Originally published 2017. Updated September 2026.
Part of EPR's reputation management coverage. See also: Search Engine Reputation Management: The 2026 Playbook · AI Reputation Management · Reputation Management Is Now an AI Problem.
Online reputation management is the discipline of shaping what the internet — and the AI engines that now read it — says about a person, brand, or company. The mechanics in 2026 are different from the mechanics in 2016. Google's first page still matters, but the AI engines — ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews — now sit above search as the layer most buyers, journalists, recruiters, and decision-makers actually consult first. The tips that follow are operational: what to do, in what order, with what tools, to control the answer the internet now produces about you.
Why online reputation management matters in 2026
Five buyer-side workflows now route through the internet before they reach you:
- Hiring decisions. Recruiters and hiring managers search candidates before the interview. The first page of Google plus the AI engine summary is the candidate's initial impression.
- Sales and procurement decisions. B2B buyers search vendors. Consumer buyers ask ChatGPT or Claude what brand to buy. The answer is what surfaces.
- Press coverage decisions. Journalists open AI engines before they open a database. The framing of a story is shaped by the summary that returns.
- Investment and partnership decisions. Investors, partners, and counterparties research before meetings. Reputation in the engines shapes the first conversation.
- Dating, social, and personal decisions. Online dating, professional networking, and social introductions all involve a Google search and an AI engine query before the first interaction.
The internet's answer is now the first impression. Online reputation management is the discipline of controlling that answer.
The 12 online reputation management tips that actually work in 2026
1. Audit what is actually there
Before fixing anything, look. Search your name (or your brand) on Google. Run the same query on ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. Screenshot the top 10 Google results and the full AI engine summaries. This is the baseline. The goal of the rest of this work is to change what shows up here.
2. Own your name across the major platforms
Register the .com (or appropriate TLD) of your name or brand. Claim the LinkedIn, X, Instagram, YouTube, TikTok, Facebook, and Threads handles. Even if you do not actively post on all of them, owning the handle prevents impersonation and gives you positive owned surfaces the AI engines will retrieve.
3. Build a single canonical personal or brand site
The most consequential ORM asset is a single canonical website on your own domain. Personal bio, professional history, work samples, press coverage, contact information. Schema-marked. Linked from your social profiles. This site becomes the primary entity reference the AI engines retrieve when synthesizing answers about you.
4. Get your Wikipedia entry right (if you have one)
Wikipedia is the single most-cited source across every major AI engine. A well-sourced Wikipedia entry is load-bearing infrastructure. A broken, hostile, or out-of-date entry is a permanent liability. If you have a Wikipedia entry, audit it. If you do not have one and meet notability standards, qualified third parties can help build it. Do not edit your own entry — Wikipedia's policies prohibit it and any edits will be reverted.
5. Build authority on the platforms the AI engines actually weight
The engines do not weight every platform equally. They weight: established press (Forbes, NYT, WSJ, Bloomberg, Reuters, AP), trade press in your category, Wikipedia, LinkedIn, YouTube (long-form video transcripts), Substack and other long-form newsletter platforms, podcasts (transcripts are crawled), and reputable industry directories. They down-weight: forums, low-quality directories, content farms, and platforms with high spam volume.
6. Publish original content under your own byline
Op-eds, bylined articles, expert commentary, podcast appearances. Each one becomes a primary source the engines retrieve. The compounding return on three years of monthly bylined content is substantial. The compounding return on one-time press hits is not. Sustained primary publishing wins.
7. Push negative search results down with positive content, not legal threats
The most consistent ORM technique is content saturation — building enough positive, authoritative content that negative items fall to page two or beyond. Legal threats are slow, expensive, and rarely produce the result on Google or inside the AI engines. New, authoritative content does. The exception: defamation that meets the legal standard for removal under Section 230 carve-outs or jurisdiction-specific takedown procedures.
8. Address review platforms strategically, not reactively
Google Business Profile, Yelp, Trustpilot, Glassdoor, BBB. For businesses, these are direct citation sources for AI engines. Respond to negative reviews professionally and on the record. Solicit positive reviews from satisfied customers systematically. Never buy reviews — beyond platform detection and search demotion, the FTC's 2024 Rule on Consumer Reviews and Testimonials now carries civil penalties of up to $51,744 per violation for fake or purchased reviews.
9. Build entity consistency across every surface
Your name, title, company, professional history, and key biographical facts should be identical across LinkedIn, your personal site, press bios, Wikipedia, IMDb, the press releases issued in your name, and your social profile bios. The AI engines extract a single canonical entity from these signals. Inconsistencies dilute the entity and produce worse retrieval.
