Part of the EPR GEO Scorecard series, Everything-PR's quarterly measure of AI Citation Share by vertical.
"McKinsey published its way into AI dominance twenty years before AI existed. Every consulting firm that held its research behind a gate is now invisible to the engines."
Executive Summary
McKinsey 88 (A). BCG 82 (A). Accenture 79 (B+). Bain 75 (B). Deloitte Consulting 72 (B). Oliver Wyman 58 (D). Kearney 55 (D). Roland Berger 51 (D).
A 37-point gap separates the top and bottom of the management consulting category — the widest spread measured in any GEO Scorecard volume to date. McKinsey's dominance is not a surprise. The reason for it is. Decades of freely published research, frameworks, and executive commentary built a retrieval infrastructure the AI engines now treat as ground truth. The firm that invented the modern white paper also invented modern AI citation authority — without trying.
The category ceiling is the highest of any professional-services vertical in the Scorecard series. McKinsey's 88 exceeds every PR holding company (Publicis 79), every law firm, and every B2B-services firm previously scored. Consulting firms invested in public-facing trade research for forty years. That content is now the structural advantage.
The comparison to Vol. 8's law firm study is the sharpest illustration of that advantage. Kirkland & Ellis, the highest scorer among the AmLaw 10, topped out at 72 (B). McKinsey clears that ceiling by 16 points while sitting at the bottom of its own cohort's top tier. Both industries guard client confidentiality. Only one of them also publishes forty years of freely available research that the engines can retrieve.
But the same openness that feeds citation share also feeds controversy retrieval. Asked about McKinsey on any engine, Purdue Pharma, the opioid crisis, and Saudi Arabia appear alongside strategy and operations. The retrieval graph does not forget.
Audience: C-suite executives evaluating consulting engagements, procurement leads running RFPs, PE operating partners selecting advisors, consulting firm CMOs and partners managing brand, MBA candidates researching employers, and communications professionals advising professional-services clients on AI visibility.
15 Publishable Findings
- McKinsey is cited on 94% of strategy-consulting queries across all five engines — the highest single-firm citation rate in any GEO Scorecard volume.
- The 37-point gap between McKinsey (88) and Roland Berger (51) is the widest #1-to-#8 spread in any volume. Consulting has the deepest inequality of any category scored.
- McKinsey's negative-retrieval rate is also the highest in the series. Asked "McKinsey reputation," five of five engines surface Purdue Pharma, the opioid crisis, or Saudi Arabia within the first three sentences.
- BCG is the only firm within 10 points of McKinsey. The growth-share matrix, Bruce Henderson's founding story, and the BCG Henderson Institute give the engines a three-layer retrieval anchor.
- Accenture scores 79 (B+) overall but drops to 64 (C) on pure "strategy consulting" queries. The engines categorize Accenture as technology/IT consulting first. The $64.1B revenue and 774,000 employees are retrieved — the strategy credential is not.
- Bain dominates private equity advisory queries — 5/5 engines name Bain first on "best consulting firm for PE due diligence." Outside PE, Bain drops to third in MBB.
- Deloitte Consulting has the highest raw mention frequency of any firm in the cohort — but the lowest extractability score (62). The engines conflate Deloitte's consulting arm with audit, tax, and advisory. The brand is everywhere. The consulting identity is unclear.
- Asked "which consulting firm should I hire for digital transformation," Accenture is named first on five of five engines. McKinsey is named on four of five. BCG on three of five.
- Perplexity produces the highest scores for six of eight firms — its real-time web-search architecture pulls McKinsey Global Institute reports, BCG Henderson Institute publications, and recent earnings data.
- Oliver Wyman scores 58 (D) overall but 78 (B+) on financial-services-specific queries. The firm's specialist positioning is a GEO asset within its lane — and a liability everywhere else.
- Kearney's 2020 rebrand from A.T. Kearney cost an estimated 8-12 points of citation authority. Three of five engines still retrieve "A.T. Kearney" as the primary entity. The old name outranks the new one.
- Roland Berger is the lowest-scoring firm in the cohort (51, D) — the only major German-origin global strategy firm, penalized by English-language retrieval bias. German-language queries score 15-20 points higher.
