On this page
Online reputation management now includes what AI assistants say about your brand. ChatGPT, Gemini and Google AI Overviews build their answers from reviews, directories and third-party articles. So the work is auditing those answers, fixing the sources behind them, and earning real reviews. Fake or paid reviews break FTC rules in force since October 21, 2024.
We first published this guide in 2023 as a list of seven reputation strategies. Most of the original advice still holds: monitor what people say, ask for honest reviews, respond to every one, and never fake them. What changed is where people read about you. A buyer who used to scan your star rating and the first page of Google may now ask an assistant, “Is this company any good?” and get a one-paragraph verdict. That verdict is built from sources you can influence, which is why reputation work now overlaps with local SEO and AI search.
This rewrite keeps the seven strategies, updates them for 2026, and adds a brand prompt audit you can run yourself.
What does online reputation management cover now?
Online reputation management covers every place a buyer forms an impression of you before they talk to you: reviews, branded search results, social channels, directories, press and now AI-generated answers. The job is to make sure each of those places is accurate, current and fair, and to respond well when something negative shows up.
Here’s how the scope has shifted:
| Area | What it meant before | What it includes in 2026 |
|---|---|---|
| Reviews | Star rating and volume on major sites | Rating, volume, recency, response quality, and how review themes get summarized by AI |
| Branded search | Owning page one for your brand name | Page one plus the AI Overview that may sit above it |
| Social | Responding to complaints quickly | Same, plus knowing which public threads AI tools may pull from |
| Directories and listings | Consistent name, address and phone | Consistent facts everywhere, since assistants cross-reference them |
| AI answers | Not a channel | What ChatGPT, Gemini, Perplexity and AI Overviews say when asked about you |
Most SEO work is really reputation and brand work, and AI search makes that literal. It acts like PR for your business across the entire internet, and every source needs to tell the same accurate story.
How do AI assistants form an opinion of your brand?
AI assistants describe your brand by combining what they learned in training with what they find on the web when the question is asked. For a brand question, the sources they pull from tend to be your own site, review platforms, directories, news coverage, comparison articles and community discussions. The answer reflects the balance of those sources, not just your homepage.
From what we’ve seen running brand prompts across engines, a few patterns repeat:
- When review sites show a recurring complaint, assistants often mention it, even if the complaint is a few years old.
- When facts disagree across sources (founding year, services, locations, pricing approach), answers get vague or pick the wrong version.
- When your own site doesn’t answer a common question, the assistant answers it from someone else’s page.
- Thin brands, with few third-party mentions, get thin answers or get confused with similarly named companies.
We measure this in two ways: mentions (your brand named in the answer) and citations (a link to a source). Our State of AI Visibility 2026 report tracked both across 55 brands and five AI engines, and the two numbers can tell different stories. For reputation, citations are especially useful because they show you exactly which page the assistant trusted. If that page is an outdated directory listing or a critical forum thread, you now know what to fix.
How do you run a brand prompt audit?
Run a brand prompt audit by asking each major assistant the same set of questions about your brand, recording what it says and which sources it cites, and flagging anything inaccurate, outdated or unfairly negative. Do it before you change anything so you have a baseline, then repeat on a schedule.
- Write 10 to 20 prompts a real buyer would use. Mix neutral, comparison and skeptical questions.
- Run each prompt in ChatGPT, Gemini, Perplexity and Google (to see AI Overviews and AI Mode).
- Record the answer, whether it’s accurate, its tone, and every cited source.
- Group the problems by source, since one bad source often causes several bad answers.
- Assign each fix to an owner: website, listings, reviews, PR or customer service.
Here’s a template you can copy into a spreadsheet:
| Prompt | Engine | Brand mentioned? | Accurate? | Tone | Sources cited | Fix and owner |
|---|---|---|---|---|---|---|
| What is [brand] known for? | ChatGPT | Yes / No | Yes / Partly / No | Positive / Neutral / Negative | List URLs | e.g. Update About page (web team) |
| Is [brand] legit? | Gemini | |||||
| [Brand] reviews and complaints | Perplexity | |||||
| [Brand] vs [competitor] | AI Overviews | |||||
| Best [category] in [area or industry] | AI Mode | |||||
| How much does [brand] cost? | ChatGPT |
Answers vary from one run to the next, so don’t panic over a single bad response. Look for patterns across engines and across repeat runs. A wrong fact that shows up in three engines almost always traces back to a source you can find and correct.
