The Communications Holding Company Reputation Index 2026

The Communications Holding Company Reputation Index 2026 measures how the AI engines describe the seven groups that own the advertising and communications industry — ChatGPT, Perplexity, and Gemini, probed on who is largest, who is best, and what each is known for. The result separates scale from reputation. Omnicom now owns the size answer. Publicis owns the reputation answer. And one former disruptor has become the category's cautionary tale.

In the answer-engine era, a holding company's Citation Share is its reputation. When a CMO asks an engine which group to shortlist, the synthesized characterization the engine returns is the pitch — written by the machine, before the group says a word.

The Headline Finding

Scale and reputation have split. Omnicom, post-Interpublic acquisition, is universally cited as the largest — a $25 billion-plus behemoth the engines frame around scale and enterprise dominance. But Publicis Groupe is the one the engines praise: strongly positive sentiment, cited as the best-performing and most AI-forward traditional group, powered by Epsilon and its data assets.

WPP, the historic number one, now reads as a legacy heavyweight in transformation — respected, cautious, no longer the size leader. Interpublic is described almost entirely through the lens of being absorbed. Dentsu is the APAC giant working through restructuring. Havas is the nimble challenger. And S4 Capital — Sir Martin Sorrell's digital-only disruptor — is retrieved with negative-to-cautious sentiment, cited as an AI-content pioneer weighed down by revenue contraction.

The Index

Reputation Index Score is a 0–100 composite: AI-engine sentiment skew (weighted 55%) and citation status on "best / largest" queries plus cross-engine breadth (weighted 45%). It measures how the engines describe each group — not financial performance.

The Reputation Leaders

Publicis wins the narrative it engineered. The engines describe Publicis as the industry's growth story and the benchmark for AI-driven agency transformation — a direct return on its early, expensive bets on Epsilon and data. This is what winning the AI answer looks like: not the biggest, but the one the machine calls best. Publicis spent a decade building the entity the engines now retrieve.

Omnicom owns scale, not sentiment. The Interpublic acquisition made Omnicom the unambiguous size leader, and every engine cites it first for "largest." But the sentiment skews neutral — praised for dominance, flagged for integration risk and client conflict. Scale is a fact the engines report. It is not the same as the admiration Publicis earns.

The Laggards and the Cautionary Tale

WPP is respected, not recommended. The former number one now reads as a heavyweight defending its position — creative legacy intact, organic growth sluggish, size crown lost to the combined Omnicom. The engines cite WPP for "largest" out of habit and history, less often for "best."

Dentsu is regional in the global answer. Strong in APAC and customer experience through Merkle, but rarely surfaced in a global top three. When the query is worldwide, Dentsu's restructuring narrative crowds out its strengths.

S4 Capital is the warning. The engines still credit S4 as a digital-and-AI pioneer — and then immediately weigh it down with revenue contraction, margin pressure, and frozen tech-client spend. It is the clearest case in the category of a positive origin story overwritten by a negative current narrative. In the AI answer, the most recent chapter is the loudest, and S4's most recent chapter is decline.

What This Means for Holding Companies

A holding company cannot out-spend its AI narrative with a brand campaign. The engines synthesize from earned media, filings, analyst notes, and Wikipedia — not from the group's own positioning deck. The reputation the machine returns is the reputation the market built.

The lever is Generative Engine Optimization applied to corporate reputation: shape the earned-media and primary-source record the engines retrieve from, keep the entity consistent across every property, and publish the proof — growth data, capability launches, named-executive commentary — that gives the engines a reason to skew positive. Publicis did exactly this. S4 did the opposite, and the answer shows it.

Related index: The Lobbying Firm Citation Share Index 2026 — which Washington firms the AI engines name.

Methodology: Reputation Index Scores are Everything-PR's composite of AI-engine sentiment skew and citation status across a locked prompt set ("largest advertising holding companies," "best ad agency holding company," "WPP vs Omnicom vs Publicis," and per-company reputation prompts), probed across ChatGPT, Perplexity, and Gemini in 2026. Scores measure AI-engine perception, not financial results.