- McKinsey Global Institute (MGI) publications are cited more often than any single consulting firm's website across all five engines. MGI is McKinsey's retrieval engine — more valuable than the corporate site.
- Asked "best employer in consulting," Bain ranks #1 on four of five engines — driven by 14 consecutive years atop Glassdoor's "Best Places to Work" list. McKinsey ranks #1 on zero engines for this query.
- The consulting category ceiling (88) is the highest of any professional-services vertical in the series — 9 points above PR holding companies (79) and 16 points above the AmLaw 10's top score, Kirkland & Ellis's 72 in Vol. 8. Consulting firms' forty-year investment in public research created a structural GEO advantage no other services category matches, confidentiality norms included.
Methodology. Scores are based on observed AI engine outputs during a July 2026 test window. Five engines: ChatGPT, Claude, Gemini, Perplexity, Google AI Overviews. 50 prompts per company across five query buckets (Recommendation, Comparison, Capability, Reputation, Corporate), 10 prompts each. Total: 2,000 individual response audits across eight firms. Each response is scored on a binary (cited/not cited) and qualitative (primary mention, secondary mention, absent) basis. Raw observations are normalized into the five-dimension framework: Citation Frequency 40%, Cross-Engine Breadth 20%, Query-Type Breadth 20%, Extractability 15%, Crawl Access 5%. The methodology is identical to Vols. 1–9 and is reproduced quarter-over-quarter. Full protocol at the EPR GEO Scorecard hub.
The Buyer Test: What the Engines Actually Say
We ran the queries a CEO, a PE operating partner, and a board member would actually type. Here are the results.
| Prompt | ChatGPT Names | Perplexity Names | Gemini Names |
|---|---|---|---|
| "Best management consulting firm" | McKinsey first, then BCG, Bain | McKinsey first, then BCG, Bain | McKinsey first, then BCG, Bain |
| "Which consulting firm for digital transformation" | Accenture first, McKinsey second | Accenture first, then Deloitte | Accenture first, McKinsey second |
| "Best consulting firm for PE due diligence" | Bain first (PE heritage) | Bain first, then McKinsey | Bain first (Bain Capital link) |
| "Top strategy consulting firm for financial services" | McKinsey first, Oliver Wyman second | Oliver Wyman first, McKinsey second | McKinsey first, Oliver Wyman third |
| "Largest consulting company in the world" | Accenture ($64.1B revenue) | Accenture (by revenue + headcount) | Accenture (774,000 employees) |
| "McKinsey controversy" or "McKinsey reputation" | Purdue Pharma, opioids, Saudi Arabia | Opioid crisis, Purdue, ICE contracts | Purdue Pharma, Saudi, bankruptcy |
| "Best consulting firm for operations" | McKinsey first, Kearney mentioned | McKinsey first, Kearney third | McKinsey first, Deloitte second |
| "Best consulting firm to work for" | Bain first (Glassdoor #1) | Bain first, then McKinsey | Bain first (Glassdoor 14 years) |
The finding: McKinsey owns the default answer. On any general "best consulting firm" query, five of five engines name McKinsey first. But the category fragments by use case. Accenture owns digital transformation. Bain owns PE advisory. Oliver Wyman owns financial services. Deloitte is mentioned everywhere but named first almost nowhere. MBB is not a brand — it is a retrieval pattern. The engines reproduce it because the training data reinforces it.
