Last updated:
📊 The OSINT Demand Index 2026
What do 1.6 million OSINT practitioners actually search for? We harvested autocomplete from six engines across 12,786 queries — 26,139 cross-validated suggestions, each scored by engine consensus. The full dataset is open. This is the first public map of real OSINT search demand.
What this is: an open dataset of real OSINT search demand, built from six independent search engines (Google, Bing, DuckDuckGo, Brave, Yahoo, YouTube) across 12,786 seeded queries. Result: 26,139 cleaned unique suggestions, each scored 1–6 by how many engines independently returned it. Only 188 hit 6/6 — those are the hardest signals of genuine demand in the field. Download the full CSV →
Key findings
We harvested autocomplete from six independent search engines across 12,786 seeded queries, yielding 26,139 cleaned unique suggestions, each scored 1–6 by how many engines returned it independently. Only 188 hit 6/6 — full cross-engine consensus, the hardest signal of real demand in the dataset. The headline finding: cross-referenced against real Search Console data from an established OSINT site, 58 of those 188 top-tier terms had little or no ranking footprint at all — validated demand that nobody is serving. Four clusters dominate the gap: reverse image search, messaging-platform OSINT, data-broker removal, and the entire ALPR/Flock surveillance category, which barely existed as a search term before 2025. The full CSV is free.
How it was built
The headline: the strongest demand is unserved
Cross-referencing the 188 top-tier (6/6) terms against real Google Search Console data from an established OSINT tools site, 58 of them had little or no ranking footprint. The demand is validated across six engines — and nobody is serving it well. These are the clearest opportunities in the field right now:
Four clusters dominate the unserved list: reverse image and face search, messaging-platform OSINT (Telegram, Discord, TikTok), data-broker removal, and the entire ALPR / Flock Safety surveillance cluster — the last of which barely existed as a search category a year ago.
Demand by theme
The 5/6-and-above terms, grouped. Each is validated by at least five of six independent engines.
🤖 AI-era OSINT
| Query | Engines |
|---|---|
| osint ai | 6/6 |
| osint ai tools | 6/6 |
| reverse image search ai | 6/6 |
| osint gpt | 5/6 |
| people search ai | 5/6 |
| reverse image search ai free | 5/6 |
| flock safety ai | 5/6 |
| google dorks ai | 5/6 |
💬 Messaging-platform OSINT
| Query | Engines |
|---|---|
| osint discord | 6/6 |
| osint tools discord | 6/6 |
| telegram osint | 6/6 |
| telegram osint tools | 6/6 |
| discord osint bot | 6/6 |
| tiktok osint | 6/6 |
| tiktok osint tools | 6/6 |
| opsec discord | 6/6 |
🖼️ Reverse image & face
| Query | Engines |
|---|---|
| reverse image search | 6/6 |
| reverse image search google | 6/6 |
| reverse image search free | 6/6 |
| reverse image search people | 6/6 |
| reverse image search ai | 6/6 |
| reverse image search tools | 6/6 |
| reverse image search instagram | 6/6 |
| reverse image search app | 6/6 |
🧹 Privacy & removal
| Query | Engines |
|---|---|
| data broker removal | 6/6 |
| data broker sites | 6/6 |
| best data broker removal service | 6/6 |
| free data broker removal | 6/6 |
| best data broker removal | 6/6 |
| digital footprint | 6/6 |
| digital footprint meaning | 6/6 |
| digital footprint checker | 6/6 |
⚖️ Legality
| Query | Engines |
|---|---|
| is osint legal | 6/6 |
⚙️ Tooling & developer
| Query | Engines |
|---|---|
| osint github | 6/6 |
| osint tools github | 6/6 |
| osint kali linux | 6/6 |
| best osint tools github | 6/6 |
| osint tools termux | 6/6 |
| reverse image search api | 6/6 |
| maltego kali linux | 6/6 |
| osint api | 5/6 |
₿ Crypto
| Query | Engines |
|---|---|
| crypto osint | 5/6 |
The surveillance story
The fastest-emerging cluster in the whole dataset is automated license-plate recognition. flock safety, flock safety raven, flock safety lpr, what is flock safety, alpr police and alpr lite all hit 6/6 — a search category that was negligible before 2025. This maps precisely onto the wave of Flock contract terminations, officer-misuse cases and state legislation documented over the past year. When a surveillance vendor becomes a search term, the public has noticed.
Download the dataset
The complete 26,139-row CSV — every cleaned suggestion, its engine-consensus score, which engines returned it, and how many seeds surfaced it — is published on GitHub, free to download, analyse and cite.
osint-demand-index-2026.csv. Columns: query, engines (1–6 consensus), engine_list, seed_count. Licensed for free use — a link back is appreciated. We re-run the harvest annually.Methodology, seed list and harvester are reproducible. Harvest date: July 2026. Engines reflect live autocomplete at that time and will drift.
Frequently asked questions
What is the OSINT Demand Index?
It is an open dataset of what people actually search for in open-source intelligence, built by harvesting autocomplete suggestions from six independent search engines — Google, Bing, DuckDuckGo, Brave, Yahoo and YouTube — across 12,786 seeded queries. The result is 26,139 cleaned, unique suggestions, each scored by how many of the six engines independently returned it. A term returned by five or six engines is real, cross-validated demand rather than one engine's quirk. The full CSV is published on GitHub under a permissive licence.
Why autocomplete instead of a keyword tool?
Autocomplete reflects what people are typing right now, aggregated by the engines themselves, and it is free and reproducible. Paid keyword tools give volume estimates but are proprietary, US-skewed and expensive. Autocomplete gives no volume — its strength is breadth and recency, and cross-engine consensus is a strong signal that demand is genuine. We ran a gibberish control that returned nothing, confirming the engines are not simply echoing our seeds.
What does a 6/6 score mean?
That all six engines independently suggested the term from our seeds. Of 26,139 cleaned suggestions, only 188 hit 6/6 — these are the hardest, most cross-validated signals of live demand in the whole dataset. A further 411 scored 5/6. Everything at 5/6 or above is what a practitioner or a tool builder should treat as a confirmed, unmet-or-met need worth serving.
Is the data free to use?
Yes. The full 26,139-row CSV is on GitHub, free to download, analyse and cite. If you use it in research or writing, a link back to this page is appreciated but not required. We will re-run the harvest annually so the Index tracks how OSINT demand shifts year over year.
How was the noise filtered?
Suggestions shorter than 6 or longer than 60 characters were dropped, along with a blocklist of unrelated namesakes — the video game LSPDFR, the drug alprostadil, the @OSINTdefender account, and generic job/salary/Wikipedia noise. What remains is OSINT-topical. The raw unfiltered pull is available on request; the published set is the cleaned one.
What is the single biggest finding?
That a large share of the strongest demand is unserved. Cross-referencing the 6/6 terms against real Search Console data for an established OSINT tools site, 58 of the 188 top terms had little or no ranking footprint — meaning the demand exists and nobody is ranking well for it. Reverse image search, messaging-platform OSINT, data-broker removal and the whole ALPR/Flock surveillance cluster are the clearest gaps.