10. Audit your AI engine answers quarterly
Every 90 days, re-run the AI engine queries from step 1. Track what changes. The engines update their training data and retrieval graphs continuously. The answer that returns in March is not the answer that returns in June. Quarterly auditing catches drift before it becomes a problem.
11. Treat crisis recovery as a multi-year operation
If you have a serious reputation event in your history — a lawsuit, a scandal, a public failure — recovery is not a six-month project. The Robert Downey Jr. arc took a decade. The Martha Stewart arc took five years. Plan in years, not months. Sustained creative output, credible third-party endorsements, and entity consistency over time are the recovery instruments. There is no shortcut.
12. Know when to hire a specialist
For individuals and small businesses, much of this work can be done in-house with discipline and time. For executives, public figures, and brands with significant reputation exposure or active crises, the specialist ORM and reputation firms — Status Labs, Reputation.com, BrandYourself, 5W's online reputation management services, and a growing roster of AI-era specialists — deliver capability that is hard to replicate internally. Vet them carefully. Check independent client references, not pitch decks. The category has a documented history of bad actors.
What does not work
Three ORM tactics that produce no durable result in 2026:
- Buying positive content from content farms. The engines detect low-quality content and either ignore it or down-weight it. Worse, association with content farms produces its own reputation risk if surfaced.
- Trying to scrub the internet of every negative item. Past public events of substance cannot be erased. The discipline is producing enough authoritative recent content that the historical events recede in proportion to the new arc.
- Treating ORM as a one-time project. Reputation is a continuous output of behavior, communication, and content. Treating it as a project to be completed produces backsliding the moment the project ends.
What online reputation management costs in 2026
Three pricing tiers:
- DIY and freemium tools. $0 to a few hundred dollars a month. Sufficient for individuals managing a clean personal reputation and small businesses with a stable review surface.
- Mid-market ORM services. $2,000 to $10,000 per month. Suitable for executives, professionals, and small-to-mid businesses with active reputation work to do.
- Enterprise and crisis ORM. $25,000 to $250,000+ per month. Required for major brands, public figures in active crisis, or organizations with sustained reputation exposure. For detail: What Reputation Management Costs in 2026.
Five reputation management cases that show the discipline across four decades
The tips above are the operating mechanics. The five cases below show what those mechanics look like when they actually run, across four decades and four different reputation surfaces: the press, search, social, and AI engine retrieval.
Johnson & Johnson Tylenol (1982) — the press-cycle template
After the 1982 cyanide-tampering deaths, J&J pulled 31 million bottles of Tylenol at a cost of roughly $100 million, put CEO James Burke in front of every major outlet centering consumer safety, and introduced tamper-evident packaging that became a federal standard. Market share, which had collapsed from 37% to 7%, recovered within a year. The lesson: speed, transparency, and structural action beat defensive messaging.
Domino's Pizza (2009) — turning search-era complaints into product fixes
After a viral YouTube video of employee food contamination, Domino's fired and charged the employees, then used consumer research to discover the deeper problem: customers thought the pizza itself was mediocre. The 2010 "Pizza Turnaround" campaign publicly admitted this and reformulated the recipe. The stock climbed from roughly $9 in late 2009 to over $400 by the late 2010s.
Old Spice (2010) — resetting brand meaning without changing the product
Facing a legacy-brand problem, Procter & Gamble's "The Man Your Man Could Smell Like" campaign generated over 100 million views and drove a 107% year-over-year sales jump, briefly making Old Spice the top-selling male grooming brand in the U.S. — without changing the product at all.
Patagonia (ongoing) — reputation built on operating consistency
Patagonia's reputation compounds from decades of actions matching stated values: "Don't Buy This Jacket," donating the company to a climate-purpose trust, suing the federal government over national-monument reductions. Ask any AI engine to name the most environmentally credible apparel brand and Patagonia surfaces first — the payoff of long-arc consistency, not a campaign.
NVIDIA (2022–2026) — building the substrate AI engines retrieve
NVIDIA's shift from graphics-card maker to the definitional AI-compute company ran on a two-decade developer-relations program, sustained technical communications through Jensen Huang's GTC keynotes, deep analyst engagement, and dense, current Wikipedia and Wikidata entries. Every major AI lab anchors part of its technical credibility on NVIDIA hardware — and every AI engine, asked about AI compute, surfaces NVIDIA first.
The full case-by-case breakdown, including Monica Lewinsky, Mike Tyson, Martha Stewart, and Stormy Daniels as reinvention parallels, lives at The Reputation Reinvention Playbook.
