mapou Visibility Index · Financial services → Banking · May 2026
Which banks does AI cite most?
When buyers ask AI assistants questions like “What are the best options for checking and savings accounts in 2026?” or “Best affordable checking and savings accounts for everyday banking under $50?”, a small set of bankingget cited every time. Most don't. This report measures which.
How we measured. MVI is a 0–100 score per brand: 0 means AI never cites you in banking, 100 means it cites you in every prompt. We tested 20 brands across 5 AI assistants (ChatGPT, Perplexity, Gemini, Claude, Grok) using 20 fixed prompts, reused every monthly run for replicability.
How we describe AI visibility
- Stage 1, First encounter. The brand is discovered and cited occasionally in AI answers for buyer-intent prompts.
- Stage 2, Repeat use. The brand is cited regularly enough that it feels familiar and reliably present across prompts and engines.
- Stage 3, Default choice. The brand is the go-to recommendation in AI answers within its segment, often appearing first or most consistently.
Bottom line
Ally Bank leads banking on AI search visibility with MVI 76, sitting firmly in the default-choice tier, by a wide margin ahead of Chase (MVI 45).
Who AI cites most
Ally Bank is cited in 72 of 100 prompt-engine pairs (72%). 95% confidence interval 66-82.
Concentration
The top 3 brands (Ally Bank, Chase, Capital One) capture 48% of all citations in this segment. 12 of 20 tracked brands are cited in fewer than 1 in 10 prompt-engine pairs.
Where the field sits
Of 20 brands tested: 1 in default-choice, 6 in first-encounter, 13 not yet cited. Overall, AI cites a brand from this segment in 15% of buyer-intent prompt-engine pairs.
Engine asymmetry
Marcus by Goldman Sachs is cited in 78% of Claude prompts but only 8% on Perplexity, visibility is engine-specific, not universal.
Notable absence
TD Bank is not yet cited, 0 prompts across all 100 prompt-engine pairs. A recognizable brand AI is not yet surfacing.
Analyst note
Ally Bank leads with a 76 MVI, far ahead of Chase's 45 MVI.
Ally Bank is the leader in this segment with an MVI of 76. It performs well across engines, especially on ChatGPT and Claude at 88 each. Chase follows at 45 MVI. Many brands are not making an impact, with 13 out of 20 falling into the invisible category. This creates a situation where Ally Bank is the primary choice for consumers.
Risk:Discover Bank has a 73 MVI on Claude but is nearly invisible on Grok at 8, showing engine concentration risk.
Headline finding
Ally Bank leads banking on AI search visibility with MVI 76, sitting firmly in the default-choice tier.
Average MVI
18
Default choice
1
Of 20 brands
Repeat use
0
First encounter
6
Not yet cited
13
Citation rate per engine
How often each engine cites a brand from this category as a recommendation, averaged across all 20 brands tested.
ChatGPT
19%
12 / 20 brands cited at least once
Perplexity
9%
12 / 20 brands cited at least once
Gemini
14%
10 / 20 brands cited at least once
Claude
21%
13 / 20 brands cited at least once
Grok
14%
12 / 20 brands cited at least once
Phase strength across the category
Which buyer-intent phases are easiest vs hardest to win in banking. Citation rate averaged across all brands tested. Phase weights are part of the MVI formula.
Discovery · 30%
17%
Top: Ally Bank
Filtered discovery · 25%
14%
Top: Ally Bank
Comparison · 25%
13%
Top: Ally Bank
Evaluation · 20%
17%
Top: Ally Bank
Who wins which buyer phase
Top 12brands by MVI mapped against the four buyer-intent phases. Each cell shows the brand's citation rate for that phase, color-coded so the visual pattern tells the story: a brand strong across all four phases reads as a horizontal orange band; a brand strong only at Discovery but weak at Evaluation reads as a left-heavy gradient. This is the segment's findings against the panel.
