The First HR Software Buyer Leaderboard Is Not a Trade-Press List
The first HR & Talent buyer-question segment to clear the Machine Relations Index evidence floor is how buyers choose HR software. The top ten is not a trade-press list: six of the ten domains are still unclassified, two are community platforms, and only two are vendor-owned.
The HR software shortlist layer just became measurable.
In the September 21, 2026 Machine Relations Index release (mri_score_v2.0+2026-09-21+d73e239e7c5b), the how buyers choose segment for HR & Talent cleared the evidence floor: 101 observed answer runs across seven run dates, with 235 cited source domains. The day before, the same segment was still collecting at 95 runs across six run dates. One more run date made the leaderboard publishable.
The first read is not what an HR software marketer would expect. The top ten domains are not the HR trade press. They are not analyst firms. They are not a clean vendor shortlist. They are a mixed machine-readable surface: one vendor at the top, LinkedIn and Reddit immediately behind it, six domains the Index still classifies only as other observed sources, and another vendor at ninth.
That is the useful finding. When answer engines explain how to choose HR software, they are not reciting a publication list. They are assembling a shortlist layer from vendor pages, community platforms and lightly classified operator surfaces.
The newly published HR software shortlist leaderboard
The published segment is HR & Talent paired with the how_choose question shape. The denominator is 101 observed answer runs. A domain's rate is the share of those runs in which at least one observed answer cited the domain.
| Rank | Domain | Source role | Runs cited | Segment citation rate |
|---|---|---|---|---|
| 1 | Pin | Vendor-owned source | 14 of 101 | 13.86% |
| 2 | Community and social platform | 13 of 101 | 12.87% | |
| 3 | Community and social platform | 9 of 101 | 8.91% | |
| 4 | Selenios | Other observed source | 9 of 101 | 8.91% |
| 5 | Umantis | Other observed source | 9 of 101 | 8.91% |
| 6 | US Tech Automations | Other observed source | 9 of 101 | 8.91% |
| 7 | ClearCompany | Other observed source | 8 of 101 | 7.92% |
| 8 | CVViZ | Other observed source | 8 of 101 | 7.92% |
| 9 | Gem | Vendor-owned source | 8 of 101 | 7.92% |
| 10 | Lever | Other observed source | 8 of 101 | 7.92% |
Read as source roles, the top ten break this way:
| Source role | Top-ten slots |
|---|---|
| Other observed source | 6 |
| Vendor-owned source | 2 |
| Community and social platform | 2 |
| Editorial publication | 0 |
| Analyst and consulting research | 0 |
| Academic and government source | 0 |
The absence of editorial publications is the point. The first editorial publication is People Managing People at rank 11, also cited in 8 of 101 runs. SelectSoftware Reviews appears at rank 19, cited in 7. Both matter. Neither owns the shortlist question the way HR marketers usually imagine trade publications own it.
A buyer asking how to choose HR software is not being routed first through a news desk. The engine is reaching for the pages that look like selection surfaces.
Why this differs from HR news answers
HR & Talent already had a published news-shaped segment. That one is older, thicker and completely different: 608 observed runs across 53 run dates, with 1,249 cited domains.
Its top ten are:
| News-topic rank | Domain | Source role | Runs cited | News-topic rate |
|---|---|---|---|---|
| 1 | Unleash | Editorial publication | 69 of 608 | 11.35% |
| 2 | Landbase | Vendor-owned source | 59 of 608 | 9.70% |
| 3 | Crunchbase | Market and company database | 54 of 608 | 8.88% |
| 4 | Rippling | Vendor-owned source | 51 of 608 | 8.39% |
| 5 | SHRM | Academic and government source | 50 of 608 | 8.22% |
| 6 | Outsail | Analyst and consulting research | 48 of 608 | 7.89% |
| 7 | HR Executive | Editorial publication | 47 of 608 | 7.73% |
| 8 | People Managing People | Editorial publication | 45 of 608 | 7.40% |
| 9 | HR Tech Feed | Editorial publication | 43 of 608 | 7.07% |
| 10 | Community and social platform | 43 of 608 | 7.07% |
News answers cite editorial properties. Shortlist answers do not, at least not in the first ten slots of this newly published segment.
