Client Onboarding Checklist for Reputation Management

Why onboarding gaps cause problems later, not just at the start

Skipping an onboarding step rarely blows up immediately. It shows up three weeks later as a reply written in a tone the client hates, a platform nobody realized needed monitoring, or a request going out to a customer list that was never supposed to get an SMS. By the time the gap surfaces, you've already spent a month building a relationship on an incomplete foundation, and fixing it mid-stream costs more trust than getting it right on day one would have.

The fix isn't more effort — it's a checklist you run the same way for every client, so nothing depends on whoever happens to be running the kickoff call that week. Use the sections below as that checklist.

Platform access

Review platforms vary in how access works, and getting this wrong is the single most common source of a slow first week.

• Confirm who currently has owner/admin access to the Google Business Profile — this is often the client, sometimes a previous marketing vendor, occasionally no one anyone can locate
• Request manager-level access to the Google Business Profile under your agency's account, rather than asking for the owner's personal login
• Confirm Facebook Page admin access, and identify who on the client's side needs to approve the access request if it isn't the point of contact
• Check for any additional platforms relevant to the client's industry (Yelp, Tripadvisor, App Store/Google Play for app-based businesses, industry directories) and confirm whether claimed profiles already exist
• If a previous agency or employee still holds admin access anywhere, get written confirmation from the client that access should be transferred or revoked

Review links for every platform

You need the direct review link for each platform the client is on, not just the general business page — this is what goes into every request you send.

• Pull the direct "write a review" link for Google Business Profile (not the general listing URL)
• Pull direct review links for Facebook, Yelp, and any other active platform
• Test every link from a fresh browser session to confirm it opens the review form directly, not just the business page
• Store all links in one place accessible to anyone on your team working the account

Brand assets for client-branded touchpoints

If review requests, reports, or a client-facing dashboard are going to carry the client's branding, collect the assets before the first request goes out — not after a client notices a generic logo on their first report.

• Logo files (vector or high-resolution PNG, both light and dark background versions if available)
• Brand color codes (primary and secondary hex values)
• Approved business name variations (is it "Smith & Sons Plumbing" or "Smith & Sons" in customer-facing copy?)
• Any required legal name or disclaimer language that must appear on customer-facing messages
• Preferred sender name/email or phone number for SMS and email requests, if the client wants their own

A short reply tone guide, specific to this client

AI-drafted replies still need a human-approved tone baseline for each client — a plumbing company and a pediatric dental office should not sound the same answering a 3-star review. Build this once per client, not once per reply.

• Formal or casual? Get a specific answer, not "whatever sounds professional" — ask for an example of how the client would text a regular customer
• Identify words or phrases the client never wants used (common ones: admitting fault in specific legal terms, competitor names, anything that sounds scripted)
• Flag industry-specific sensitivities before the first reply goes out — healthcare and legal clients in particular need replies that never confirm a patient/client relationship or reference specifics of someone's visit, even when the reviewer mentioned it themselves
• Confirm how the client wants negative reviews handled in public replies: acknowledge and move the conversation offline, or address specifics publicly
• Get two or three example replies the client has written themselves, if any exist, to calibrate voice

Approval rules for AI-drafted replies

Decide the approval threshold with the client up front, rather than guessing and finding out in month two that they wanted more oversight than you gave them.

• Confirm the star rating above which AI-drafted replies can auto-post without manual review (many clients land on 4-5 stars once they trust the tone)
• Confirm that every review below that threshold gets manual review before posting, with no exceptions
• Identify who on the client's side, if anyone, wants visibility into drafts before they post, even for auto-approved ranges
• Agree on a response time target for manually reviewed replies so they don't sit waiting on approval for days

Request channels

Confirm which channels the client wants enabled and flag consent practices before the first request goes out — don't go deep into the legal weeds, just make sure the client's own opt-in process matches the channel.

• Confirm which channels are enabled: email, SMS, QR code (in-store/on invoices), or direct link
• For SMS specifically, confirm the client's customers have actually opted in to text messages through the client's own intake process — this is the client's compliance responsibility, but it's worth a direct question before requests start going out
• For email, confirm the client isn't adding review requests to a list that was collected for something unrelated without disclosure
• Confirm requests go out to every customer the same way, regardless of how the job or visit went — private feedback is available as an additional channel for a customer to flag an issue, but it never replaces or filters the public request

Staff training on the client's side

A short walkthrough for the right people on the client's team prevents a lot of the "wait, how does this work?" messages you'd otherwise field weeks later.

• Identify who needs to know how review requests get triggered (often front-desk or field staff who complete the job)
• Walk whoever triggers requests through the actual process — is it automatic after a CRM status change, or does someone tap a button?
• Identify who should see negative reviews immediately rather than waiting for the monthly report
• Confirm who approves manually-reviewed replies and make sure that person knows what their inbox will look like

Baseline report

Capture where things stood before you touched anything. This is the reference point every future report gets compared against, and it's impossible to recreate accurately later.

• Snapshot current rating and review count on every platform, dated
• Document current response rate and whether any reviews have ever been responded to
• Note current review request process, if any existed before you started
• Record any visibility baseline available (local pack position for key terms, if already tracked)
• Share this baseline with the client directly — it's the "before" half of the ROI story you'll be telling every quarter

That baseline matters for more than a tidy kickoff doc — it's the number you'll point back to when the client asks what they've gotten for their money. Our guide to proving reputation management ROI to clients builds directly on this same baseline idea.

Make it a system, not a one-off

The value of a checklist like this compounds. The first time you run it, it's just a list. By the tenth client, it's the reason your onboarding takes three days instead of three weeks, and the reason every account starts with the same baseline quality regardless of who on your team handled the kickoff. Keep this list in whatever project tool your agency already uses, assign owners to each section, and treat any skipped item as a flag to revisit before the account goes live — not something to circle back to eventually.

If platform access, request channels, and reporting currently live across five different tools and a shared spreadsheet, onboarding will always feel harder than it needs to. Tailwind for agencies gives every new client one connected setup — monitoring, requests, AI-drafted replies with your approval rules, and a baseline report — so onboarding is a checklist, not a project.