A gated asset is invisible to AI. Crawlers do not fill in forms, so everything behind your lead capture, the research, the benchmark report, the buyer's guide, contributes nothing to what engines say about you.
Worse than nothing, actually. When your original research is gated, the blog posts other people write about it are not, so the engine cites their summary. Your data, their citation.
That is the trade B2B SaaS made a decade ago, when the asset's job was capturing an email. The asset's job now includes teaching machines who you are, and the two jobs are in direct conflict.
What exactly can AI not see?
Anything that requires an action a crawler cannot take. The list is longer than most teams think.
- Form-gated PDFs. The classic. The landing page gets crawled; the asset does not.
- Login-walled help docs. Your documentation answers exactly the how-to questions engines get asked, from behind a door.
- Email-gated pricing. "Contact us for pricing" means engines answer the pricing question from stale third-party guesses instead.
- JavaScript-dependent content. Not gated on purpose, but if the body copy needs a script to render, retrieval sees an empty shell.
The pattern is one-way glass. You can see the leads; the engines cannot see the content.
How do you find your own blind spots?
Fetch your key pages with JavaScript off and no session, which is roughly what a crawler gets, then compare what survives against what a buyer sees. The gap is your invisible surface.
Do the same check on the competitor you lose to most. Teams are routinely surprised to find the rival's documentation wide open while theirs sits behind a login, which is a retrieval advantage handed over for nothing.
Why does this cost more than it used to?
Because the answer layer resolves questions without a click, and it resolves them from whatever it can read. Absence does not mean silence. It means someone else supplies the answer.
Who gets cited instead of you?
Whoever wrote the ungated version. Analyst recaps, competitor comparisons, community threads summarising your report from the press release.
We see this shape constantly in scan logs: the brand that produced the research is absent, and three pages describing that research take the slots. The free 20 minute check will show you whether it is happening to you.
What is the compounding effect?
Engines build an entity picture from what they can read, and a company whose best material is invisible reads as thinner than it is. That weakens retrieval on every prompt, not just the ones the gated asset would have answered.
Meanwhile the ungated summarisers accrue citation history on your topic. Persistence compounds for them instead of you.
Should you de-gate everything?
No. The honest position is a split, decided by what each asset does for retrieval versus what the form actually captures.
What should be ungated, no argument?
Documentation, pricing information, integration lists, and the findings of any research you want cited. These are the pages engines retrieve for exactly the questions your buyers ask.
Help docs are the sharpest case. They answer how-to queries in your category better than any blog post, and a login wall removes them from the game entirely.
Integration pages are the quiet second. "Does X work with Y" is a constant assistant query, and the answer should come from your integrations directory rather than a partner's changelog.
What still earns its gate?
Assets whose value is the artifact, not the information. Templates, calculators, tools, certification courses. The engine cannot cite a spreadsheet usefully, and the lead intent is real.
The test: would an engine quoting this content answer a buyer question? If yes, the gate is costing citations. If the value only exists in using the thing, gate away.
What about original research?
Ungate the findings, gate the extras. Publish every headline number, chart and methodology note on a crawlable page, and keep the formatted PDF, the raw data or the workbook behind the form.
You keep most of the leads, because people who want the deck still fill the form. And the citations for your own numbers come to you instead of the recap writers. Structure the findings page so every stat is liftable on its own.
How do you make the case internally?
The objection is always lead volume, so bring the two numbers that reframe it.
| Metric | Gated | Ungated findings page |
|---|---|---|
| Form fills | Higher | Somewhat lower |
| AI citations of your data | Zero | Available to win |
| Who gets cited for your research | Summarisers | You |
| Compounding entity value | None | Accrues to your domain |
Then check what the gate is protecting. Most gated assets are downloaded, skimmed and forgotten; the emails were the point.
If the same page captures slightly fewer emails while making your data the cited source in AI answers your buyers read, that is not lost demand generation. It is reallocated demand generation.
Run the before and after on your prompt set, the same way you would measure any visibility change. Movement takes weeks, not days, so set the expectation before someone judges it on day five.
What does this look like in practice?
A sequence, not a purge. Start where the citation upside is highest and the lead cost is lowest.
De-gate documentation first; the leads from doc gates were never good anyway. Then publish a crawlable pricing page, even if it is ranges and models rather than exact numbers. Then convert your flagship research into a findings page plus a gated extras pack.
Announce each change to your own team as it ships. Sales especially, because reps who have been emailing PDFs for years will keep gating by habit long after the wall comes down.
Leave the tools and templates gated. Re-run your prompt set after each step, and fold the whole thing into answer engine optimization planning rather than treating it as a one-off cleanup.
Where Unveilr fits
Unveilr's audits check crawler access across your whole surface, including what sits behind forms and logins, then track whether de-gated pages start earning citations on your prompt set. Scan, detect, update, re-scan; the loop shows whether each de-gating decision actually moved an answer.
The findings-page pattern is the one we push hardest, because it is the rare change that wins citations without a real lead cost.
In one D2C case study, the brand moved from the 9th most-cited domain in its category to number 1, with ChatGPT visibility rising from 3.3% to 44.7%.