Nationwide Ranks Last on AI Placement—but First on Citation Rate

1. The prominence floor

AIVI™ measures placement two ways. Top-3 rate is the share of a brand’s mentions landing in the first three positions of an answer, and it carries 5% of the composite. Position weighting, at 20%, is the finer-grained version: a score in which a higher number means the brand appears earlier in the response. Nationwide is last in the top ten on both.

The numbers are 46.43% and 2.67. The next-lowest top-3 rate in the top ten belongs to Allstate at 55.80%, so the gap between worst and second-worst is 9.37 points — wider than the 4.91-point band that holds the next five brands together, Allstate at 55.80% through Chubb at 60.71%. Against a top-ten median top-3 rate of 60.06% and a median position score of 3.17, Nationwide is not marginally behind on placement. It is the floor.

This is the exact inverse of the first teardown in this series. The Hartford holds the highest top-3 rate in the study, 88.06%, and rank eight overall. Nationwide holds the lowest in the top ten and rank five. Placed side by side the two profiles establish something the composite is trying to say and that a single-metric dashboard cannot: placement matters, but it does not determine the value of an AI appearance by itself. Recommendation strength and citation support can materially change what that appearance is worth.

2. What we measured, and what we did not

This section is deliberately placed before the analysis rather than after it.

• Public model outputs only. Every figure comes from organic brand mentions in responses generated by ChatGPT, Gemini, Perplexity, Copilot, and Claude. No advertising data, no analytics access, no internal reporting from any carrier.
• No client relationship. Brainpan.AI has no commercial relationship with Nationwide. The company was not contacted, did not participate, and did not review this page. It appears here because it is the clearest case in the Q2 2026 top ten of a composite result being carried by something other than reach or placement.
• Organic-only, first-mention attribution. Sponsored placements and paid surfaces are excluded. Where a response names several brands, attribution follows the first-mention rule described in the methodology.
• A sample, not a census. One quarter, 300 stratified prompts, 1,500 model responses. Prompt sets, model versions, and retrieval behavior all move between quarters. Treat quarter-over-quarter movement as directional until a second cycle confirms it.
• Open data, CC BY 4.0. The underlying benchmark is published under a Creative Commons Attribution 4.0 license. Reuse it, check it, disagree with it in public.

Two derived measures appear below. Visibility Points are the sum of position, recommendation, and citation scores; the line-of-business totals in section 6 are computed inside a single line and are not normalized 0–100 AIVI™ scores, so they should not be compared across lines of different sizes. Recommendation points weight a strong recommendation at 5 and a moderate one at 3. Both weightings, and the direction of the position score, are set out in the versioned methodology.

3. Eighty-two placements behind, two points behind

The clearest way to isolate the value of placement is to compare Nationwide with the brand immediately above it. GEICO is the study’s clearest prominence performer outside The Hartford: it converts three-quarters of its mentions into top-three positions and places materially earlier in the answer when it appears.

GEICO wins eight of the twelve rows, including every placement row and every reach row, and it does so decisively — 82 more top-three placements on only 13 more mentions, and a top-3 rate 28.66 points higher. The composite gap at the foot of the table is 2.26 AIVI™ points (the displayed scores of 63.4 and 65.7 round to a 2.3-point gap; the underlying unrounded difference is 2.26). Visibility Points aggregate the position, recommendation, and citation outputs shown in the table above; the AIVI™ score applies the published component weights, folds in Share of Model separately, and normalizes every brand against the category leader. The two figures sit on different scales and should not be compared as though they were the same measurement.

Nationwide takes the other four rows: 11 more recommendation points, 19 more strong recommendations, 25 more citations, and a citation-to-mention ratio 11.75 points higher. Four line items against eight, and the net of the whole table is 2.26 points.

The same arithmetic runs the other way one rank below. Allstate is named 24 more times than Nationwide, holds more share of model, lands 37 more top-three placements, posts a top-3 rate 9.37 points higher, and places earlier on average — and finishes behind Nationwide by 6.72 Visibility Points. Nationwide is the hinge between the two profiles: it loses to the brand above it on prominence and beats the brand below it despite losing on prominence there too. The short version: GEICO wins the prompt, but Nationwide wins the argument.

