The Best-Tools List and the Comparison Answer Are Different Contests: A Six-Broker Rank Gap
Across six named robo-advisor and brokerage brands, every one ranks far higher in AI comparison answers than in best-tools roundups for the same category — measured against the Machine Relations Index.
Which question shape a buyer asks decides who gets cited, not just what the brand is: measured against the Machine Relations Index (MRI) release mri_score_v2.0+2026-09-23+fdec6388001a, six named vendor-owned brokerage and robo-advisor domains — Acorns, Wealthfront, Betterment, Charles Schwab, Fidelity and Vanguard — rank dramatically higher when an AI engine answers a head-to-head "X vs Y" comparison than when it answers an open-ended "best tools" question, in the same Consumer Finance category, over the same 2026-05-10 to 2026-09-23 window.
The measured cross-section
MRI splits buyer questions into shapes. Two of them sit inside Consumer Finance: "Best tools" (an open-ended roundup, 132 observed runs across 7 dates) and "Comparisons" (a named head-to-head, 138 observed runs across 7 dates). Reading each of the six brands' own domain profile in the same release gives a clean side-by-side across the two shapes — a paired cross-section from the same window and the same 16,475-run population, not a before-and-after over time — no recomputation:
| Domain | Best tools rank | Best tools rate | Comparisons rank | Comparisons rate |
|---|---|---|---|---|
| Acorns.com | not ranked (0 of 132 cited runs) | — | #1 of 170 | 38.41% |
| Schwab.com | #211 of 236 | 0.76% | #6 of 170 | 26.09% |
| Wealthfront.com | #85 of 236 | 2.27% | #7 of 170 | 24.64% |
| Fidelity.com | #99 of 236 | 1.52% | #14 of 170 | 10.87% |
| Betterment.com | #136 of 236 | 0.76% | #15 of 170 | 10.14% |
| Vanguard.com | #230 of 236 | 0.76% | #22 of 170 | 6.52% |
Every one of the six clears the top 25 of 170 sources in Comparisons. In Best tools, the best-placed of them (Wealthfront) sits at #85 of 236, and Acorns has zero cited runs in that segment at all — it does not appear on the Best-tools leaderboard, while it is the single most-cited source in Comparisons. MRI publishes a separate confidence field for each segment row, distinct from a domain's overall confidence grade; for all twelve cells behind this table, across all six domains, that segment-level field reads Unavailable. The ranks above are the finding; the segment confidence field does not certify them, and neither does the overall grade discussed below.
The Best-tools leaderboard for the same category, same window, is led by Forbes (39.39%), NerdWallet (29.55%), CNBC (28.79%) and Bankrate (27.27%) — third-party editorial roundups, not the vendors themselves. Reddit and YouTube also outrank every one of the six brokers in Best tools.
The mechanic
A "best robo-advisor" prompt asks the engine to assemble an unbounded field on its own, and it reaches for the sources that already did that work: editorial roundups built for exactly that comparison. A "Wealthfront vs Betterment" prompt already names the field, and the engine's shortest path to an answer is each named brand's own comparison or product page. MRI measures citations at the root-domain level, not the specific URL, so this dataset cannot confirm which page on wealthfront.com an engine actually drew from in any given run. As an example of the kind of asset that shape of question rewards, Wealthfront publishes its own head-to-head against Betterment with a feature-by-feature table, on Wealthfront's own domain — read live on 2026-09-23, returning HTTP 200 — the kind of page a comparison prompt would plausibly reach for. That is a difference in what each question shape rewards, not a difference in how good any one broker is: correlation, not causation, and this design cannot separate "a vendor page like this one answers the named question well" from "the engine defaults to whichever page already contains both brand names."
What this means for a brand in this position
A brokerage or robo-advisor site optimized only for "best X" roundup placement is optimizing for the segment where vendor domains structurally rank worst in this measured set — sixth-worst here belongs to Vanguard at #230 of 236. The segment where these six brands already win, decisively, is the one triggered by a buyer who already knows the field and is choosing between two names. The visible gap is evidence for where a named comparison page (this vendor vs. that one, written on the vendor's own domain) is undersupplied relative to where the citations already are.
Method
Source: Machine Relations Index, public view v2.0, release mri_score_v2.0+2026-09-23+fdec6388001a, generated 2026-09-23, window 2026-05-10 to 2026-09-23, 16,475 monitored answer runs across six engines (ChatGPT, Claude, Gemini, Google AI Mode, Google AI Overviews, Perplexity). Category: Consumer Finance. Question shapes: Best tools (132 observed runs, 7 distinct run dates) and Comparisons (138 observed runs, 7 distinct run dates), both above the evidence floor of 10 runs across 7 dates. Per-domain ranks and rates read directly from each domain's published MRI profile (https://machinerelations.ai/index/domains/