That difference is the category lesson. A publication strategy built from the HR news leaderboard will put operators toward Unleash, HR Executive, People Managing People and HR Tech Feed. A visibility strategy for the buyer-choice question has to ask a different question: what makes a selection page, comparison page or community result useful enough to be cited when an engine is helping a buyer decide?
The Index now has enough evidence to separate those two layers. Averaging them would erase the only useful signal.
What is still collecting
This is not an HR & Talent category verdict. It is one buyer-question shape.
The other HR & Talent buyer shapes are still below the evidence floor:
| HR & Talent segment | Status | Observed runs | Run dates |
|---|---|---|---|
| Best tools | Collecting | 102 | 6 |
| How buyers choose | Published | 101 | 7 |
| Is it worth it | Collecting | 96 | 6 |
| Problem-first | Collecting | 95 | 6 |
| Top lists | Collecting | 64 | 4 |
| Comparisons | Collecting | 65 | 3 |
| News topic | Published | 608 | 53 |
That table matters because the wrong move is available. A vendor could take the how-choose leaderboard and claim to understand HR software visibility. It does not. It understands one question shape at the first moment that shape cleared the publication floor.
Best-tools, is-it-worth-it and problem-first are all one run date short. They may publish on the next release. If they do, the source list may change again.
The published result should therefore be used as a first evidence point, not a finished category map.
What the top-ten shape says
Three things are visible already.
First, community survives the buyer layer. LinkedIn ranks second and Reddit third. That does not mean HR buyers are making vendor decisions from social threads. It means answer engines are using those surfaces as evidence when they explain selection logic. A community platform does not need to sell the tool to help the model describe how buyers talk about choosing one.
Second, vendor-owned pages can win the shape. Pin is first at 14 of 101 runs. Gem is ninth. ClearCompany, CVViZ and Lever sit near them but are still classified as other observed source in this release, which is itself a boundary: the public source-role field is not yet precise enough to make a clean vendor-versus-publisher claim from the top ten. The source-role corrections published this week already showed that source-role coverage is a live instrument limit; this segment demonstrates the practical consequence in a new category.
Third, editorial appears just below the line. People Managing People at rank 11 and SelectSoftware Reviews at rank 19 are not absent. They are outside the ten slots the answer interface is most likely to compress into a narrative. That is a different diagnosis than "trade press does not matter." It matters as an adjacent source layer, not as the first answer.
For an HR software company, that changes the work. A press win can help with news visibility. For buyer-choice visibility, the page that explains fit, tradeoffs, selection criteria and category vocabulary is not optional. The engine is already citing that kind of surface.
The first judgment to run next
The next release should be read before anyone turns this into a playbook. The test is simple:
- Does HR & Talent best-tools clear the floor on its next run date?
- If it does, do the first ten slots look like the how-choose list or like the news list?
- Do the same unclassified domains keep appearing, or does a different evidence layer surface?
- Does People Managing People move into the top ten when the question shape changes?
- Does any vendor-owned domain carry across two buyer shapes?
Those are the questions worth asking because the 2026-09-21 release creates a baseline, not a conclusion.
The clean statement today is narrower: the first HR & Talent buyer-choice leaderboard is published, and it does not look like an HR trade-media leaderboard.
Sources and method
All Index figures in this piece come from the public Machine Relations Index release generated 2026-09-21, retrieved live at machinerelations.ai/data/machine-relations-index.json and identified in this wake by SHA-256 prefix d73e239e7c5b. The observation window is 2026-05-10 through 2026-09-21. The release records 16,146 observed answer runs, 126,924 citation events and 22,377 cited source domains across six engines: ChatGPT, Claude, Gemini, Google AI Mode, Google AI Overviews and Perplexity.
The HR & Talent / how-buyers-choose segment is published at machinerelations.ai/index/categories/hr-talent/how_choose. The HR & Talent news-topic segment is published at machinerelations.ai/index/categories/hr-talent/news_topic. Domain identities and whole-index confidence grades are read from the corresponding public domain profiles under machinerelations.ai/index/domains. The September 20 release used for the collecting-to-published comparison is the local prior release mri_score_v2.0+2026-09-20+47973f373a20; it showed HR & Talent / how-buyers-choose collecting at 95 observed runs across six run dates. I did not infer per-engine rates, raw cited URLs or prompt text, because those are outside the public boundary of the release.