What’s allowed when you ask for reviews?
You can ask any customer for an honest review, and you should make it easy. What you can’t do is fake reviews, pay for positive ones, hide who wrote them, or pressure people out of leaving negative ones. Those rules come from the FTC in the US and from each review platform’s own policies, and Google’s are stricter than the FTC’s on incentives.
The FTC announced its final rule on fake reviews and testimonials on August 14, 2024. According to the Federal Register notice, it took effect October 21, 2024. The rule bans:
- Fake reviews and testimonials, including AI-generated ones.
- Buying reviews when the incentive is conditioned on a particular sentiment, positive or negative.
- Reviews from company insiders, such as employees, without clear disclosure.
- Company-controlled review sites that claim to be independent.
- Suppressing reviews through threats or intimidation.
- Buying or selling fake social media followers or views.
Google’s Maps content policy goes further. It doesn’t allow fake engagement, and it prohibits paid or incentivized reviews, review gating (only asking customers you expect to be happy), and discouraging negative reviews. So a “leave a review and get 10% off” offer that might be arguable under the FTC rule is still against Google’s policy.
The safe, effective process looks like this:
- Ask every customer, not just the ones you think loved you.
- Ask at a natural moment: after delivery, after a resolved support ticket, after a stay or a move-in.
- Send a direct link to the review page so it takes under a minute.
- Don’t offer anything in exchange.
- Have employees disclose their relationship if they ever post about you.
Real reviews are slower to collect than bought ones. They’re also the only kind that hold up, and they’re the kind AI assistants summarize when someone asks whether you’re any good. Manufacturing signal poisons the very measurement you’re trying to improve.
How should you respond to negative reviews?
Respond to negative reviews quickly, calmly and briefly: acknowledge the problem, apologize where it’s warranted, say what you’re doing about it, and move the details to a private channel. The reply is for future readers as much as for the reviewer, and a good one often does more for trust than the review did damage.
A reply structure that works:
- Thank them for the feedback, using their name if it’s shown.
- Name the specific issue so it’s clear you read the review.
- Apologize for the experience if it fell short, without arguing over details in public.
- Say what changed or what you’ll do.
- Give a direct contact (a named person or a specific email) to continue the conversation.
A few things to avoid. Don’t share private customer information in a public reply, even to correct the record. Don’t copy and paste the same reply on every review, because readers notice. Don’t threaten legal action over an honest negative review; beyond looking bad, using threats to suppress reviews is one of the practices the FTC rule bans.
If a review is fake or breaks the platform’s policy, report it through the platform’s process and reply once, politely, noting you can’t find a record of the customer. Then leave it. For senior living, healthcare and other sensitive categories, have whoever handles privacy review your response templates before anyone uses them.
What are the 7 reputation management strategies that work in 2026?
The seven strategies that work are: monitor every source, ask for honest reviews, respond to every review, handle social complaints fast, never fake or suppress reviews, own your branded search results, and manage what AI says about you. The first six come from our original guide and still hold. The seventh is now the one that ties the rest together.
1. Monitor what people say about you
Set up alerts for your brand name and key people, track the review platforms that matter in your industry, and use social listening where volume justifies it. Assign an owner to each source and a check schedule, because monitoring without an owner turns into nobody checking.
2. Ask happy customers, and everyone else, for honest reviews
Build review requests into your normal customer process so they go out automatically. Ask all customers, make it easy, and offer nothing in return. Steady recent reviews matter more than a burst from three years ago.
3. Respond to every review
Thank positive reviewers specifically, not with a template. Handle negative ones with the structure above. Response rate and response time are visible to anyone reading your profile, and the replies themselves become part of the text that gets summarized.
4. Handle social complaints quickly
Reply publicly within hours, resolve the issue privately, then post a short public follow-up once it’s fixed. Keep a written escalation plan for serious issues (safety, legal, data privacy) so the person on social knows exactly who to call.
5. Never fake or suppress reviews
This one is now a legal line, not just an ethical one. No fake reviews, no AI-written reviews, no sentiment-conditioned incentives, no undisclosed insider reviews, no intimidation, no fake followers. Platforms remove fake reviews, and the FTC rule gives regulators a clear basis to act.
