# Reputation management brief: a template for agencies A reputation management brief you can send to agencies: the fields to fill, the goals and limits to set, and how to score the proposals that come back. --- A reputation management brief is a one to three page document that tells an agency who is being protected, what is wrong in search results and AI answers, what outcome you want, what methods are off limits, and how you will judge the work. Send the same brief to every agency you speak to, because proposals can only be compared when they answer the same questions. Most buyers skip this step and describe the problem on a call. The agency then fills the gaps with its own assumptions, and each proposal solves a slightly different problem at a different price. A written brief moves that work to you, where it is cheap, and it exposes how each agency thinks before any money moves.

The short version

Write the brief before you contact anyone: subject, audience, evidence file, goal, constraints, success measures, decision process.
Separate what is inaccurate, outdated or unlawful from what is accurate but unflattering. Only the first group is a removal or correction target.
State the goal as a change in what people find, never as a promise of deletion.
Put lawful methods and confidentiality in writing, and ask each agency to confirm them in its proposal.
Fix the scoring weights before the proposals arrive, and treat any promised outcome or rule-breaking method as an automatic fail.

## A reputation management brief fixes the problem before anyone prices it A brief is the document that defines the problem, the outcome and the limits, so that every agency answers the same question. It is not a request for a price. A price attached to a vague problem is a guess, and you cannot compare guesses. The brief does three jobs. It forces you to decide what you actually want, which is often different from "make it go away". It gives the agency enough evidence to diagnose rather than sell. And it gives you a fixed yardstick: when proposals come back, you read each one against your brief, not against the best-sounding pitch. ## Build the evidence file before you write a word of the brief The evidence file is the dated record of what exists today, and it is the part of the brief that most changes the quality of the proposals you receive. An agency that can see the real search results can diagnose. One that cannot will describe a generic program. Collect these items first: - **Your own searches.** Search your name, your company name and your name plus the obvious qualifiers in a private window, on more than one search engine, and note the position of each result. Our free page one score tool gives you a first read if you want a fast baseline. - **AI answers.** Ask two or three AI assistants what they say about you or your company, with the exact prompts you used, and save the replies with the date. - **A list of problem items.** For each: the address, the type of page (news article, forum thread, complaint listing, review, social post, profile), the date you first saw it, and where it ranks. - **A claim-by-claim reading.** For each item, write the specific statement that is wrong, outdated or unlawful. If you cannot name a statement, write "accurate, unflattering". This line decides which route is even possible. Keep the file shareable but not public. If it contains sensitive material, such as private messages or medical, financial or family details, do not attach it to the first email. Ask for a mutual confidentiality agreement first. Our standard mutual NDA shows what a reasonable one covers. ## The brief template, field by field The template has ten fields, and each one answers a question an agency would otherwise ask on the first call. Copy the table into a document and fill the middle column. Keep each answer to a few lines; the evidence file carries the detail.

Field	What to write	Why the agency needs it
1. Subject	Person or organization, the name as searched, spelling variants, same-name people	Sets the entity to protect and the confusion risk
2. Audience	Who searches (investors, sponsors, customers, hiring managers, journalists) and the decision they make	Decides which results matter most
3. Problem inventory	Each item: address, page type, date found, rank, the exact statement at issue	Lets the agency diagnose rather than guess
4. History	Reports filed, requests sent, refusals, legal letters, court documents	Avoids repeating failed routes
5. Goal	The change in what people find, in plain words, with a time horizon	Defines what the proposal must deliver
6. Constraints	Confidentiality, allowed methods, approvals, jurisdictions, counsel involvement, deadlines	Sets the boundary of the work
7. Owned assets	Sites, profiles, domains, who holds access	Shows what can be built on
8. Budget and term	The envelope you can commit to and the longest term you would accept	Keeps proposals signable
9. Success measures	How you will judge progress and what reporting you expect	Makes results checkable
10. Decision process	Who decides, the date for questions, the date for proposals, the date you decide	Sets a fair, equal process