| Brand | MVI | Discovery | Filtered | Comparison | Evaluation |
|---|---|---|---|---|---|
| Ally Bank | 76 | 82% | 57% | 80% | 88% |
| Chase | 45 | 45% | 35% | 57% | 40% |
| Capital One | 41 | 33% | 40% | 45% | 48% |
| Discover Bank | 36 | 43% | 27% | 25% | 50% |
| Goldman Sachs | 34 | 50% | 27% | 10% | 48% |
| Marcus by Goldman Sachs | 34 | 47% | 27% | 20% | 43% |
| Bank of America | 32 | 28% | 30% | 40% | 30% |
| Wells Fargo | 13 | 17% | 3% | 18% | 13% |
| Axos Bank | 9 | 13% | 3% | 15% | 0% |
| Aspiration | 6 | 3% | 13% | 5% | 0% |
| Citi | 5 | 2% | 13% | 3% | 3% |
| Amalgamated Bank | 4 | 0% | 13% | 3% | 0% |
How to read it. Strong horizontal band = durable brand, AI cites it across the entire funnel. Left-heavy gradient = brand with awareness (Discovery) but weak recommendation (Evaluation), the demand-leak pattern from Finding 03. Right-heavy gradient = brand AI considers in Comparison and Evaluation but does not surface in initial Discovery, the anti-leak pattern. Color steps: dark orange ≥75%, orange ≥50%, peach ≥30%, light ≥10%, beige >0%, neutral 0%.
For CMOs in banking
What this report means for your banking portfolio
Each bullet is a category-specific decision derived from this month's data, with the mapou service that operationalizes it.
Concentration risk
In banking, AI effectively recommends 8.5 of 20 tracked brands. The top brand captures 22% of citations. The next 2 capture another 12%. Visibility is concentrated but not winner-takes-all.
How mapou helps: GEO & Citation Architecture restructures your entity data so you can break into the top set.
Engine convergence
Unusually for our panel, banking engines mostly agree. Cross-engine correlation is 0.75 (1.0 = perfect). Improvements on one engine are likely to lift others. Lower-effort optimization than fragmented categories.
How mapou helps: AI Visibility Audit confirms your position is consistent across engines before you commit budget.
Persona-volatile category
Banking rankings shift meaningfully by buyer persona. Ally Bank is the baseline #1 brand, but loses the top spot under at least one buyer signal (budget, premium, professional, first-time, values-driven). Top-3 overlap with baseline is only 60% across personas. Your baseline visibility number is incomplete.
How mapou helps: Persona-Tuned MVI computes the visibility number for your actual buyer mix.
Visibility tier landscape
In banking, 1 of 20 tracked brands clear MVI 75 (default-choice tier). 13 are below MVI 25 (not yet cited). The strategy differs at each tier. If you are below 25, you need foundational visibility infrastructure before tactical optimization.
How mapou helps: AI Visibility Audit identifies your tier; GEO & Citation Architecture moves you up.
The mapou Visibility Index
What is MVI?
The mapou Visibility Index (MVI) is a 0-100 proprietary score combining four weighted dimensions: Discovery (open recommendations, 30%), Filtered Discovery (budget, persona, use-case, 25%), Comparison (head-to-head authority, 25%), and Evaluation (decision-criteria authority, 20%).
Citations count fully; mentions count at half weight. Engines are equally weighted (no market-share gymnastics). Wilson 95% confidence intervals are shown alongside every score. The same 20 prompts run every month so MVI deltas are paired comparisons, not noise.
How to read this ranking
- Default choice (MVI 75+). AI's go-to recommendation in banking. The tier other brands are competing into.
- Repeat use (50–74). Cited often enough to feel reliably present across prompts and engines. One signal away from default.
- First encounter (25–49). Discovered and cited occasionally, but visibility is inconsistent. The brand is real to AI, not yet trusted.
- Not yet cited (0–24). AI does not surface this brand for buyer-intent prompts in banking. Effectively invisible in AI-driven discovery.
Ranked by MVI score (Wilson 95% CI shown). The Spread column shows the gap between each brand's best and worst engine, under 15pp is durable, 50pp+ is engine-dependent. Per-engine columns show the count of prompts where each engine cited the brand as a recommendation (out of 20). Read each column as a signal: when ChatGPT cites you but Gemini doesn't, your gap is engine-specific. When all five miss you, the gap is foundational.
