Online reputation management
Online reputation management (ORM) is the work of shaping what people see when they search your brand: page one of Google, review sites, Reddit, Quora, YouTube and what ChatGPT says. I run ORM for Masai School.
- youOfficial site
- youLinkedIn profile
- youGoogle Business Profile
- themReview aggregator
- youPress coverage
- youYouTube channel
- themForum thread
- youKnowledge panel
Six of eight owned
- 0-97%
- of consumers read reviews before buying, across surveys
- ~10
- slots on page one of a branded search
- 0
- workstreams a real ORM programme covers
- 0-12mo
- before a one-time cleanup decays without upkeep
Range across published consumer surveys
Google SERP
ORM framework, 2026
Practitioner estimate
ORM does not make criticism disappear. It makes what people find about you accurate and easy to trust. Many brands buy only review management, or only suppression, and the rest stays broken. Monitoring shapes replies, replies bring more reviews, more reviews lift the rating, and pages you own take the page-one slots. AI assistants read the same sources.
- Most legitimate negative reviews cannot be removed, so response quality and review velocity are the actual levers
- A branded search is roughly ten slots, and the assets you control are the only ones you can move
- AI assistants synthesise reputation from the same public sources search engines use
- A cleanup with no monthly rhythm decays inside a year
How the work actually runs.
Six phases. Each one produces something you can hold, not a status update.
- 01
Audit the branded SERP
Every result on page one of your brand name, logged and classified as owned, neutral, or hostile, in a logged-out browser so personalisation does not flatter the result.
- Owned versus earned versus hostile
- Review profiles and aggregate ratings
- What ChatGPT and AI Overviews currently say
- 02
Fix the review engine
A compliant way to ask for reviews at the right moment, plus reply templates by complaint type. Most reviews cannot be deleted, so volume and replies are what you control.
- Policy-compliant request flow
- Response frameworks by complaint category
- Escalation path for the removable minority
- 03
Build the owned stack
The assets that can legitimately take page-one slots: the site, profiles, press, video, and structured data that ties the entity together.
- Entity and Organization schema
- Profiles that actually rank
- Video and press as slot-holders
- 04
Run it monthly
Monitoring and replies every month, and a full re-audit each quarter, including AI answers. Without this, a cleanup fades within a year.
- Mention monitoring across the niche
- Monthly SERP diff
- Quarterly AI answer audit
Branded SERP audit
Every page-one result classified, with the gap between what you own and what you could.
Review operating system
Request workflow, timing, and a response library by complaint type.
Owned-asset roadmap
Which assets to build, in which order, to take the most slots for the least work.
AI answer baseline
What the major assistants say about the brand today, so drift is visible later.
Monthly rhythm
The operating cadence that stops the whole thing decaying.
The shortcuts are what get brands removed. These are the rules the work runs under.
- Buy or incentivise reviewsEarn velocity with a compliant request flow and a real trigger. Fabricated reviews breach every platform policy and cost the whole profile.
- Gate reviews by sentimentAsk every customer the same way. Routing happy customers to Google and unhappy ones to a private form breaches Google policy.
- Promise removalsAppeal only genuine policy breaches, and answer everything else. Most legitimate negatives cannot be removed by anyone.
- Build thin pages to push results downBuild assets that deserve the slot. Suppression pages do not hold and they make the entity messier for the models.
Each is measured against a baseline taken before any work starts.
Owned share of page one
Slots you control on a branded search, logged out
Review velocity
New reviews per month, not lifetime total
Response rate and time
Share answered, and how fast
AI answer accuracy
What assistants say, re-checked quarterly
A read of where you stand on ORM right now, what the gap is, and the three things worth doing first. It is free because it is also how I work out whether I can help.
- Where your brand appears today, and where competitors appear instead
- The specific threads or questions worth owning first
- What the first ninety days would actually involve
Can you remove a negative Google review?
Only if it breaches Google's policies, and most legitimate complaints do not. The realistic path is the redressal and appeal process for genuine violations, and response plus velocity for everything else. Anyone promising removal as a service is describing an outcome they do not control.
How long does ORM take to show results?
Review velocity and response metrics move within weeks. Page-one composition usually takes three to six months because it depends on assets ranking. AI answers lag further, since they are downstream of everything else.
Is ORM different from SEO?
Same mechanics, different target. SEO wins visibility for what you sell. ORM shapes what appears for your name.
Does ORM affect what AI assistants say about us?
Yes, indirectly. Assistants draw on reviews, press and your own pages, so the work that fixes page one also changes what they repeat.
Do we need this if we have no negative reviews?
Yes. Building pages that hold your branded search takes months, and you will not have months during a crisis.
- AEOReputation now includes what an assistant says when nobody is watching.
- RedditThe community half, where criticism is specific and public.
- QuoraWhere an unanswered question about you becomes the answer.
- LinkedInThe employer-brand half, and the one candidates check first.
- YouTube ORMWhere the comment under a funded video is the review nobody moderates.