## State the goal as a change in what people find, not as a deletion A good goal names the audience, the result you want them to find, and a horizon, and it never promises that a third party's page will disappear. Removal depends on the publisher, the platform, the law and the facts, none of which an agency controls, so a goal written as a deletion sets the wrong test. Here is a scenario, invented for illustration. A founder is preparing for a funding round in about six months. A search on her name shows an outdated news story that has since been corrected by the publisher, a listing on a complaint site, and a profile page she does not control. Compare two ways of writing the goal. The weak version: "Remove the negative results about me." It cannot be measured and it invites a proposal that promises more than anyone can deliver. The strong version: "Before the funding round, a search on my name on page one should show my own site, my professional profiles and accurate coverage first. The outdated story should be corrected at source if the publisher agrees, the complaint listing should be reviewed against the site's rules, and AI assistants should describe my current role correctly." The strong version contains three jobs with three routes: correcting a publisher's article, disputing a listing against a platform's own rules, and building accurate pages that rank above what remains. The last is what search suppression means in practice: lawful push-down by better, accurate content. Our pages on removal and suppression explain how the two differ. Describe the outcome and let agencies propose the routes. ## Write the constraints that keep the work lawful and confidential Constraints belong in the brief because the agency cannot infer them, and because a written constraint is something you can hold the proposal against. Six are worth stating every time. 1. **Methods.** Say that all work must be lawful, accurate and within the rules of each platform and search engine. Ask for a statement of the methods the agency does not use. 2. **Reviews and testimonials.** Say that no fake, paid-for or insider reviews may be created, bought or sold on your behalf. 3. **Confidentiality.** State who may see the evidence file, whether subcontractors are allowed, and where files are stored. 4. **Approvals.** Say which actions need your sign-off before they happen: contacting a publisher or complainant, filing a legal request, publishing content in your name. 5. **Jurisdiction.** Name the countries and legal systems involved, since removal and correction rules differ by country. 6. **Counsel.** Say whether your lawyer will review anything that could become a legal step. The review constraint is not a formality. The US Federal Trade Commission published a final rule banning fake reviews and testimonials. Its announcement describes bans on creating or selling fake or false consumer reviews, on buying reviews that carry a required sentiment, on insider reviews without disclosure, and on using unfounded legal threats or intimidation to suppress negative reviews, as the FTC announcement sets out. It also covers businesses that spread fake reviews they knew or should have known were fake. In practice, both the firm that supplies fake reviews and the business that buys or uses them are in view. The announcement was read for this guide; the rule's full text and the law in your own country are what to check before relying on this outline. Platforms have their own rules. Google's contribution policy states plainly that "Fake engagement is not allowed and will be removed." It also bars reviews that have been paid for and content based on a conflict of interest. The page was read for this guide and can change, so read it again before you rely on a detail. Search engines also publish spam policies that cover paid links, and the practical point for a buyer is simple: a tactic that breaks a rule can leave you worse off than the original problem, and the brief should say so. ## Choose success measures before the proposals arrive Success measures are the checks you will run to decide whether the work is moving, and they must be written before the proposals so that no agency can define success after the fact. Pick measures you can verify yourself, in your own searches, not only in the agency's report. Good measures are observations of the results people actually see: - **Page one composition.** How many of the first ten results are pages you own or accurate pages you welcome, against the baseline in your evidence file. - **Position of named items.** Where each problem item ranks on the date of the brief and on each review date. - **Status of each request.** For each correction or removal request: filed, answered, refused, appealed, closed. - **AI answer accuracy.** Whether named AI assistants describe you correctly for the prompts saved in your evidence file. Our AI search reputation page covers why these answers can differ from search. - **Review integrity.** The number of genuine reviews gathered through lawful requests, never a rating target set by the agency. Ask each agency to propose its own reporting too: what is sent, how often, in what form, and whether you can see the underlying searches. Monthly is common for ongoing work, but ask the agency for its cadence and its reason. The baseline in your brief is what makes any report meaningful, because progress is only visible against a starting point. ## Ask these questions in the briefing call Send the written brief first, then hold a short call with each agency using the same questions. The point of the call is not to be persuaded; it is to hear how the agency reasons about your case. Ask for the answers in writing afterwards so they enter the proposal. **On diagnosis** - Which of my items do you think are correctable or removable, and which are not? What is your reasoning for each? - What would you want to see that is not in the evidence file? - Which route would you try first, and what happens if it is refused? **On method** - What does your team do itself, and what does it hand to others? Who exactly will work on my file? - Which methods do you not use, and why? - How do you handle a request that a platform has already refused once? **On risk** - What could make my situation worse, and how do you reduce that risk? Draw out the chance of a removal attempt drawing attention to the item, sometimes called the Streisand effect. - What do you need my approval for before you act? - What happens to my files and logins when the work ends? **On exit** - How do I stop, what do I keep, and what do I pay for if I stop? ## What a good proposal contains A good proposal answers your brief field by field and adds a plan, a method statement, a reporting scheme and terms. Hold each proposal against the list below. Missing items are not always fatal, but each is a question to put back to the agency. 1. **A restatement of your situation** that shows the agency read the evidence file and noticed what is in it. 2. **A claim-by-claim assessment**, separating what looks removable or correctable from what looks like a suppression job and what is unlikely to move at all. 3. **A route for each item**, with the reason and the fallback if it is refused. 4. **A method statement**: the practices used and the ones excluded, with an explicit statement on reviews and links. 5. **A timeline in phases**, with ranges and the word "usually" where outcomes depend on third parties. 6. **Named responsibilities**: who does the work, who you speak to, who approves. 7. **A reporting scheme** matched to your success measures and baseline. 8. **Your approvals** written into the process. 9. **Terms**: length, notice to stop, what is delivered when, and how your data is handled. 10. **A cost structure** you can map to the work, shown against the budget you stated. Item 2 is the most revealing. A proposal that treats every item the same way, with one program for everything, has not engaged with your evidence. A proposal that says "this one is unlikely to move, here is why and here is what we would do instead" has. ## Score the proposals against a fixed scorecard A weighted scorecard turns proposals into numbers you can compare, and it works only if the weights are set before you read the first proposal. Five criteria cover most of what matters, and the weights below are an example you can change to suit your case.

