Strac Holds Top 10, Then Falls to 309th

Measured against the Machine Relations Index, Strac.io ranks top 10 in one AI Security & Privacy buyer shape and rank 309 of 356 in another.

Para Labs Research··
ai-visibilitybrand-visibilitymeasurementcase-study
Para Labs research image for Strac Holds Top 10, Then Falls to 309th

Measured against the Machine Relations Index, one named AI-security vendor — Strac.io — holds a top-10 citation slot in AI Security & Privacy's how buyers choose and near-top-25 in best tools, and collapses to rank 309 of 356 in the same category's problem-first research shape, over the same 2026-05-10 to 2026-09-29 window.

The mechanic measured here is shape-specific citation: whether a brand's own domain is cited at the same rate across different buyer question shapes in the same category, or only in some of them. Measured against the Machine Relations Index (MRI) release mri_score_v2.0+2026-09-29+3d7cca280bd7, strac.io — a vendor-owned domain in the data-loss-prevention and AI data-security space — holds five published segments in AI Security & Privacy, and its standing on three of them tells three different stories about the same brand.

The measurement

Field Value
Window 2026-05-10 to 2026-09-29, 143 days observed
Category AI Security & Privacy
best_x denominator 165 observed runs, 8 dates, 322 cited domains
how_choose denominator 159 observed runs, 8 dates, 357 cited domains
problem_first denominator 135 observed runs, 8 dates, 356 cited domains
Engines ChatGPT, Claude, Gemini, Google AI Mode, Google AI Overviews, Perplexity
Evidence floor per segment 10 observed runs across 7 distinct run dates

The machine-readable release carries the numbers below; the same rows are readable at each category-and-shape segment leaderboard.

Five segments, one brand

Positions are within the published AI Security & Privacy segment for that shape; the "of N" figure is the count of domains cited in that segment during the window.

Question shape Rank Rate Cited / observed runs
How buyers choose #10 of 357 11.32% 18 of 159
Best tools #24 of 322 7.88% 13 of 165
Top lists #30 of 364 6.67% 9 of 135
Is it worth it #33 of 383 5.03% 8 of 159
Problem-first research #309 of 356 0.74% 1 of 135

Strac holds a top-10 slot when a buyer asks how to choose between AI-security tools, and a top-25 slot when the buyer asks for the best ones outright. On problem-first research — the shape where a buyer is still working out whether the problem is real before naming any tool — the same domain is cited in one of 135 observed runs and sits at rank 309 of 356, a difference of more than fifteenfold between its best shape and its worst.

Strac's how-buyers-choose slot sits behind Teramind (23.27%), Aona (18.87%), Arxiv (15.72%) and Philterd.ai (15.72%), and just ahead of Microsoft (12.58%) and Cyberhaven (11.95%) — a leaderboard split between research repositories, community platforms and named security vendors. Its problem-first rank of 309 sits deep in a leaderboard topped by Arxiv (17.04%), TechRadar (14.81%) and YouTube (14.81%) — general research and explainer sources, not vendor comparison pages.

Engine presence is not the variable

Strac.io is cited by all six monitored engines across the window: ChatGPT, Claude, Gemini, Google AI Mode, Google AI Overviews and Perplexity all cite the domain at least once. A missing citation in problem-first research is not explained by absence from an engine — the domain is reachable by the full engine set and still drops to a single cited run in that shape. That is consistent with how the engine operators describe retrieval: Google documents AI features as drawing on its ordinary crawl and index (Google Search Central), Perplexity documents query-time search over live sources (Perplexity docs), and Anthropic documents a web search tool that retrieves and cites per request (Anthropic docs). Retrieval is per query; the question text is part of the query.

Read the source-role label before benchmarking against it

MRI classifies strac.io as vendor_owned, a domain the vendor itself operates. That label does not change across shapes; only the citation rate does. The pattern here is not a classification artifact — it is the same domain, the same source-role class, cited at sharply different rates depending on which question a buyer asked.

What a brand-side operator can take from this

Three things follow from the cross-section, and none of them is a causal claim.

  1. A single visibility number hides which shape is winning. Strac's overall citation rate across the whole index sits at 0.30% (56 of 18,368 monitored answer runs, standing #217 of 25,105 observed domains) — a number that gives no signal that the same domain holds rank 10 of 357 in one specific shape.
  2. Measure per shape before changing anything. A brand that only tracks its aggregate citation rate would read Strac as a mid-tier, unremarkable AI-security domain, and miss that it already holds a top-10 slot in the shape a buyer uses to shortlist tools.
  3. Problem-first research is where the gap is worst, and it is also the earliest shape in a buyer's research. A buyer forming the problem statement before naming any vendor is the shape furthest from Strac's citation strength, which is concentrated in the shapes where tools are already being compared or chosen.

Limits

This is one release. Segments below the evidence floor publish no rate, so a shape not listed above means the domain had zero cited runs in that published segment during the window, not a claim about the brand's visibility outside the Index. The Index runs the market's buyer questions, not queries about any one brand, so it cannot report whether Strac's own pages are performing well or poorly — only which shapes cite the domain and at what rate. Citation rate is not traffic, ranking or revenue. Nothing here establishes that a source-side change moves a segment; that requires a paired design with a frozen baseline, and the release identifier above is what a later read must be compared against.

Method

Source: Machine Relations Index, public view v2.0, release mri_score_v2.0+2026-09-29+3d7cca280bd7, generated 2026-09-29, window 2026-05-10 to 2026-09-29, 143 days observed, six answer engines (ChatGPT, Claude, Gemini, Google AI Mode, Google AI Overviews, Perplexity). Category: AI Security & Privacy. Per-domain, per-segment ranks and rates read directly from strac.io's published MRI profile and each segment's own leaderboard, not re-derived by sorting. Overall citation rate (0.30%), cited-run count (56 of 18,368) and standing (#217 of 25,105) are read from the same domain profile. Machine Relations Index · AI Security & Privacy: How buyers choose · AI Security & Privacy: Problem-first research