Machine Relations Index

Machine Relations Index (MRI)

Name: Machine Relations Index
Published: 2026-05-10
License: https://creativecommons.org/licenses/by/4.0/

A neutral public benchmark for how AI answer engines cite sources, built from observed root-domain citations across six engines. MRI reports citation rates by category and question type once minimum sample thresholds are met.

31,497 domains198,008 source events26,794 answer runs119/151 measurable segments validated

This release

Published October 17, 2026

The Machine Relations Index release mri_score_v2.1+2026-10-17+a2034c7ccdc9, published October 17, 2026, counts 31,497 cited source domains across 26,794 observed answer runs from 6 answer engines (2026-05-10 to 2026-10-17). Methodology mri_score_v2.1; 119 of 151 measurable segments publish a rate.

From the weekly Index Report of October 12, 2026, which compares release mri_score_v2.1+2026-10-12+7925b6c15feb with release mri_score_v2.0+2026-10-05+0a0c7e02957b. A change is the difference between two published cumulative rates, in percentage points; it records that a rate moved, not why. The two releases use different methodology versions (mri_score_v2.0 and mri_score_v2.1), so part of each change comes from the method.

Largest rises among graded domains: rework.com 0.29% (+0.14 pts); nhimg.org 0.52% (+0.12 pts); guideflow.com 1.37% (+0.11 pts)
Largest falls among graded domains: reddit.com 10.02% (-0.75 pts); youtube.com 7.32% (-0.30 pts); medium.com 3.94% (-0.23 pts)
Segments that crossed the evidence floor: Industrial / Best tools, Industrial / How buyers choose, Industrial / Is it worth it, Industrial / Problem-first research, Industrial / Top lists, Industrial / Comparisons, Local Services / Best tools, Local Services / How buyers choose, Local Services / Is it worth it, Local Services / Problem-first research, Local Services / Comparisons, Logistics & Freight / Problem-first research

Weekly Index Report, October 12, 2026 All weekly Index Reports Categories and leaderboards

Measurement window2026-05-10 to 2026-10-17

Published 2026-10-17 from the current dataset release covering 31,497 cited domains and 198,008 source events.

Sample threshold10/7

Rates publish after enough answer runs across distinct observation dates.

Segment coverage78.8%

119 of 151 measurable segments validated, 32 still collecting observations. A further 6 of the 157 segments in the full universe cannot accumulate evidence and are excluded from every rate.

Citation leaders

Top cited domains

Reddit#1 overall · Community and social platform100th percentile overall · 2,450 cited runs · 4 engines YouTube#2 overall · Search or media platform100th percentile overall · 1,876 cited runs · 6 engines LinkedIn#3 overall · Community and social platform100th percentile overall · 1,486 cited runs · 6 engines Forbes#1 of 1027 classified editorial publications · Editorial publication100th percentile overall · 1,041 cited runs · 6 engines Medium#5 overall · Community and social platform100th percentile overall · 965 cited runs · 6 engines TechRadar#2 of 1027 classified editorial publications · Editorial publication100th percentile overall · 758 cited runs · 6 engines Arxiv.org#7 overall · Academic and government source100th percentile overall · 712 cited runs · 6 engines G2#8 overall · Market and company database100th percentile overall · 662 cited runs · 6 engines

Source universe

Cited source index

Initial view shows the top 50 domains from 31,497 total.