The Scorecard
| Dimension (weight) | McKinsey | BCG | Accenture | Bain | Deloitte | Oliver Wyman | Kearney | Roland Berger |
|---|---|---|---|---|---|---|---|---|
| Citation Frequency (40%) | 92 | 84 | 82 | 76 | 78 | 56 | 52 | 48 |
| Cross-Engine Breadth (20%) | 90 | 84 | 80 | 76 | 72 | 60 | 58 | 54 |
| Query-Type Breadth (20%) | 86 | 80 | 72 | 74 | 68 | 58 | 56 | 54 |
| Extractability (15%) | 82 | 78 | 76 | 72 | 62 | 58 | 54 | 52 |
| Crawl Access (5%) | 78 | 80 | 82 | 74 | 70 | 60 | 56 | 50 |
| FINAL GRADE | 88 · A | 82 · A | 79 · B+ | 75 · B | 72 · B | 58 · D | 55 · D | 51 · D |
Engine Heatmap
| Company | ChatGPT | Claude | Gemini | Perplexity | Google AIO | Avg |
|---|---|---|---|---|---|---|
| McKinsey | 90 (A) | 88 (A) | 86 (A) | 92 (A) | 84 (A) | 88 |
| BCG | 84 (A) | 82 (A) | 80 (A) | 86 (A) | 78 (B) | 82 |
| Accenture | 80 (A) | 78 (B) | 76 (B) | 84 (A) | 76 (B) | 79 |
| Bain | 76 (B) | 74 (B) | 72 (B) | 80 (A) | 72 (B) | 75 |
| Deloitte Consulting | 74 (B) | 70 (B) | 72 (B) | 78 (B) | 66 (C) | 72 |
| Oliver Wyman | 60 (C) | 58 (D) | 54 (D) | 64 (C) | 52 (D) | 58 |
| Kearney | 56 (D) | 54 (D) | 52 (D) | 62 (C) | 50 (D) | 55 |
| Roland Berger | 52 (D) | 50 (D) | 48 (F) | 58 (D) | 46 (F) | 51 |
Engine reads. McKinsey is the only consulting firm to score A across all five engines — the first firm in any GEO Scorecard volume to achieve a clean sweep. BCG scores A on four of five (Google AIO 78, B is the exception). Perplexity is the strongest engine for the category — its real-time architecture pulls McKinsey Global Institute and BCG Henderson Institute reports within hours of publication. Google AIO is consistently the weakest, with three firms scoring D or below.
Company Deep Dives
McKinsey & Company, 88 (A)
Founded: 1926 (James O. McKinsey). HQ: New York. Private partnership. Revenue ~$16.4B (2023). ~45,000 consultants. 130+ countries. Global Managing Partner: Bob Sternfels (since 2021).
What's working. McKinsey Global Institute (MGI) is the single most powerful retrieval asset in the consulting category. Forty years of freely published research on labor markets, productivity, AI, and global economics gave the engines a deep, structured knowledge base that no competitor can replicate retroactively. The McKinsey Quarterly — free, ungated, regularly updated — compounds this advantage. McKinsey published its way into AI dominance before AI existed.
Five-of-five-engine consistency on any general strategy query. The "MBB" acronym — McKinsey, BCG, Bain — is itself a retrieval pattern that places McKinsey first in the ordering. Wikipedia depth is extraordinary: the McKinsey article runs 12,000+ words with 300+ citations. Alumni network (Pete Buttigieg, Sundar Pichai, Sheryl Sandberg, Jeff Immelt) reinforces entity linkage on executive queries.
What's hurting. Controversy retrieval is permanent. Purdue Pharma ($573M settlement), the opioid crisis, Saudi Arabia advisory (Jamal Khashoggi), and ICE contracts appear on every reputation query across all five engines. This is not a reputational risk — it is a structural feature of the retrieval graph. The negative content will never decay because the sources (New York Times, ProPublica, congressional testimony) are the same tier-1 outlets the engines trust most.
What moves the score. McKinsey's ceiling is not extractability or breadth — it is controversy dilution. New positive-signal content at MGI publication velocity is the only countermeasure. The firm cannot delete the retrieval. It can only outproduce it. Estimated ceiling: 92 without structural controversy; 88-90 with it.
Boston Consulting Group (BCG), 82 (A)
Founded: 1963 (Bruce Henderson). HQ: Boston. Private partnership. Revenue ~$12.3B. ~32,000 consultants. 100+ offices globally. CEO: Christoph Schweizer (since 2021).
What's working. Three-layer retrieval anchor: the growth-share matrix (the most-cited framework in business-school curricula), the Bruce Henderson founding story (the entrepreneurial origin that engines love to retrieve), and the BCG Henderson Institute (the research arm that mirrors MGI's content-publishing model). BCG's research output is the second-deepest in the category — not as prolific as McKinsey, but freely published and well-structured.
BCG scores A on four of five engines. The firm's AI-specific positioning — its partnership with Anthropic and published AI adoption research — gives it a forward-looking narrative the engines weight heavily on technology queries.
What's hurting. Google AIO (78, B) is the soft spot — thinner Wikipedia-to-publications linkage compared to McKinsey. BCG does not own a single dominant query category the way McKinsey owns "best consulting firm" or Bain owns PE advisory. BCG is always second — a strong second, but second.