| # | Brand | MVI | 95% CI | Spread | Per-engine | ChatGPT | Perplexity | Gemini | Claude | Grok | Tier |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | ▲ Evaluation (17/20) · top-brands lists (5/5) ▼ Filtered Discovery (17/30) · premium (0/5) | 76 | 66–82 | 45pp | 16 | 8 | 15 | 17 | 16 | Default choice | |
| 2 | ▲ Comparison (8/20) · top-brands lists (4/5) ▼ Filtered Discovery (10/30) · emerging brands (0/5) | 45 | 34–53 | 50pp | 10 | 2 | 5 | 8 | 12 | First encounter | |
| 3 | ▲ Evaluation (8/20) · budget-friendly (4/5) ▼ Discovery (10/30) · emerging brands (0/5) | 41 | 31–50 | 40pp | 10 | 8 | 9 | 2 | 10 | First encounter | |
| 4 | ▲ Evaluation (10/20) · top-brands lists (3/5) ▼ Comparison (3/20) · premium (0/5) | 36 | 27–46 | 60pp | 10 | 2 | 7 | 13 | 1 | First encounter | |
| 5 | ▲ Discovery (15/30) · filtered persona pro (4/5) ▼ Comparison (2/20) · premium (0/5) | 34 | 26–44 | 70pp | 8 | 1 | 7 | 15 | 3 | First encounter | |
| 6 | ▲ Discovery (14/30) · open recommendation (4/5) ▼ Comparison (3/20) · filtered use case specific (0/5) | 34 | 26–44 | 65pp | 8 | 1 | 7 | 14 | 2 | First encounter | |
| 7 | ▲ Comparison (6/20) · top-brands lists (4/5) ▼ Discovery (6/30) · open recommendation (0/5) | 32 | 23–41 | 30pp | 6 | 1 | 3 | 4 | 7 | First encounter | |
| 8 | ▲ Comparison (3/20) · top-brands lists (3/5) ▼ Filtered Discovery (1/30) · open recommendation (0/5) | 13 | 7–20 | 10pp | 3 | 1 | 2 | 3 | 1 | Not yet cited | |
| 9 | ▲ Comparison (3/20) · open recommendation (1/5) ▼ Evaluation (0/20) | 9 | 4–15 | 40pp | 0 | 8 | 0 | 0 | 0 | Not yet cited | |
| 10 | ▲ Filtered Discovery (4/30) · filtered values driven (4/5) ▼ Evaluation (0/20) · open recommendation (0/5) | 6 | 3–12 | 10pp | 2 | 1 | 0 | 2 | 1 | Not yet cited | |
| 11 | ▲ Filtered Discovery (4/30) · premium (4/5) ▼ Discovery (0/30) · open recommendation (0/5) | 5 | 2–12 | 5pp | 1 | 0 | 1 | 1 | 1 | Not yet cited | |
| 12 | ▲ Filtered Discovery (4/30) · filtered values driven (4/5) ▼ Discovery (0/30) · open recommendation (0/5) | 4 | 2–11 | 5pp | 1 | 1 | 0 | 1 | 1 | Not yet cited | |
| 13 | ▲ Evaluation (2/20) · comparison alternative to leader (1/5) | 4 | 1–9 | 5pp | 0 | 0 | 1 | 1 | 1 | Not yet cited | |
| 14 | ▲ Evaluation (2/20) · evaluation decision criteria (1/5) | 4 | 1–9 | 10pp | 0 | 2 | 0 | 0 | 0 | Not yet cited | |
| 15 | ▲ Comparison (0/20) | 3 | 1–7 | 0pp | 0 | 0 | 0 | 0 | 0 | Not yet cited | |
| 16 | ▲ Comparison (0/20) · filtered use case specific (1/5) | 3 | 1–8 | 5pp | 1 | 0 | 0 | 0 | 0 | Not yet cited | |
| 17 | ▲ Discovery (0/30) | 1 | 0–5 | 0pp | 0 | 0 | 0 | 0 | 0 | Not yet cited | |
| 18 | ▲ Filtered Discovery (1/30) · budget-friendly (1/5) | 1 | 0–6 | 5pp | 0 | 0 | 0 | 1 | 0 | Not yet cited | |
| 19 | 0 | 0–4 | 0pp | 0 | 0 | 0 | 0 | 0 | Not yet cited | ||
| 20 | 0 | 0–4 | 0pp | 0 | 0 | 0 | 0 | 0 | Not yet cited |
Strategic insights for banking
Five derived metrics computed from the same data, surfacing how this segment behaves on AI search. See the State of AI Search for cross-segment comparison.