4. The corroboration engine

Citation influence carries 15% of the composite, and it is the component Nationwide owns outright.

One hundred and nineteen citations from 252 mentions. That is 92% of State Farm’s citation volume produced on 55% of State Farm’s mention volume, and it is a 47.22% citation-to-mention ratio against State Farm’s 28.23% — the highest and the lowest in the published top ten respectively. Against the top-ten median of 40.35%, Nationwide runs nearly seven points clear.

The comparison that matters more is against the carrier population rather than the leaderboard. Sorting the full 134-brand study cohort by entity type gives an all-carrier average of 0.278 citations per mention. Nationwide returns 0.472. It is not that Nationwide gets cited more than its peers by a visible margin; it gets cited about 1.7 times as often per mention as the average carrier in the study, while carrying the sixth-largest mention base in it.

Recommendation strength, at 25% of the composite, points the same direction. Nationwide’s 104 recommended mentions split exactly 52 strong and 52 moderate — 416 recommendation points, third in the top ten at 1.651 per mention, behind only USAA and Amica. Both of those brands are working from a smaller base. Among the six brands in the study with more than 250 mentions, only USAA returns more recommendation value per mention than Nationwide does.

5. Cited—and independently corroborated

Raw citation rates can overstate independent authority when a brand or publisher is frequently supported by its own domain. The benchmark separates raw citation from independently backed citation — the share of a brand’s mentions carrying a citation that survives a filter for brand-owned or affiliated sources, meaning a citation whose source is not controlled by the brand itself. That distinction is where most of the raw citation leaderboard falls apart.

Three publisher domains hold the top of the raw citation panel: MoneyGeek at 100.0%, Bankrate at 97.0%, NerdWallet at 53.2%. On the right panel they fall to 27.9%, 18.2% and 7.3%. NerdWallet retains under a seventh of its own citation-to-mention ratio once brand-owned and affiliated sources are removed. The six carriers in the published set retain nearly all of theirs.

Nationwide has the highest citation-to-mention ratio among the published carriers and retains 43.3 percentage points after brand-owned and affiliated sources are removed — a retention of 92%. Amica’s independently backed rate is marginally higher, at 43.5%, but it is based on less than half Nationwide’s mention volume. Nationwide appears relatively late, but its mentions are supported by independent sources at an unusually high rate: when a model needs to source an insurance claim, Nationwide is disproportionately the carrier a third party has already referenced.

That is worth stating precisely, because it is the diagnosis this page exists to make. Nationwide’s fifth-place result is not built on being mentioned first. It is built on unusually strong recommendation performance and a high rate of independently sourced citation support.

6. Where AI visibility exceeds market share

Aggregate share of model says how much of the category a brand holds. Benchmarking that share against real-world premium tells you where AI visibility runs ahead of it — a representation index against a market-share baseline, not proof that the visibility was earned. The study compares AI visibility share against NAIC 2025 market share for every brand-and-line pair it can, and publishes the five most over-represented.

Nationwide is the only brand in the study appearing twice on that list. Its Auto visibility runs 7.1× its NAIC Auto market share and its Commercial visibility 4.2× its Commercial share. Every other brand on the over-represented list is over-represented in exactly one line, and no brand appears twice on the under-represented list either except Farmers.

The two tables have to be read together, because they say different things. Nationwide is fifth in Auto on 293 Visibility Points — a real position in the largest line reported. In Commercial it is over-represented against market share by 4.2× and still does not reach the published top five, which The Hartford leads on 598 points. Over-representation is a statement about ratio, not about rank: Nationwide’s share of measured AI visibility is substantially larger than its NAIC premium market share in both Auto and Commercial — not evidence that it wins either line.

Read against sections 4 and 5, the shape of the profile is consistent. The pattern holds across two downstream measures: Nationwide’s mentions produce unusually strong recommendation value and unusually high independent citation support, despite relatively weak placement.