Data through 2026-10-17

Row	Source	Class			Confidence
1	Redditreddit.com · #1 of 31,497 observed domains · source site	Community and social platform	9.14%	2,450	Confidence A
2	YouTubeyoutube.com · #2 of 31,497 observed domains · source site	Search or media platform	7.00%	1,876	Confidence A
3	LinkedInlinkedin.com · #3 of 31,497 observed domains · source site	Community and social platform	5.55%	1,486	Confidence A
4	Forbesforbes.com · #1 of 1027 classified editorial publications · source site	Editorial publication	3.89%	1,041	Confidence A
5	Mediummedium.com · #5 of 31,497 observed domains · source site	Community and social platform	3.60%	965	Confidence A
6	TechRadartechradar.com · #2 of 1027 classified editorial publications · source site	Editorial publication	2.83%	758	Confidence A
7	Arxiv.orgarxiv.org · #7 of 31,497 observed domains · source site	Academic and government source	2.66%	712	Confidence A
8	G2g2.com · #8 of 31,497 observed domains · source site	Market and company database	2.47%	662	Confidence A
9	NIHnih.gov · #9 of 31,497 observed domains · source site	Academic and government source	2.35%	629	Confidence A
10	Gartner.comgartner.com · #10 of 31,497 observed domains · source site	Analyst and consulting research	2.00%	536	Confidence A
11	Yahoo Newsyahoo.com · #3 of 1027 classified editorial publications · source site	Editorial publication	2.00%	536	Confidence A
12	Healthlinehealthline.com · #4 of 1027 classified editorial publications · source site	Editorial publication	1.70%	455	Confidence B
13	Microsoft.commicrosoft.com · #13 of 31,497 observed domains · source site	Vendor-owned source	1.65%	443	Confidence A
14	Amazon.comamazon.com · #14 of 31,497 observed domains · source site	Other observed source	1.49%	399	Confidence A
15	NerdWalletnerdwallet.com · #5 of 1027 classified editorial publications · source site	Editorial publication	1.48%	397	Confidence A
16	Guideflow.comguideflow.com · #16 of 31,497 observed domains · source site	Vendor-owned source	1.37%	366	Confidence A
17	Ibm.comibm.com · #17 of 31,497 observed domains · source site	Vendor-owned source	1.35%	363	Confidence A
18	TechTargettechtarget.com · #6 of 1027 classified editorial publications · source site	Editorial publication	1.34%	358	Confidence A
19	Dev.todev.to · #19 of 31,497 observed domains · source site	Community and social platform	1.16%	310	Confidence A
20	Substack.comsubstack.com · #20 of 31,497 observed domains · source site	Community and social platform	1.15%	307	Confidence A
21	Paloaltonetworks.compaloaltonetworks.com · #21 of 31,497 observed domains · source site	Vendor-owned source	1.13%	304	Confidence A
22	Landbase.comlandbase.com · #22 of 31,497 observed domains · source site	Vendor-owned source	1.13%	303	Confidence A
23	Hubspot.comhubspot.com · #23 of 31,497 observed domains · source site	Vendor-owned source	1.07%	287	Confidence B
24	The New York Timesnytimes.com · #7 of 1027 classified editorial publications · source site	Editorial publication	1.07%	286	Confidence B
25	Crunchbase.comcrunchbase.com · #25 of 31,497 observed domains · source site	Market and company database	1.04%	279	Confidence B
26	Wikipediawikipedia.org · #26 of 31,497 observed domains · source site	Academic and government source	1.03%	277	Confidence B
27	Facebook.comfacebook.com · #27 of 31,497 observed domains · source site	Community and social platform	1.02%	272	Confidence B
28	Github.comgithub.com · #28 of 31,497 observed domains · source site	Community and social platform	0.98%	262	Confidence B
29	Sentinelone.comsentinelone.com · #29 of 31,497 observed domains · source site	Vendor-owned source	0.95%	254	Confidence B
30	CNBCcnbc.com · #8 of 1027 classified editorial publications · source site	Editorial publication	0.91%	243	Confidence B
31	Alibaba.comalibaba.com · #31 of 31,497 observed domains · source site	Other observed source	0.88%	236	Confidence B
32	Zapier.comzapier.com · #32 of 31,497 observed domains · source site	Vendor-owned source	0.86%	231	Confidence B
33	PR Newswireprnewswire.com · #33 of 31,497 observed domains · source site	Wire and press-release distribution	0.85%	228	Confidence B
34	People Managing Peoplepeoplemanagingpeople.com · #9 of 1027 classified editorial publications · source site	Editorial publication	0.85%	227	Confidence B
35	PCMagpcmag.com · #10 of 1027 classified editorial publications · source site	Editorial publication	0.84%	226	Confidence B
36	Erpresearch.comerpresearch.com · #36 of 31,497 observed domains · source site	Market and company database	0.82%	220	Confidence B
37	Experian.comexperian.com · #37 of 31,497 observed domains · source site	Vendor-owned source	0.81%	218	Confidence B
38	Rippling.comrippling.com · #38 of 31,497 observed domains · source site	Vendor-owned source	0.81%	217	Confidence B
39	Stripe.comstripe.com · #39 of 31,497 observed domains · source site	Vendor-owned source	0.80%	214	Confidence B
40	Tom's Guidetomsguide.com · #11 of 1027 classified editorial publications · source site	Editorial publication	0.79%	211	Confidence B
41	Deloitte.comdeloitte.com · #41 of 31,497 observed domains · source site	Analyst and consulting research	0.78%	210	Confidence B
42	Harvard.eduharvard.edu · #42 of 31,497 observed domains · source site	Academic and government source	0.75%	202	Confidence B
43	Worldmetrics.orgworldmetrics.org · #43 of 31,497 observed domains · source site	Analyst and consulting research	0.75%	201	Confidence B
44	Bankrate.combankrate.com · #44 of 31,497 observed domains · source site	Other observed source	0.74%	198	Confidence B
45	Nvidia.comnvidia.com · #45 of 31,497 observed domains · source site	Vendor-owned source	0.72%	192	Confidence B
46	Grandviewresearch.comgrandviewresearch.com · #46 of 31,497 observed domains · source site	Market and company database	0.71%	189	Confidence B
47	Monday.commonday.com · #47 of 31,497 observed domains · source site	Vendor-owned source	0.69%	185	Confidence B
48	Databricks.comdatabricks.com · #48 of 31,497 observed domains · source site	Vendor-owned source	0.68%	183	Confidence B
49	Apple Newsapple.com · #49 of 31,497 observed domains · source site	Other observed source	0.68%	182	Confidence B
50	WebMDwebmd.com · #12 of 1027 classified editorial publications · source site	Editorial publication	0.68%	181	Confidence B

Categories and question-shape leaderboards

25 categories and 151 question-shape leaderboards in release mri_score_v2.1+2026-10-17+a2034c7ccdc9. Each leaderboard ranks sources on its own observed runs; a collecting segment publishes no rate until it clears the evidence floor.