Engine agreement
0.75
Broad consensus
Effective brands
8.5 / 20
Moderately concentrated · top 2 take 34%
Top demand-leak brand
Axos Bank
+13pp Discovery vs Evaluation
Top mention-only brand
Alliant Credit Union
100% of visibility is mention-only
Kingmaker engine by funnel phase
discovery
Claude
100pp spread
filtered
ChatGPT
83pp spread
comparison
ChatGPT
75pp spread
evaluation
Gemini
100pp spread
For each phase, the engine where the gap between most-cited and least-cited brand is widest, i.e. where positioning matters most. Win that engine, win that phase.
On the record for banking
Pre-registered claim for the next monthly run.
Each of these is a falsifiable, dated prediction we'll grade green or red against next month's data. The full set across all segments lives on the State of AI Search page.
Banking remains a high-agreement segment
PendingOn the next monthly run, banking will retain mean engine agreement at or above 0.75 (currently 0.84). Banking is one of the segments where the five engines materially agree on ranking, and we expect that pattern to hold.
Why we expect this: Banking has unusually consistent canonical sources (Chase, Bank of America, Wells Fargo, Capital One are referenced across nearly every comparative-finance article on the open web). This drives high cross-engine ρ. Categories with weak source consensus (fragrance ρ 0.36, cruise ρ 0.42) are the volatile ones, banking should be among the sticky ones.
Brands cited most across the category
Aggregated across every (brand × prompt × engine) combination tested. The most-cited brands here are the names AI consistently surfaces when buyers ask about banking.
Emerging brands AI is citing in banking
Brand names AI engines surfaced for banking prompts that are not currently on the mapou tracked panel. Ranked by mention count and engine breadth. These are panel candidates, brands AI considers part of the category even though we are not yet measuring them.
Method. Aggregated across the canonical run for banking. For every (panel brand × prompt × engine) we record the brand names the analyzer extracted (capped at 6 per response), then drop names that match the tracked panel or its aliases, plus a denylist of generic category terms. Threshold to qualify: at least 3 mentions across at least 2 of 5 engines. Click any row to see the AI quote that surfaced the brand. Some entries may be tracked elsewhere on mapou but not in this segment, in which case AI considers them cross-category competitors. Reviewed monthly to inform panel additions.
How banking rankings shift by buyer persona
The MVI score is calibrated to a generic shopping-assistant prompt. But buyers don't arrive generically. We re-ran the same 20 canonical prompts five more times, each with a different buyer-persona signal in the system prompt: budget-conscious, premium, working professional, first-time, values-driven. Top-3 overlap with baseline: 60%. Leader holds across all personas: no. Ally Bank loses the #1 spot to a different brand under at least one persona.
| Brand | Baseline | Budget | Premium | Pro | First-time | Values |
|---|---|---|---|---|---|---|
| Ally Bank | 85% | 100% | 30% | 70% | 90% | 60% |
| Discover Bank | 60% | 70% | 15% | 40% | 30% | 0% |
| Bank of America | 55% | 5% | 30% | 60% | 55% | 0% |
| Chase | 50% | 5% | 60% | 45% | 50% | 0% |
| Capital One | 45% | 65% | 5% | 35% | 25% | 10% |
Each cell is the citation rate (out of 20 canonical prompts) for that brand under that persona, ChatGPT only. Cells are tinted green when a brand gains 5+ percentage points vs baseline, orange when it loses 5+. Strong tints flag a 20+ percentage-point swing. Top 8 baseline brands shown; full per-persona data is in data/research/persona-robustness/2026-05-07-1625/. The full methodology is on the State of AI Search page.
The 20-prompt taxonomy
Every brand in this report is tested against the same 20 canonical prompts, spanning the four MVI dimensions (Discovery, Filtered Discovery, Comparison, Evaluation). The prompt set is fixed at methodology v1.0 and reused every monthly run, so MVI deltas are paired comparisons not noise.
The exact prompt templates and phase-weighting formula are part of mapou's proprietary methodology, shared with paying clients alongside custom benchmarks for their specific brand.
See the framework →Methodology v1.0. MVI is mapou's proprietary 0-100 visibility score across 5 AI engines and 4 buyer-intent dimensions. 95% Wilson confidence intervals. Equal engine weighting. See the framework →
Run yours
Want to see your brand on this leaderboard? Run a free visibility check on your own brand. We'll show you exactly which prompts you're missing and which engines are losing you the most ground.