What moves the score. Henderson Institute publication velocity. Schema-level Organization markup linking BCG.com to its research output. A signature proprietary framework launch — something the engines can index alongside the growth-share matrix. Estimated lift: 3-5 points within two quarters.
Accenture, 79 (B+)
Founded: 1989 (as Andersen Consulting; Accenture since 2001). HQ: Dublin (operationally global). NYSE: ACN. Revenue ~$64.1B (FY2024). ~774,000 employees. CEO: Julie Sweet (since 2019).
What's working. Absolute dominance on digital transformation and technology consulting queries — five of five engines name Accenture first. Revenue ($64.1B) and headcount (774,000) are retrieved on every corporate query. The engines treat Accenture as the largest consulting firm in the world by default. Julie Sweet's CEO profile is well-anchored — Accenture's public-company disclosure requirements feed the retrieval graph quarterly.
What's hurting. Strategy-consulting identity gap. Asked "best strategy consulting firm," three of five engines do not name Accenture at all. The engines categorize Accenture as IT consulting/technology services first, strategy second. The Andersen Consulting origin story still surfaces on historical queries — the Arthur Andersen and Enron connection is retrievable. Extractability drops on "MBB" queries because the engines have learned that MBB does not include Accenture.
What moves the score. Repositioning content that bridges technology consulting to strategy consulting. Accenture Strategy-branded research at MGI-level publication velocity. Wikipedia enhancement emphasizing the strategy practice alongside technology. Estimated lift: 4-6 points within two quarters if strategy narrative is prioritized.
Bain & Company, 75 (B)
Founded: 1973 (Bill Bain, ex-BCG). HQ: Boston. Private partnership. Revenue ~$6.5B. ~18,000 employees. Global Managing Partner: Manny Maceda (since 2018).
What's working. Private equity advisory dominance. Five of five engines name Bain first on PE-specific queries. The Bain Capital connection — founded by Bain & Company partners in 1984, including Mitt Romney — gives the engines a PE retrieval anchor no competitor can match. Bain's Glassdoor record (14 consecutive years as #1 Best Place to Work) makes it the default answer on employer-brand queries. Bain Brief publications are well-structured and ungated.
What's hurting. Narrower general-consulting citation breadth. Outside PE and employer-brand queries, Bain drops to third in MBB. Revenue ($6.5B) is half of BCG and a fraction of McKinsey — the engines retrieve this as a scale limitation. The Mitt Romney political association surfaces on reputation queries, particularly during election cycles. Bain's Wikipedia article is significantly shorter than McKinsey's or BCG's.
What moves the score. Broader industry-vertical research publications. Wikipedia depth enhancement. Increased executive commentary beyond PE advisory. Estimated lift: 3-5 points within three quarters.
Deloitte Consulting, 72 (B)
Founded: 1845 (William Welch Deloitte; consulting arm formalized 1990s). HQ: London (Deloitte Touche Tohmatsu); consulting operations distributed globally. Revenue: part of Deloitte's $67.2B total (consulting estimated ~$26B). Largest Big 4 consulting practice by headcount.
What's working. Highest raw mention frequency in the cohort. Deloitte appears in more engine responses than any other firm — the Big 4 brand is retrieved on audit, tax, advisory, risk, and consulting queries alike. Deloitte Insights (the research platform) publishes at high volume. Deloitte's public-sector consulting practice is the most frequently cited on government-related queries.
What's hurting. Extractability is the structural problem. Deloitte Consulting scores 62 on extractability — the lowest in the top five — because the engines cannot cleanly separate "Deloitte the consulting firm" from "Deloitte the audit firm" or "Deloitte the tax advisor." Asked "best consulting firm," the engines often cite "Deloitte" without specifying the consulting arm. This conflation dilutes the consulting brand's authority. Google AIO (66, C) is the weakest engine — it defaults to the parent entity rather than the consulting practice.
What moves the score. Entity disambiguation. Deloitte Consulting needs its own distinct Wikipedia presence, its own schema markup, and its own research-publication brand that does not share a URL with Deloitte audit or Deloitte tax. Estimated lift: 6-8 points within two quarters if entity separation is prioritized.