AI InfrastructureBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
AI Security & PrivacyBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
AI Visibility & GEOBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
Consumer FinanceBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
Consumer HealthBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
Consumer ProductsBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
CybersecurityBest tools · How buyers choose · Is it worth it · News-driven citations · Problem-first research · Top lists · Comparisons
Deep Tech & HardwareBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
Education & TrainingBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
Emergent ProsumerBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
Enterprise SoftwareBest tools · How buyers choose · Is it worth it · News-driven citations · Problem-first research · Top lists · Comparisons
Family SoftwareBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
FintechBest tools · How buyers choose · Is it worth it · News-driven citations · Problem-first research · Top lists · Comparisons
Healthcare ServicesBest tools · How buyers choose · Is it worth it · News-driven citations · Problem-first research · Top lists · Comparisons
HR & TalentBest tools · How buyers choose · Is it worth it · News-driven citations · Problem-first research · Top lists · Comparisons
iGaming & BettingBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
IndustrialBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
Legacy News TopicsNews-driven citations
Local ServicesBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists · Comparisons
Logistics & FreightBest tools · How buyers choose · Is it worth it · Problem-first research · Top lists (collecting) · Comparisons (collecting)
Martech & AdvertisingBest tools (collecting) · How buyers choose (collecting) · Is it worth it (collecting) · News-driven citations · Problem-first research (collecting) · Top lists (collecting) · Comparisons (collecting)
Physical Consumer ProductsBest tools (collecting) · How buyers choose (collecting) · Is it worth it (collecting) · Problem-first research (collecting) · Top lists (collecting) · Comparisons (collecting)
Professional ServicesBest tools (collecting) · How buyers choose (collecting) · Is it worth it (collecting) · Problem-first research (collecting) · Top lists (collecting) · Comparisons (collecting)
Sales, GTM & CRMBest tools (collecting) · How buyers choose (collecting) · Is it worth it (collecting) · Problem-first research (collecting) · Top lists (collecting) · Comparisons (collecting)
VC & Private EquityBest tools (collecting) · How buyers choose (collecting) · Is it worth it (collecting) · Problem-first research (collecting) · Top lists (collecting) · Comparisons (collecting)

Agent access

Use this data in your agent

Every figure on this index can be read by machines. The MCP endpoint and the REST API return the same domain, category and segment records you see here. Each record carries its release identifier, window and denominators, and segment records also carry the evidence floor, so an agent can cite the exact number and when it was measured. The data is published under CC BY 4.0.

Read a segment rank as a statement about the questions behind it, not about a whole market. Some segments rest on only a few questions, and published rates still include runs where the question names the domain's own brand. Pulling from the API reproduces the published numbers; it does not re-run the measurement.

Read the returned questionContext, questionSet and naming fields before interpreting a rank. Missing or null context means unavailable for that release, never zero. When present, read questionContext.readAs and limitations, including namesReviewedCoverage and sourceRoleBasis; a rank describes the sampled questions, not an entire market.

MCP endpoint: https://machinerelations.ai/mcp

The Machine Relations Index measures how often AI answer engines cite each source domain when people ask real buying and research questions. It observes six engines — Perplexity, ChatGPT, Gemini, Claude, Google AI Mode, and Google AI Overviews — across a fixed basket of 24 measured subject categories plus a legacy news-topic bucket (25 taxonomy nodes in all). Findings are grouped into strata, where a stratum is one subject category paired with one question type. A stratum publishes a citation rate only after it clears the evidence floor of at least 10 observed runs across at least 7 distinct dates; below that line it is marked collecting rather than scored, so early or thin signal is never shown as settled. An engine-day is left out of the denominator when its runs cannot reach the numerator. History collected before per-engine-day collection health was recorded is excluded when that engine was asked at least five times and returned no eligible citation; each such day is marked with that backfill rule. From the day collection health is stamped at collection, a day is excluded only when that stamp says the engine was silent or dark. Quiet days and days whose health was not measured stay in the denominator. Every excluded engine-day is listed, with its run count and the rule that excluded it, in the denominator_exclusion block of each release. Every domain an engine cites is measured on equal terms — membership in any external catalog is recorded only as a label and is never used to decide what gets scored. The public Index reports rates and rankings; it excludes raw cited URLs, internal query identifiers, and provider payloads. Validated source segments currently cover 78.8% of the measurable benchmark, which is the 151 segments that can accumulate evidence out of 157 in the full stratum universe.

Release mri_score_v2.1+2026-10-17+a2034c7ccdc9. Degraded coverage: 5 of 6 answer engines are currently returning answers. gemini quiet (22 of 121 collected calls returned an answer, 18%). Coverage across the window is uneven: gemini answered on 132 of 154 observed days (2,122 of 6,145 returned runs produced a citation); google_ai_mode answered on 112 of 154 observed days (1,489 of 1,910 returned runs produced a citation); google_ai_overviews answered on 95 of 154 observed days (883 of 1,343 returned runs produced a citation). Engines below 90% of the window are named here; per-engine day and run counts for all 6 are in engine_roster.