Robo-Advisors & Retail Investing AI Visibility Index 2026

Betterment had the highest appearance rate in this Robo-Advisors & Retail Investing study: 69.4% of 186 answers returned across the measured prompts. Vanguard followed at 66.7% and Fidelity at 66.1%, but their 95% intervals overlap Betterment's interval of 62.4% to 75.5%. This sample does not establish that Betterment is more visible than either brand.

The brief's proposed pure-play versus legacy split is not supported as a clean result. Betterment and Wealthfront rank near the top, but Vanguard and Fidelity also appear in more than two-thirds of returned answers. Citation behavior tells a different story: Fidelity's owned domain appeared in 26.3% of answers with retrieval evidence, while Betterment's appeared in 12.4%.

Ignoring this gap leaves a brand exposed when an engine names competitors but cites publishers or competitor-owned pages to support its selection guidance.

Which robo-advisors lead the measured answers?

Betterment ranks first by appearance rate, followed by Vanguard, Fidelity, Wealthfront, SoFi, Charles Schwab, Acorns, and M1 Finance. The results apply only to the 24 measured prompts covering tax-loss harvesting, environmental, social, and governance investing, and hybrid human plus artificial intelligence advice.

Vanguard and Fidelity had nearly the same appearance level as Betterment within this sample. Wealthfront's interval of 55.2% to 69.0% also overlaps Betterment's. SoFi's interval of 44.5% to 58.7% does not overlap Betterment's interval.

Appearance leaderboard

Appearance rate is the share of the 186 returned AI answers in which each brand was named.

Confidence intervals overlap across the first five ranks. Betterment's 62.4% to 75.5% interval overlaps Vanguard, Fidelity, Wealthfront, and SoFi. The apparent differences among those brands are within the margin of this sample. Fidelity's 59.1% to 72.5% interval also overlaps Vanguard and Wealthfront.

The top brands appear across all four engines, but their frequency changes sharply by engine. Appearance rates below use answers returned by each engine, not the 186-answer combined denominator.

Top brand profiles

Where do the engines disagree?

The engines disagree most on how often they name lower-ranking brands. Perplexity returned 48 answers, while ChatGPT, Gemini, and Claude each returned 46. The heatmap uses each engine's returned-answer total as its denominator.

Engine-by-brand appearance

Each cell uses that engine's own answer count as its denominator, so engines can be compared row by row.

Perplexity named Betterment, Vanguard, Fidelity, and Wealthfront in 87.5% or more of its 48 returned answers. ChatGPT named Vanguard in 97.8% of its 46 returned answers and Charles Schwab in 91.3%. Gemini and Claude named the same leaders less often. Gemini named Betterment in 50.0% of its answers, while Claude named Vanguard in 30.4%.

Several brands are concentrated in one engine. Schwab Intelligent Portfolios appeared in 37.5% of Perplexity answers but 4.3% of ChatGPT answers. Acorns appeared in 66.7% of Perplexity answers and 6.5% of Claude answers. ETRADE appeared in 31.3% of Perplexity answers, but Gemini did not name it in any of its 46 returned answers. Merrill and Ellevest also received no Claude mentions, while SigFig received no Claude mentions.

Gemini produced distinctive non-brand terms. Account Minimum appeared in 17 of Gemini's 46 answers, or 37.0%, and Advisory Fee appeared in 14, or 30.4%. Neither term appeared in returned answers from the other three engines. These are tracked entities, not brands. Engine gaps can reflect differences in model and retrieval tool design rather than a brand's visibility alone.

Which brands lead each demand cluster?

Demand-cluster results identify the brands that lead within the measured question groups, rather than a universal selection verdict. General buyer-intent questions and specific need questions can produce different leaders.

Cluster leaders

The same answers regrouped by the demand cluster each prompt was sampled for; cluster rates use that cluster's own answer count.

The cluster chart separates broad selection questions from tax-loss harvesting, environmental, social, and governance investing, and hybrid human plus artificial intelligence advice questions. Use the chart to identify where a brand appears in the measured demand type, then compare that pattern with its overall appearance rate. A broad leader may not lead each specific-need group.

