Key takeaways
- White-label is a tier, not a toggle. Most tools gate branded reports, client portals or custom domains behind a specific plan, so check the plan before you promise a client anything.
- Compare cost per active client, not cost per account. Per-domain pricing multiplies with every new client, while pooled plans with isolated projects usually get cheaper as you grow.
- Set up every client the same way: one project, about ten real buyer prompts, named competitors, a baseline run, then a reviewed report.
- Report each AI engine separately. The number of sources an engine cites differs a lot, so one blended score hides the story.
- The report is the easy part. The money is in the service layer: explaining the data and fixing what it finds.
What a white-label AI visibility report is
A white-label AI visibility report shows where a client appears in AI answers, which competitors get named instead, and which sources the AI cites. It carries your logo, colors and sender address, not the vendor's. That's the whole definition. Everything else is packaging.
The packaging matters more than you'd think, though. Clients ask "does ChatGPT recommend us?" and expect an answer they can trust and forward to their boss. If the PDF says "Powered by SomeVendor" in the footer, you've just introduced them to your supplier.
A quick sanity check on the data itself. AI engines don't behave alike. In Promptwatch Data's sources-per-response report (daily averages, published September 2025), ChatGPT cited about 5 sources per web-search answer, while Google AI Overviews and Perplexity each cited roughly 10. Microsoft Copilot swung from under 2 to nearly 17 within weeks. Practical consequence: never present a single blended number. Show ChatGPT, AI Overviews and Perplexity as separate lines, and don't treat one as a forecast for another.
What to check before you buy
The vendor pages and third-party roundups I read disagree with each other on pricing and limits in several places. Treat every number in this guide as a starting point and confirm it on the vendor's own pricing page before you quote a client.
Here is the checklist I'd run on any tool:
- Which plan unlocks white-label reports? Which unlocks a client portal or read-only seats?
- Can the dashboard run on your own domain, or only carry your logo? A logo swap still leaks the vendor in the address bar.
- What's the pricing unit: per domain, per seat, per prompt, or per credit?
- Are there pitch workspaces so you can audit a prospect before the sales call?
- Can you write your own prompts per client, or is it seeded from a generic database?
- Is there an API, Looker Studio connector or CSV export for your own reporting stack?
- Is there a trial without a card, so you can test with a real client brand?
Tool comparison for agency reporting
Prices below come from third-party listings (Semrush's blog, BrandLight, SearchEye, vendor-published comparison posts) gathered in 2026. They conflict in places. Verify before use.
| Tool | Entry price (listed) | White-label and client access | Pitch or prospect audits | Watch out for |
|---|---|---|---|---|
| Promptwatch | Agencies from $199/mo (Kick-off) | White-label workspace, client portal, white-label scheduled PDFs | Pitch projects included | Highest-value features sit on Growth and Scale |
| Peec AI | Essential $245/mo, Scale $795/mo | Agency-branded reports, read-only dashboards | Pitch projects | Credit math takes work; seat limits are reported inconsistently |
| Otterly.AI | Lite $29/mo, Premium $489/mo | Agency workspaces reportedly on Premium | Pitch environment | White-label depth varies by source |
| Semrush AI Visibility Toolkit | $99/mo per domain | White-label PDFs need a $20/mo add-on | Not a focus | Per-domain cost scales badly past a few clients |
| SE Ranking | From about $103/mo annual, Agency Pack extra | 30 client seats, scheduled white-label reports | Not a focus | AI prompts share limits with keywords |
| Nightwatch | Professional from EUR 159/mo | White-label from Professional, unlimited seats | Not listed | Rank tracking first, AI second |
| Rank Prompt | $149/mo Agency | White-label reports, client portal, 500 brands | Prospecting on Agency Plus ($299/mo) | Vendor's own claims, so test it |
| Profound | Agency Growth $99/mo plus add-ons | Client workspaces are paid add-ons | 10 pitch workspaces per month | Agency pricing not fully transparent |
| Scrunch | Agency Core about $500/mo | White-label reporting, pitch environment | Yes | Packaging has changed; plan details have conflicted |
The universal setup, before any tool
The tool changes. The method doesn't. Do this once and reuse it for every client.
Step 1: Write the prompts like a customer, not a marketer
A tracker only measures what you ask it. A local plumber's customers ask "who can fix a water heater in Mesa today," not "plumbing services Arizona." Start from sales calls, support tickets, Search Console queries and People Also Ask. Load about ten real questions per client on day one, then grow to 30-50 as the baseline settles.