Oliver Wyman, 58 (D)
Founded: 1984 (as Mercer Management Consulting; Oliver Wyman since 2007). Part of Marsh McLennan ($23B+ total revenue). Revenue ~$3B+. Financial services specialist.
What's working. Specialist positioning is a narrow advantage. On financial-services-specific queries — "best consulting firm for banking" or "top risk management consultants" — Oliver Wyman scores 78 (B+). Marsh McLennan's public-company disclosures feed the retrieval graph. The Oliver Wyman Forum publications are well-structured.
What's hurting. The specialist advantage is also the ceiling. Outside financial services and risk, Oliver Wyman is absent from engine responses. Four of five engines do not mention Oliver Wyman on general "best consulting firm" queries. The Marsh McLennan parent brand is more visible than Oliver Wyman itself — but the parent is not a consulting brand in the engines' classification. Wikipedia depth is thin. Cross-engine breadth (60) is the lowest dimension for any firm scoring above F.
What moves the score. Cross-industry research publications. Wikipedia enhancement. Earned media in tier-1 business outlets on topics beyond financial services. Estimated lift: 8-12 points within three quarters.
Kearney (f/k/a A.T. Kearney), 55 (D)
Founded: 1926 (Andrew Thomas Kearney). HQ: Chicago. Private partnership. Revenue ~$1.8B. Operations and procurement specialist.
What's working. Operations and procurement queries retrieve Kearney in three of five engines. The firm's annual FDI Confidence Index is a well-cited proprietary data product. Founded the same year as McKinsey — the 100-year heritage is a retrieval asset when it surfaces.
What's hurting. The 2020 rebrand from A.T. Kearney to Kearney is a measurable citation-share loss. Three of five engines still retrieve "A.T. Kearney" as the primary entity name. The old name has deeper Wikipedia anchoring, deeper media archival presence, and deeper alumni linkage. The rebrand effectively split the firm's retrieval graph in two. Google AIO (50, D) is the floor — it retrieves neither name consistently.
What moves the score. Entity consolidation — redirects, Wikipedia disambiguation, schema markup linking the old name to the new. Operations-specialist research at higher publication velocity. Estimated lift: 10-15 points within three quarters if entity consolidation is aggressive.
Roland Berger, 51 (D)
Founded: 1967 (Roland Berger). HQ: Munich. Private partnership. Revenue ~€1B+. The only major German-origin global strategy firm.
What's working. European industrial and automotive consulting queries retrieve Roland Berger in two of five engines. The firm's founder-named brand carries a distinctive origin story — the engines retrieve it on "European consulting firms" queries. German-language queries score 15-20 points higher than English-language queries.
What's hurting. English-language retrieval bias is the structural ceiling. The engines are predominantly trained on English-language corpora. Roland Berger's research publications, media coverage, and Wikipedia depth are significantly thinner in English than in German. Asked "best consulting firm" in English, zero of five engines name Roland Berger in the first response. Google AIO (46, F) and Gemini (48, F) are the two lowest single-engine scores in the entire cohort. The firm is invisible to the English-language AI buyer.
What moves the score. English-language research publication at scale. English Wikipedia enhancement. Tier-1 English-language earned media on European industrial topics where Roland Berger has genuine expertise. Estimated lift: 12-18 points within four quarters if English-language content is prioritized.
Biggest Winners
| Winner | Why |
|---|---|
| McKinsey on general strategy queries | 5/5 engines name McKinsey first on "best consulting firm." 94% citation rate across all strategy queries. MGI is the single most powerful retrieval asset in professional services. |
| BCG on research-driven queries | Henderson Institute + growth-share matrix = three-layer retrieval anchor. A on four of five engines. The strongest #2 position in any Scorecard volume. |
| Accenture on digital/technology queries | 5/5 engines name Accenture first on digital transformation. $64.1B revenue and 774,000 employees retrieved on every corporate query. Owns the technology-consulting answer. |
| Bain on PE advisory + employer brand | 5/5 engines on PE due diligence. 4/5 engines on employer brand. Two distinct retrieval lanes no competitor can touch. |
| The consulting category vs. all other services | Category ceiling (88) is the highest of any professional-services vertical. Forty years of published research created a structural GEO advantage. |
Biggest Risks
| Risk | Who It Hurts | Severity |
|---|---|---|
| Permanent controversy retrieval (Purdue Pharma, opioids, Saudi Arabia) | McKinsey | High |
| Entity conflation with audit/tax/advisory | Deloitte Consulting | High |
| Strategy-consulting identity gap on AI queries | Accenture | Medium |
| Rebrand entity fragmentation (A.T. Kearney → Kearney) | Kearney | Medium |
| English-language retrieval bias | Roland Berger | Medium |
| Narrow PE/employer-brand lane limits general-consulting growth | Bain | Medium |
| Specialist ceiling caps total citation share | Oliver Wyman | Medium |
Q4 2026 Predictions
Who gains next quarter: BCG (+2-4 points). Henderson Institute publication velocity is accelerating. AI-partnership narrative (Anthropic, Google Cloud) is fresh and retrieval-ready. BCG is the most likely firm to close the gap on McKinsey.
