How to protect your market research
To protect your market research, treat what you sell and what you publish differently. Full reports and datasets are protected by copyright, a licence agreement that limits who may use them, and delivery through a secure portal or a DRM-protected PDF watermarked with each client's name, so a leaked copy can be traced. Headline findings are meant to travel, so the protection there is a citation policy that says how to quote you and links to the original. Credibility comes from disclosing your methods, through the AAPOR Transparency Initiative for polls, ISO 20252 certification for market and social research, and the ICC/ESOMAR Code.
Then sign each published report, chart, topline and press release under your firm's organisation TIP-ID on the Trust Identity Protocol (TIP) by The AI Lab, an open standard. Each one gets a permanent public record of who published it, when, whether it was made by people or with AI, and a fingerprint that shows whether a circulating copy has changed. When an AI answer, a vendor deck or a social post quotes one of your numbers, anyone can check it against the version you actually published, and a fake poll or report using your name has no record from you.
What needs protecting
What research organisations produce, and what goes wrong with each
| What you produce | What goes wrong |
|---|---|
| Headline statistics and press releases | The most quoted thing you make. AI answers and blog posts repeat a figure with no link, round it, drop the date or credit the wrong firm, and AI tools sometimes invent figures and attribute them to real research firms. |
| Syndicated reports and market size forecasts | Sold under licence and costly to produce. Copies are forwarded beyond the licensee or posted on document sharing sites, and forecast numbers are lifted into other firms' reports and pitch decks. |
| Public opinion polls, toplines and crosstabs | In 2026 several polls from firms that turned out not to exist were shared by campaigns and listed by poll aggregators before they were exposed. Real toplines get cropped or altered in screenshots. |
| Analyst evaluations and vendor rankings | Vendors quote placements and rankings in their marketing. Analyst firms run quote policies because excerpts get taken out of context or outlive the report they came from. |
| Survey datasets, questionnaires and methodology statements | The work behind the numbers. Without a dated original it is hard to show which version of a dataset or questionnaire produced a published finding. |
| Indices and trackers | Brand trackers and confidence indices are published on a schedule and cited for years. An old release resurfaces as if it were current, or a chart from one wave is passed off as another. |
| Charts and infographics | Shared as images, cut off from the report, relabelled or redrawn with different numbers while keeping your logo. |
| Client deliverables and qualitative recordings | Confidential by contract and often holding respondents' personal data. These belong in your secure portal under your data protection obligations. Sign only what you are ready to make public. |
Every option, honestly
What each way of protecting your market research does, and where it stops
These work together. Use more than one, and read the last column for each, including ours.
| Option | What it does | Where it stops |
|---|---|---|
| Copyright and licence agreement | Copyright covers your report text, charts and the way you select and arrange data. The licence sets who may read, share and quote the report. | Individual facts and figures are generally not protected by copyright, though the EU and UK database right can cover a substantial dataset. A licence binds only the people who signed it. |
| Secure portal, DRM and per-client watermarks | Limits printing, copying and forwarding, and stamps each copy with the client's name so a leak can be traced. | Does nothing once someone photographs a screen or retypes the numbers, and does not cover the findings you publish on purpose. |
| Citation and quote policy | Tells journalists, vendors and clients how to quote your research and when they need your approval. | Works on people who read it. AI answers and scraped copies do not ask. |
| Methods disclosure (AAPOR Transparency Initiative, ISO 20252, ICC/ESOMAR Code) | Shows how the research was done, which is what separates real research from a fabricated poll. | Proves your methods. It does not let a reader check that a number in circulation came from you. |
| AI training opt-outs (robots.txt, ai.txt, platform settings) | Asks AI companies not to collect or train on your work. | A request, not enforcement. Companies that ignore it are not stopped. |
| Trust Identity Protocol (TIP) by The AI Lab | Signs each report, chart, topline or release to your firm's verified TIP-ID and records when it was published, how it was made (Original Human, AI-Assisted, AI-Generated or Mixed) and a fingerprint, in a public record anyone can check for free. | Does not stop leaks, copying or AI training, does not check your sample or your data, and is not copyright registration. It proves what you published and when. |
Step by step
How to protect your market research, in order
- Decide what is public and what is licensed. Headline findings, press releases, toplines and methodology statements are public. Full reports, datasets and client deliverables stay behind your licence and portal.