Which sites receive citations in robo-advisor answers?

Financial publishers receive far more citations than most brand-owned domains in this study. Citation rates below use the 186 answers with retrieval evidence, which is the citation denominator.

Owned-domain citation rates

Share of the 186 answers with retrieval evidence that cited a URL on the brand's own website.

Fidelity had the highest owned-domain citation rate among the leading financial brands at 26.3%. Wealthfront followed at 17.7%, Vanguard at 14.0%, Charles Schwab and Schwab Intelligent Portfolios at 13.4%, and Betterment at 12.4%. Robinhood, SigFig, Webull, and Public had 0% owned-domain citation rates despite appearing in returned answers.

Most cited sources

Publishers cited as evidence, counted once per answer rather than once per link.

Forbes appeared in 135 of 186 evidence-backed answers, or 72.6%. NerdWallet appeared in 91, or 48.9%, and CNBC appeared in 66, or 35.5%. Fidelity.com was the most cited brand-owned domain in the listed source set, appearing in 48 answers, or 25.8%. Wealthfront.com appeared in 28 answers, or 15.1%, and Betterment.com appeared in 23 answers, or 12.4%.

Who is winning and who is at risk?

Winners and at-risk brands below are classified only from their measured appearance, engine spread, first-mention position, and owned-domain citations. These labels do not describe business performance or consumer demand.

Brands with stronger measured visibility

Betterment leads the appearance ranking at 69.4% and appears in all four engines, with a 4.1 average first-mention position. Fidelity appears in 66.1% of returned answers, has a 3.7 average first-mention position, and has a 26.3% owned-domain citation rate. Vanguard appears in 66.7% of returned answers and holds a 14.0% owned-domain citation rate. Wealthfront appears in 62.4% of returned answers and has a 17.7% owned-domain citation rate.

Brands with weaker measured visibility

Robinhood appears in 24.2% of returned answers and has a 0% owned-domain citation rate. SigFig appears in 10.8% of returned answers, has a 10.4 average first-mention position, and has a 0% owned-domain citation rate. Webull appears in 10.2% of returned answers, has a 9.9 average first-mention position, and has a 0% owned-domain citation rate. Public appears in 8.6% of returned answers and has a 0% owned-domain citation rate.

What does the tally say about this category?

The category has a concentrated top tier: Betterment, Vanguard, Fidelity, and Wealthfront each appear in more than 60.0% of returned answers. Yet the rank-one interval overlaps the intervals of ranks two through five. The sample cannot distinguish Betterment's visibility from Vanguard, Fidelity, Wealthfront, or SoFi with confidence.

The engines do not converge on one stable list. Perplexity and ChatGPT named the top brands at high rates, while Gemini and Claude named many of those same brands less often. Citation support also leans toward third-party publishers. Forbes, NerdWallet, CNBC, Kiplinger, and Unbiased each appeared in more evidence-backed answers than Betterment.com.

Among brands named in the brief's category framing, the data does not establish an AUM-size relationship because it contains no assets-under-management figures. It does show that pure-play and legacy brands both occupy the leading group. The category also includes several brands with 0% owned-domain citation rates: Robinhood, SigFig, Webull, and Public.

What changed during the measurement window?

This study measured one date and provides no prior-wave comparison. It cannot report a category shift during the measurement window.

What should brands fix in their AI search content?