6. Own your branded search results
Make sure the first page for your brand name is filled with pages you control or can influence: an accurate website, consistent social profiles, a complete Google Business Profile where relevant, case studies, press and partner pages. Good SEO on those assets pushes outdated or inaccurate results down over time.
7. Manage what AI tools say about you
Run the brand prompt audit above, then fix the sources behind any bad answer. In practice that means keeping company facts identical everywhere, publishing direct answers on your own site to the questions buyers ask (including how your pricing works and what your policies are), correcting outdated listings, and earning genuine mentions on the sites assistants already cite. This is the core of our answer engine optimization work, and it’s where reputation management and AI search are now the same project.
How does personal reputation management work for executives and founders?
Personal reputation management is the same work applied to a person’s name: find what Google and AI assistants say about you, publish accurate pages you control, and handle harmful results through the right channel. For founders and executives, the person and the company often show up in the same searches, so one shapes how people see the other.
- Search your name the way others do. Search your name on its own and with your company, title, and city in Google. Then ask ChatGPT, Gemini, Claude, and Perplexity who you are. Note anything outdated, wrong, or mixed up with someone who shares your name.
- Own the first page for your name. A complete LinkedIn profile, a personal site, and a bio page on your company’s website give search engines and AI assistants accurate sources to draw on. Keep your title, employer, and background identical on all of them.
- Add Person schema to your bio page. Structured data with your name, job title, employer, and links to your official profiles helps search engines connect those profiles to you. It has to match what the page says.
- Remove your personal contact details from search. Google’s Results about you tool finds Search results that show your home address, phone number, or personal email and lets you request their removal. It removes the result from Google Search, not the page from the web. Google says it won’t remove information you control yourself, or results it considers valuable to the public, such as government, educational, news, or business websites.
- Answer criticism with facts. Google’s personal information removal options cover specific cases like contact details. They aren’t a way to remove criticism. Where a claim about you is false, ask the publisher for a correction, and keep publishing accurate, well-sourced content of your own.
How often should you monitor your reputation?
Check reviews and social mentions weekly, branded search and listings monthly, and AI answers monthly with a deeper review each quarter. Higher-volume brands, or any brand in the middle of an issue, should tighten that cadence. The point is to catch a problem while it’s one review, not after it has become the summary.
| Source | Cadence | What to measure | Typical owner |
|---|---|---|---|
| Review platforms | Weekly | New reviews, rating, response rate, response time | Customer experience or location manager |
| Social mentions | Daily to weekly | Complaints, sentiment, time to first reply | Social team |
| Branded search results | Monthly | What ranks on page one, any new negative results | SEO team |
| Listings and directories | Monthly | Name, address, phone, hours, services, categories | Local SEO team |
| AI answers | Monthly, deeper review quarterly | Mentions, citations, accuracy, tone, sources cited | SEO or AEO team |
Report these together, not in separate decks. A falling AI accuracy score often traces back to a listing problem, and a dip in rating often explains a shift in AI tone. Seeing them side by side is what makes the fix obvious.
FAQ
What is online reputation management?
It’s the ongoing work of monitoring and improving how your brand appears in reviews, search results, social media, directories and AI-generated answers. The goal is accuracy and fairness, backed by real customer experiences.
Can I offer a discount in exchange for a review?
The FTC rule bans incentives conditioned on a particular sentiment. Google’s Maps policy prohibits paid or incentivized reviews entirely, so for Google reviews the answer is no.
Can I get a negative AI answer about my brand removed?
Usually not directly. You change the answer by fixing the sources behind it: correcting listings, updating your own pages and earning accurate third-party coverage.
Should I respond to old negative reviews?
Yes, if they’re still visible. A calm, specific reply shows future readers how you handle problems, even if the original reviewer never sees it.
How do I manage my personal reputation online?
Search your name in Google and in AI assistants, keep your LinkedIn profile, personal site, and company bio accurate and consistent, and add Person schema to your bio page. Use Google’s Results about you tool to request removal of search results that show your personal contact info, and ask publishers to correct anything false.
Want to see what AI says about your brand?
We’ll run the check for you. Request a free growth audit and within about five business days you’ll get three to five ranked opportunities and a 90-day roadmap, including what AI assistants currently say about you and which sources are shaping it.