- Deliver licensed reports through a secure portal or a DRM-protected PDF watermarked with each client's name.
- Publish a citation policy that says how to quote your research and links to the original.
- Disclose your methods, through the AAPOR Transparency Initiative for polls or ISO 20252 certification, and follow the ICC/ESOMAR Code.
- Get an organisation TIP-ID at vp.theailab.org and sign every public release, report page, chart and topline on TIP as you publish it.
- Link the TIP record, or show the Protected by proof on TIP badge, next to each published finding, so a journalist, a client or an AI crawler can check a number against your original.
- When a fake or altered version circulates, point to your signed record and to the fact that the fake has no record from your firm.
About TIP
What the Trust Identity Protocol is
The Trust Identity Protocol (TIP) is an open standard from The AI Lab for proving who made a piece of content and whether it has changed. A person verifies once and gets a TIP-ID. Each time they publish, their own device signs the work with a post-quantum signature (ML-DSA-65, NIST FIPS 204), and the TIP Network records the signature, the time, an origin label and a fingerprint of the content. Anyone can check a record for free, with no account.
TIP does not stop copying, does not stop AI training and does not decide whether content is true. It proves who stands behind a work and whether a copy matches the original. It is governed in the open by the AI Trust Council, and the specification is public on GitHub. Signed work is called a Verifiable Asset.
Questions
Questions people ask
How do market research companies protect their reports?
With copyright, a licence agreement, a secure portal or DRM-protected PDFs watermarked for each client, and a citation policy for the findings they publish. To prove which version they actually published, they can sign reports, charts and releases on the Trust Identity Protocol (TIP) by The AI Lab, which gives each one a permanent, signed public record anyone can check for free.
How do we stop AI from misquoting our research?
You cannot fully stop it. AI answers repeat published figures, often without a link, and sometimes invent figures and credit them to real firms. Robots.txt and ai.txt ask AI crawlers not to collect your pages, and some honour them. What you can do is make the original easy to check. A finding signed on TIP has a public record with your firm's name, the date and a fingerprint, so a misquote or an invented figure can be shown not to match anything you published.
How can people tell a fake poll from a real one?
Real pollsters disclose who they surveyed, when, how and with what margin of error, and many belong to the AAPOR Transparency Initiative. A pollster that signs each release on TIP adds a second check, a public record under the firm's verified identity. A poll that claims to come from that firm but has no record from it is not theirs. TIP does not judge whether a poll is well designed.
Are our statistics protected by copyright?
Your report, its wording, its charts and the way you select and arrange data are. Individual facts and numbers generally are not, which is why figures travel so freely once published. In the EU and the UK a database right can protect a substantial dataset. A TIP record does not change what the law protects. It proves that you published a figure, when and in what form.
Should we sign confidential client reports on TIP?
Sign what you publish. A TIP record is public by design, so confidential client deliverables and anything holding respondents' personal data stay in your secure portal under your contracts and data protection obligations. Public report pages, press releases, toplines, charts and methodology statements are what TIP is for.
Does TIP check the quality of our data?
No. TIP proves who published a report and whether a copy has changed. It does not audit your sample, detect fraudulent survey respondents or confirm that a finding is true. Methods disclosure, ISO 20252 and your own quality controls do that.
Is TIP free for research organisations?
Nonprofit research centres, universities and government statistics offices use TIP free at any size, as do businesses under USD 100,000 in annual revenue. Larger commercial firms use a published commercial licence, with the tiers at theailab.org/tip-license. Checking any record is free for everyone, with no account.
Other guides