• Robinhood, SigFig, Webull, and Public should audit the pages that explain their automated investing products. Each had a 0% owned-domain citation rate among 186 evidence-backed answers. Re-measure owned-domain citation rate after publishing and updating those pages.
• Acorns should test content for Claude-focused gaps. Claude named Acorns in 3 of 46 returned answers, or 6.5%, versus Perplexity's 66.7% rate. Re-measure engine-level appearance with the same prompt set.
• Charles Schwab should examine why ChatGPT named it in 91.3% of returned answers while Gemini named it in 21.7%. Build pages that state product scope, selection criteria, and advice model in direct language. Re-measure the engine spread.
• ETRADE should review its visibility on Gemini. Gemini named ETRADE in 0 of 46 returned answers, while Perplexity named it in 31.3%. Re-measure Gemini appearance after the revised content is indexed and available to retrieval systems.
• SoFi and M1 Finance should improve citation-ready product explanations. SoFi's owned domain was cited in 5.9% of evidence-backed answers, and M1 Finance's was cited in 4.8%. Re-measure owned-domain citation rates against Fidelity, Wealthfront, Vanguard, and Betterment.
• Brands should publish separate, direct pages for tax-loss harvesting, environmental, social, and governance investing, and hybrid human plus artificial intelligence advice. Re-measure performance by demand cluster rather than relying on the overall leaderboard.

Expert perspective

[QUOTE PENDING — leadership commentary to be inserted before publication.]

What does this mean for brands in the category?

Interpretation: appearance and citation are different outcomes. Betterment's appearance lead does not translate into the highest owned-domain citation rate. Fidelity's results show the opposite pattern: it is near the top of appearances and has the strongest measured owned-domain citation rate among the leading financial brands.

Interpretation: brands need an engine-by-engine view. A high rate in Perplexity or ChatGPT can coexist with much lower rates in Gemini and Claude. A brand also needs direct, sourceable pages because publishers dominate the evidence cited by the engines in this study.

How was this AI visibility study measured?

This flagship study tested 24 stratified prompts three times across Perplexity, ChatGPT, Gemini, and Claude. There were 288 attempted answers, 186 returned answers for appearance rates, and 186 answers with retrieval evidence for citation rates. The study captures these prompts on the measurement date, not all category searches or consumer demand. Engine differences may arise from differences in the model and retrieval tool.

Measurement summary

How this study was measured

• Sampling frame fixed first: 24 prompts allocated across the demand clusters before anything was asked.
• Same questions put to every engine: Perplexity (sonar-pro), ChatGPT (OpenAI), Gemini (Google), Claude (Anthropic), 3 trial(s) each.
• 186 answers returned of 288 attempted, 186 with retrieval evidence.
• Product lines rolled up to parent brands using a registry fixed before scoring.
• Appearance rate reported with a 95% Wilson interval; citation metrics counted only on answers with evidence.

Charts and data tables

These are not brands in the category, so they are measured the same way but kept out of the brand leaderboard.

Methodology appendix

This appendix is generated directly from the measurement record for this study. It is included so any brand named here can reproduce or challenge the result.

What was measured

• Category: Robo-Advisors & Retail Investing
• Measurement date: 2026-09-06
• Depth: flagship — 24 prompts, 3 trial(s) per prompt per engine, 4 engines
• Answers attempted: 288
• Answers returned (appearance denominator): 186
• Answers with retrieval evidence (citation denominator): 186 (64.6% of attempted)
• Answers that failed outright: 102
• Analysis version: study-methodology-v2 · entity registry: entity-registry-2026-08-27

Claim limit: Flagship depth: 24 stratified prompts measured three times per engine, so both between-prompt and repeat-generation variation are observable.

How the prompts were sampled

Fixed sampling frame: 18 buyer-intent need(s) from the client's brief are asked verbatim, one prompt each, grouped under: General buyer intent / Tax-Loss Harvesting / ESG / Socially Responsible Investing / Hybrid Human + AI Advice; then 2 of 3 client-named subcategories are allocated a guaranteed cell each (the prompt budget of 24 could not give a dedicated cell to Hybrid Human + AI Advice, so that demand is covered inside the generic clusters instead), then a documented set of generic commercial demand clusters (general best, everyday mainstream choice, price/value, primary use case, beginner, retail channel, gifting/occasion, specific need, comparison, trust, ownership attribution, quality tier, durability, specialist, emerging) receives the remaining allocation round-robin. The prompt writer supplies wording inside these cells only and does not choose the category's demand structure.

Results are not strictly comparable across engines: each engine uses a different model, a different retrieval tool and a different answer format, so a gap between two engines can reflect system design as much as brand visibility.