Who holds steady: McKinsey (88 +/- 1 point). The ceiling is controversy-constrained. New MGI research adds citation share. New controversies — or renewed media coverage of old ones — subtract it. Net effect: flat.
Biggest upside potential: Deloitte Consulting (+6-8 points) if entity disambiguation is executed. The raw mention frequency is already the highest in the cohort. The consulting identity just needs to be separable from the Big 4 parent. This is the highest-leverage single intervention in the volume.
Biggest wildcard: Accenture's strategy narrative. If Accenture Strategy publishes at MGI-level velocity and earns tier-1 media on strategy topics (not just technology), the firm could leapfrog Bain and challenge BCG within two quarters. The revenue base ($64.1B) and headcount (774,000) already exceed MBB combined. The identity is the only gap.
Most likely first A+ grade: McKinsey — but only if controversy retrieval stabilizes and MGI publication velocity holds. No consulting firm has scored above 88 in any test window to date.
Action Items by Audience
| Audience | What This Means | What to Do |
|---|---|---|
| CEO / C-Suite | Your board members and investors are asking AI engines which consulting firm to hire before they ask you. | Ask ChatGPT "which consulting firm should I hire for [your problem]" and see who appears. If your preferred firm is absent, that is an AI visibility problem your advisor should solve. |
| PE Operating Partners | Bain owns the PE advisory answer. If you use another firm, the AI engines are not reinforcing your choice. | Include AI citation share in your consulting RFP evaluation. Ask each firm: "What does ChatGPT say about you on the queries that matter to our portfolio?" |
| Consulting Firm CMOs | Published research is now the primary driver of AI citation share. Gated content is invisible to the engines. | Ungate your research. Publish at velocity. Structure for retrieval: entity-rich, schema-marked, prompt-oriented headlines. Every white paper behind a registration wall is citation share you are leaving to McKinsey. |
| Consulting Firm Partners | Your personal AI visibility affects the firm's score. The engines retrieve named partners on capability queries. | Publish bylined commentary in tier-1 outlets. Ensure your LinkedIn profile and firm bio are structured for retrieval. Your name is a retrieval anchor. |
| MBA Candidates | The employer-brand answer is already written. Bain wins on Glassdoor. McKinsey wins on prestige. BCG wins on research culture. | Ask Claude, Gemini, and Perplexity "best consulting firm to work for" and compare the answers. The engines surface what the recruiting brochures don't. |
| Communications Advisors | Consulting firms prove the rule: public content = citation share. Every gated asset is invisible revenue. | Build your clients' GEO strategy around ungated research, structured data, and entity-rich earned media. The McKinsey playbook is the template. |
About the EPR GEO Scorecard Series
The EPR GEO Scorecard Series applies a single locked five-dimension framework to one consumer or industry vertical at a time. Published volumes: Beauty (Vol. 1), Hotels & Hospitality (Vol. 2), Luxury Brands (Vol. 3), Streaming & Entertainment (Vol. 4), QSR (Vol. 5), Consumer Tech (Vol. 6), PR Holding Companies (Vol. 7), Law Firms (Vol. 8), Management Consulting (Vol. 9), Executive Search Firms (Vol. 10), and Talent & Literary Agencies (Vol. 11). Quarterly. The methodology hub: everything-pr.com/epr-geo-scorecard.
Part of Everything-PR's Citation Share Index and Generative Engine Optimization research.
