that attention up front. In a category still this unfamiliar to a lot of small business owners, that's not a nothing cost.
We decided it was a cost worth paying anyway, but I want to be honest that it is one. The visitor who bounces before reading the second sentence because the first one didn't immediately signal "this is an AI company" is a visitor we probably lose under our approach and might have kept under the categorical one. We don't have a clean number for how often that happens. We only have a theory about which visitors matter more to keep — the ones who read past the first sentence.
What it probably costs us
I'd guess — and it is a guess, we haven't run the controlled experiment — that a punchier, more categorical headline converts a slightly higher percentage of first-time visitors who are browsing rather than actively evaluating. "AI-Powered Business Assistant" sounds bigger. It sounds like a category leader talking. Ours sounds more like a job description.
We think that's fine, for a reason connected to how we operate everywhere else: we'd rather have a visitor arrive already understanding what happens on the other end of a phone call, and decide from there, than arrive excited about a buzzword and get disappointed by the specifics three weeks into onboarding. The businesses that stay with us are the ones whose expectations matched the reality on day one. A headline is the first place that expectation gets set or mismatched. It's also worth naming who actually reads a homepage hero line closely: not the casual browser, but the owner who's about to spend fifteen minutes deciding whether to hand us their phone line. That's the reader we're writing the sentence for.
We still use the word "AI"
To be clear, this isn't AI-word-avoidance as a brand bit. We call AIVA an AI agent plenty of places, including this sentence, and the distinction between "AI agent" and "AI-powered" is itself worth being precise about — we've written separately about why the word "agent" carries more information than "bot" or "powered". The distinction here is narrow and specific: we don't let "AI-powered" stand in as the entire pitch, on the one page where we get to decide, uncontested by search intent, exactly what we want to say first. Everywhere else, we say what it does, and let the technology be the explanation, not the headline.
The rule this generalizes into
Once we'd made the call on the homepage, the same logic ended up applying more broadly than I expected. Our pricing page doesn't lead with "AI-powered pricing" either — it leads with the actual numbers, because a visitor evaluating cost wants the cost, not a category reminder. Our signup flow doesn't ask you to buy into a category before you've seen the product work. The underlying rule turned out to be smaller than "avoid AI hype" and more like a general instruction we already apply to the rest of how we write: describe the outcome before the mechanism, everywhere we're free to choose our own words rather than answer someone else's search term. It's the same instinct behind why we'd rather quote a specific resolution number than promise something unmeasurable, and it's easier to hold to on a homepage nobody funded us to grow faster than we're comfortable with — a point I've written about separately, on why we never raised venture capital. Nobody on a cap table is asking us to hit a growth number this quarter by making the headline louder. That's not a moral high ground. It's just one fewer reason to reach for the buzzword when the plain sentence is sitting right there, true and available.