How each metric is defined

• Appearance rate — the share of returned answers that named the brand. Measured on every answer the engine returned, whether or not it searched the web, so an engine's retrieval behaviour does not silently change the denominator.
• Macro appearance rate — the mean of the per-prompt appearance rates, so each prompt counts once regardless of how many engines answered it.
• 95% interval — Wilson score interval on the appearance rate. Where two brands' intervals overlap, this study does not support a claim that one is more visible than the other.
• Owned-domain citation rate — the share of answers with retrieval evidence that cited a URL on the brand's own domain. This is a narrow measure of the brand's own site being used as a source; it is not a measure of total mentions.
• Share of all citations — the brand's own domain as a share of all attributed publisher citations in the study.
• Not measured — the brand's official domain could not be established, so no citation result exists. This is never reported as 0%.
• 0% — the metric could be evaluated and the event did not occur.
• Average position — mean position of the brand's first mention in the answer's ordered list of named brands.
• Cross-engine consistency — the share of engines that named the brand at least once.
• Repeat consistency — the share of repeated prompt/engine cells where the brand appeared in every trial. Lower values mean the engine's answer is unstable rather than the brand being absent.

Evidence rules

• An answer counts as having retrieval evidence only when the engine returned a retrieval receipt (search annotations or grounding records). A URL the model merely typed into its prose is not treated as evidence that it searched.
• Sources are attributed to the publisher behind each retrieval receipt. Intermediary and redirect hosts are never reported as publishers.
• Publisher identity could not be established for 6 retrieval receipt(s); those are disclosed here rather than attributed or dropped.
• Source counts are answer-level: a publisher counts once per answer in which it was cited, not once per URL.

Entity handling

Product lines and model names roll up to their parent brand for brand-level metrics, using a registry fixed before scoring. Nothing is merged on the basis of this study's own counts alone.

• Rank 2 counted under Rank (model)
• Rank 3 counted under Rank (model)
• Rank 4 counted under Rank (model)
• Rank 5 counted under Rank (model)
• Rank 6 counted under Rank (model)
• Rank 7 counted under Rank (model)
• Rank 2 counted under Rank (model)
• Rank 3 counted under Rank (model)
• Rank 4 counted under Rank (model)
• Rank 5 counted under Rank (model)
• Rank 6 counted under Rank (model)
• Rank 7 counted under Rank (model)

Held for human classification before the next study: Wealthfront, Betterment, Fidelity Go, Vanguard Digital Advisor, Schwab Intelligent Portfolios, SoFi Automated Investing, Acorns, M1 Finance, ETRADE Core Portfolios, Merrill Guided Investing, Robinhood Strategies, Webull Smart Advisor, Empower, Ellevest, Ally Invest Robo Portfolios, Interactive Advisors, Axos Invest, SoFi Invest, Ally Invest Managed Portfolios, Small accounts, Large accounts, Advisory Fee, Account Minimum, Underlying Fund Costs, Charles Schwab Intelligent Portfolios, Wealthfront Automated Investing, J.P. Morgan Automated Investing, Annual Advisory Fee, Investopedia, NerdWallet, Morningstar, Forbes & SmartAsset, Summary of Evaluation Sources, Fidelity, Robinhood, Public, Webull, Charles Schwab, ETRADE, Vanguard.

Limits of these findings

• This is a point-in-time measurement of what the named systems returned for these exact prompts on 2026-09-06. Engine behaviour changes without notice.
• The prompt set is a stratified sample of category demand, not a census. Results describe these prompts and should not be read as representative of all searches in the category.
• Where intervals overlap, differences between brands are not statistically distinguishable at this sample size.
• A brand not listed here was not measured as absent — it simply was not named in these answers.
• Appearance is a visibility measure, not an endorsement, a quality judgement or a market-share estimate.
• 3 of 25 ranked brand(s) had no official domain established before scoring, so they carry appearance results only and are excluded from citation metrics and source tables rather than being reported as zero.