A launch video racks up millions of views. The product is called revolutionary. The founder is charismatic and the mockups are gorgeous. Six months later there's still nothing you can actually use.
That gap between a stunning announcement and a shipping product is where vaporware lives. The frustrating part is that the loudest launches and the realest ones can look identical on day one. Both have a slick site, big claims, and a waitlist. The difference shows up in the evidence, and evidence is exactly what a good demo is engineered to hide.
This is a checklist of tells you can run in a few minutes before you sink time, money, or trust into a "revolutionary" tool. It's the same kind of pass we run when a product is submitted for the Directory: excitement only counts when something backs it up.
Vaporware vs. an early-but-real product
First, a distinction that matters, because skepticism aimed at the wrong target just makes you cynical.
Vaporware is a product that is announced and promoted but never meaningfully ships, or ships as a hollow version of what was promised. The announcement does the work the product can't. Sometimes that's deliberate deception. Often it's a team that oversold, hit reality, and kept the marketing running anyway.
An early-but-real product is different. It exists, it's rough, it does less than the pitch, and the team says so. It might be in private beta with a handful of real users. It might crash. That's not vaporware. That's a normal early stage, and some of the best tools in any category looked exactly like this once.
The line between them is honesty about what's shipped. An early product that says "here's what works today, here's what's coming, here's where we're stuck" is behaving well. A "revolutionary" product that hides which parts are real is the one to watch.
At Uptism we stage genuinely early tools as emerging rather than pretending they're proven. Being unfinished is fine. Being unfalsifiable is the problem.
Red flag: the demo you can never touch
The single strongest tell is a demo you can only watch, never run.
Warning signs:
- The product only appears in a produced video, a keynote, or a scripted walkthrough.
- Every path to trying it yourself ends at a waitlist, an "early access" form, or "coming soon."
- The demo cuts away at the exact moment something would need to actually compute, save, or load.
- Screenshots are clearly renders or mockups, not screen recordings of a working interface.
- Reaching out for a hands-on trial gets you a sales call instead of a login.
A staged demo isn't automatically dishonest. Some real products gate access because they're capacity-limited or genuinely not ready for the public. But when every route to touching the thing is blocked, and it stays blocked for months, you're evaluating a promise, not a product. Treat the video as marketing, not proof.
The honest version of "not ready yet" is specific: a date, a scope, a public roadmap, a way to join a real beta with real people in it. Vague perpetual "soon" is the tell.
Red flag: no changelog, no pricing, no docs
Shipping products leave a paper trail. Vaporware tends not to, because there's nothing to document.
Look for these three artifacts, and be suspicious when all three are missing:
- A changelog or release notes. Real software changes. A public history of "v1.2: fixed X, added Y" is hard to fake and boring to maintain, which is exactly why it signals a working product with real users to keep informed. No changelog months after a splashy launch is a quiet tell.
- Actual pricing. Not "contact us for early pricing" on a consumer tool that supposedly launched. A team that has shipped usually knows what it costs to run and what it charges. Permanent "pricing coming soon" often means there's nothing to charge for yet.
- Documentation. Setup guides, an API reference, help articles. Docs are written against a product that exists. Thin or entirely absent docs after launch suggest the product exists mostly in the announcement.
Any one of these can be legitimately missing early. A brand-new tool might not have pricing figured out. But the combination — no history, no price, no docs, plus big claims — is a strong pattern.
Red flag: claims you can't falsify
Grounded products make claims you could, in principle, check. Vaporware leans on claims you can't.
Watch for:
- Superlatives with no method. "10x faster," "the most accurate ever," "99% success" with no description of what was measured, against what, or how. A number without a method is a marketing asset, not a result. Ask: faster than what, measured how? If there's no answer, treat the number as decoration.
- Category-of-one language. "The world's first," "nothing else like it," "makes every other tool obsolete." Real teams tend to name their competitors and explain the tradeoff, because they've actually been compared. Pretending no alternatives exist usually means avoiding the comparison.
- Outcomes with no mechanism. Big promised results with no explanation of how the thing produces them. "AI-powered" is not a mechanism. If you can't get a plain answer for how it works, there may not be a how yet.
- Testimonials you can't trace. Glowing quotes from unnamed users, or logos of companies that turn out to be pilots, advisors, or aspirations rather than customers.
None of this means ambition is bad. Bold goals are how progress happens. The tell isn't the size of the claim, it's whether there's any way to verify it. A confident team backs a big claim with a way to check it. A vaporware pitch makes the claim un-checkable on purpose.
Green flags: what evidence of substance looks like
Flip the checklist over and you get a picture of a real product. These are the things we look for as the substance bar we apply to every listing:
- Shipping evidence you can touch. A free trial, a live demo you operate yourself, an open-source repo with recent commits, an app you can actually download and open.
- A visible track record. A changelog, dated release notes, a public roadmap, an issue tracker. Signs the thing moves.
- Real, nameable users. Independent reviews, forum threads from people solving actual problems with it, named case studies you can look up. Use outside the company's own channels.
- Honest limitations. The team saying what it doesn't do yet, where it breaks, who it's not for. Stated limits are one of the most reliable signals of a real product, because only a real product has real edges.
- Traceable pricing and terms. You can find what it costs and what you're agreeing to without booking a call.
You won't get every green flag from an early tool, and that's fine. What you want is honesty and at least some touchable proof. A rough product with real users and stated limits beats a flawless video with neither.
A note on how we present this: this is a checklist for your own judgment, not a scoring system, and nothing here is a fabricated test result. When we curate, we're telling you what we found and where, not handing you an objective grade. There's a difference between "we looked and it holds up" and "we ran a benchmark," and we try never to blur it.
A quick vetting pass
Put it together into a five-minute pass you can run on any "revolutionary" launch before you trust it:
- Can I touch it? Find a way to actually use the product right now. If every path is a waitlist, downgrade your confidence.
- Does it have a history? Look for a changelog, release notes, or a repo with recent activity. Motion over months is hard to fake.
- Can I find the price and the docs? Both traceable without a sales call is a good sign. Both permanently "coming soon" is not.
- Can any claim be checked? Pick the boldest claim and ask how it was measured. If there's no method, treat it as marketing.
- Who uses it besides the company? Search for independent reviews and real user threads. Silence outside the company's own channels is a flag.
If it clears most of these, you're probably looking at a real product, early or not. If it fails several, you're looking at a pitch, and you should wait before spending anything on it. Waiting costs you almost nothing. A shipping product will still be there, and a better one, next month.
For more context on why the loudest tools often disappoint, what "AI slop" is covers the same failure at scale. And if you'd rather skip the vetting, the Directory and the AI category hub are where we've already run this pass for you.
In Plain English
What this is: a short, reusable checklist for telling a real product from a hyped announcement that never ships.
Who it helps: anyone about to spend time, money, or trust on a tool that's being called revolutionary and hasn't proven it yet.
When to use it: the moment a launch impresses you and before you join a waitlist, enter a card, or build a workflow around a promise.
What to do next: run the five-minute pass above. Check that you can touch the product, that it has a history and traceable pricing, that at least one bold claim can be verified, and that real people use it outside the company's own marketing. If it clears the bar, great. If not, wait and let the product catch up to the pitch.
Skepticism here isn't pessimism. It's how you stay excited about the tools that are actually real. The goal is to spend your enthusiasm on things that ship, so being able to spot the ones that won't is what keeps optimism grounded.
Want the version we run on every submission? Grab the review checklist and see the full substance bar a tool clears before it earns a spot in the Directory.