7. Strategic diagnostic priorities

Four moves follow from the data above. They are stated as diagnostic direction rather than as an execution plan — sequencing, prompt targets, and content specification are what an audit produces, and they depend on inputs no public dataset contains. The operating frameworks that carry that work — content standard, technical foundation, publishing protocol, and monthly cadence — are published in full at Brainpan.AI frameworks.

Protect the citation asset before optimizing anything else

The highest citation-to-mention ratio in the top ten, with 92% of it surviving a filter for brand-owned and affiliated sources, is likely Nationwide’s most defensible asset — and one of the most exposed to changes in third-party source coverage. It depends on third-party sources that already reference the brand and on retrieval continuing to reach them. The first diagnostic is not a growth question at all: it is an inventory of which independent domains are actually carrying the citations, how concentrated that set is, and what happens to the ratio if the top two disappear.

Treat placement as the isolated defect it is

The measured weakness is prominence: Nationwide is mentioned frequently enough to rank fifth, but it appears later in the generated answer than its top-ten peers. That is a retrieval-order problem — how quickly the material that names the brand is surfaced against a query — and it is diagnosed against specific prompt families rather than against the site as a whole. Late placement in a generated answer typically means the brand is appended as a secondary comparison option rather than surfaced as the lead entity the model reaches for first; closing the gap means earning lead-entity treatment for a wider set of prompt families, not just late-mention inclusion. The 9.37-point gap to the next-lowest top-3 rate is the size of the prize.

Close the Commercial gap between share and standing

Commercial is the line where the two tables disagree most: 4.2× over-represented against market share, absent from the top five on Visibility Points. That combination raises a testable question: is Nationwide appearing across many Commercial answers without generating enough strong recommendations, early placements, or citations to enter the category’s top five? Commercial buyers ask situational questions, and whether a carrier is decisive for them depends on publishing against those situations rather than against a product taxonomy.

Do not read the Auto ratio as headroom

A 7.1× over-representation in Auto is the most flattering number on this page and the most easily misused. It describes a lead over a market-share baseline, not a lead over competitors: GEICO, State Farm, Progressive, and USAA all sit above Nationwide on Auto Visibility Points, three of them by more than double. The ratio is a reason to defend the position, not evidence that the line is solved.

8. Data, method, and corrections

All figures are drawn from the Q2 2026 AIVI™ Insurance Benchmark, a study of 134 tracked insurance-sector brands with scored data across 300 stratified prompts and 1,500 model responses on five engines, producing 4,328 organic brand mentions and 1,742 organic citations, of which 1,348 resolved to a brand within the scored cohort (the remaining 394 resolved to sources outside it — see the entity-type table in the Allstate teardown). The independently backed figures in section 5 are the nine entities published in Finding 04 of the benchmark.

AIVI™ is an organic-only, first-brand-mention composite of share of model (35%), recommendation strength (25%), position weighting (20%), citation influence (15%), and top-three rate (5%), normalized so the category leader scores 100. The position score runs higher-is-earlier: a larger number means the brand appeared closer to the start of the answer. Recommendation points weight a strong recommendation at 5 and a moderate recommendation at 3. The top-ten machine-readable dataset is published at /api/aivi/insurance/q2-2026/top10.json and the per-brand record at /api/aivi/insurance/q2-2026/brands/nationwide.json, both under CC BY 4.0.

The controlling rules behind every number above — prompt-set construction, attribution, the component weights, and the known limitations — are published in the versioned AIVI™ methodology and correction policy. Prompt design follows the generative-engine framing set out in Aggarwal et al., GEO: Generative Engine Optimization (KDD 2024). The published dataset is declared as a schema.org Dataset under CC BY 4.0, and the carrier universe is framed against the National Association of Insurance Commissioners record.

If any figure on this page is wrong, we would like to know. Corrections go to kwalsh@brainpan.ai and are handled under the published correction policy — acknowledged in two business days, determined in ten, and applied to the page with a dated note rather than silently.