Criterion	Weight
Diagnosis fit	25
Method and lawfulness	25
Measures and reporting	20
Team and process	15
Terms and exit	15

Score each criterion from 1 to 5, multiply by the weight, add the results, and divide by 5 to get a score out of 100. Here is a worked example with three invented proposals. The scores are made up to show the arithmetic. Proposal A scores 5, 4, 4, 3 and 4. That is 125 plus 100 plus 80 plus 45 plus 60, a total of 410, or 82 out of 100. Proposal B scores 3, 2, 3, 5 and 3: 75 plus 50 plus 60 plus 75 plus 45, a total of 305, or 61. Proposal C scores 4, 5, 2, 4 and 5: 100 plus 125 plus 40 plus 60 plus 75, a total of 400, or 80.

Weighted scorecard, three invented proposalsScore out of 100. Weights: 25, 25, 20, 15, 15. Scores of 1 to 5 per criterion.

Proposal A82

Proposal C80

Proposal B61

Computed from the worked example in the text. Illustrative figures, not data about any real proposal.

A and C are two points apart, within the noise of a subjective 1 to 5 score. For a close result, find the criterion where they differ most and ask which weakness is easier to fix after signing. C is weakest on reporting, which a contract can usually improve. A is weakest on team and process, which is harder to change. Apply the automatic fail described above before scoring, so a proposal that promises an outcome never reaches the table. ## Read proposals for behaviors, not for brand The behaviors in a proposal tell you more than its credentials, and the table below pairs the ones that matter with the follow-up question that resolves them. None of these is proof of bad faith alone. Each is a prompt to ask.

What the proposal does	What it can signal	Ask this
Names a result and a date	Outcome sold as certain	What happens if the publisher refuses?
One program for every item	Evidence not read	Which of my items did you assess first?
Silent on methods	Methods not stated	Which methods do you exclude?
Talks about generating reviews	Possible rule breach	How are reviews gathered and disclosed?
Offers to remove accurate coverage	Ignores the lawful limit	On what ground would that be removable?
Reports only activity	Weak accountability	Can I see the searches behind the report?
Wants access to everything first	Weak data handling	Why do you need it, and who sees it?
Pressure to sign quickly	Process risk	Can the offer stand until my decision date?

## Send the brief this week, in a fixed sequence A fair selection usually takes a few weeks, and the sequence matters more than the speed. The steps below fit a single person or a small company. A larger organization with a procurement process will add stages, not remove these.

A selection sequence, usually two to four weeks

Build the evidence fileCollect searches, AI answers, the problem inventory and the history. This is usually the longest step, and it is yours alone.
Write the brief and the scorecardFill the ten fields and fix the weights before contacting anyone.
Send the brief under confidentialityShare with a short list of agencies, with a mutual confidentiality agreement for anything sensitive and one date for questions.
Hold the briefing callsUse the same questions with each agency and ask for the answers in writing.
Receive proposals by one dateGive every agency the same deadline. Answer clarifying questions to all of them, not just one.
Score, compare, follow upApply the automatic fail, score, send follow-up questions on the weakest criteria, then decide by the date you announced.

The next step is to open a document, copy the ten fields and fill the evidence file before you speak to anyone. If you want an independent view of where you stand first, a private audit gives you a baseline you can attach to the brief. Reputation Partners offers one, and it carries no obligation to hire us.

If you are about to brief an agency

Start with a baseline. A private audit maps what shows up for your name today, what is correctable and what is not, so your brief starts from facts. Request a private audit.

## Questions people ask ### How long should a reputation management brief be? One to three pages is enough. The ten fields should take a few lines each, and the evidence file carries the detail as an attachment. A brief that runs longer usually mixes in contract terms, which belong later. Short briefs get read by every agency you send them to, and a clear one signals that you will read the proposals just as carefully. ### Should I tell agencies my budget? State the envelope you can commit to and the longest term you would accept. Without it, proposals range from a narrow fix to a full program and you cannot compare them. You do not have to give a single figure; a ceiling is enough. Ask each agency to show how its cost structure maps to the work it describes. ### How many agencies should I brief? Three to five is enough for most cases. Fewer leaves you without a real comparison, and more produces proposals you cannot read properly. Brief each one the same way, on the same dates, and score them with the same weights. If fewer than three fit your situation, brief the ones that do and say so in your notes. ### Do I need a lawyer to review the brief? Not always, but have counsel look at the constraints and approvals if any step could become a legal action, such as a demand letter or a court filing. The agency can describe routes, and your lawyer decides whether a legal step is right for you. This guide is information, not legal advice. ### What if I cannot tell which items are inaccurate? Say so in the brief and list them as "unassessed". Ask each agency to assess them in its proposal. The way an agency separates accurate items from inaccurate ones tells you a lot about how it works. Where you can, check facts against primary documents such as court records, company filings or the publisher's own corrections page.