Keep every client's prompt set in the same shape: a few category questions, a few comparison questions, a few branded questions, a few problem-led questions. That consistency is what makes twenty clients manageable.
Step 2: Define the competitor set with the client
Ask the client who they think they compete with, then check who the AI actually names. They're often different lists. The gap is usually the most interesting slide in the first report.
Step 3: Run a baseline and look before you publish
Never send the first run unreviewed. Read the actual answers. You'll find prompts that are too generic, competitors that don't belong and answers that mention the brand in a mildly negative way. Fix those first, because the client will find them in about thirty seconds.
Step 4: Choose the reporting layout
Whatever the tool, a client-ready report answers four questions: where do we stand, why are competitors winning, which sources shape the answers, and what happens next. Add a short note from you at the top. That note is the product.
Setting up Promptwatch for client reporting
Promptwatch is the platform 1001 SEO Media uses for its own AI visibility work, so I'll be upfront about that. Its agency plans are built around this exact workflow.
Workspace and plan
Agency plans come in three tiers (Kick-off $199/mo, Growth $399/mo, Scale $799/mo; confirm on promptwatch.com/pricing). All include unlimited projects, pitch projects and prompts, 10 seats, state and city tracking, Ads Radar, crawler logs and visitor analytics. The tiers differ mainly by monthly response volume (10,000, 25,000 and 65,000). That makes budgeting simple: you aren't repricing every new client, you're watching one response pool.
Per-client setup
- Create one project per client with the domain and country or language.
- Add prompts. Prompt Explorer suggests candidates from Search Console, ranking keywords and People Also Ask, and the tool shows monthly volumes and difficulty so you can prioritize. Personas help when a client sells to distinct buyer types.
- Add competitors and tag prompts by topic so the report can be sliced later.
- Connect Search Console and, if the client allows it, AI crawler logs through Cloudflare, Vercel, Netlify, a CDN or the WordPress plugin. Crawler logs are what let you answer "why aren't we cited?" instead of only "are we cited?"
- Install the white-label workspace and client portal, then add your branding.
What goes in the report
Use the Visibility Score, Share of Voice, position and sentiment as the headline set, and read them together. Low Share of Voice means a coverage problem. Late position means prominence. Mentions without self-citations usually point to a content or crawlability problem. Scheduled white-label PDF reports go out on your cadence, and Looker Studio, Excel and CSV exports cover clients who live in their own dashboards.
The Action Board then gives you the next step to attach to each report, and the Content Agent can publish fixes to WordPress, Webflow or Framer once the client approves them. That matters for pricing, which I'll get to below.
Promptwatch also advertises a live session, "Auditing & Scaling AI Search for Clients," on Thursday, October 22, if you want to see an agency setup run live (promptwatch.com/live-sessions).
Setting up Peec AI
Peec AI is strong on competitor benchmarking, which makes for good client narratives.
Setup is credit-driven. Plans are listed by credits (for example 10,000, 25,000 and 65,000), and credits can be allocated flexibly across client projects. Create an isolated project per client, load prompts, then decide how many credits each client gets per month. SearchEye warns that this credit math takes real effort and that execution remains with the agency, so estimate burn on one pilot client before you quote a retainer. White-label reporting is reported on higher tiers, so confirm which one before relying on it.
Setting up Otterly.AI
Otterly is the low-friction option. Qwery's agency review praises setup in minutes and a $29 Lite entry point that lets you add a small account without a budget conversation.
The catch is agency workspaces, which one source places on the $489 Premium tier. For a solo consultant with a handful of clients, Lite or Standard plus hand-branded exports can be enough. At fifteen clients that manual branding gets painful.
Setting up Semrush and SE Ranking
If you already pay for one of these, the AI add-ons are the cheapest way in.
Semrush's AI Visibility Toolkit is listed at $99/month per domain, and white-labeled PDFs need the $20/month My Reports Pro add-on. BrandLight's math puts 15 client domains at about $1,485/month, so it's lovely for three clients and expensive for thirty.
SE Ranking pools AI prompts and keywords in one project, and its Agency Pack adds 30 client seats and unlimited scheduled reports. Practitioners on Reddit keep naming it for exactly that. The trade is that AI answers sit one tab over from rankings, so lead your client report with the AI answer, not the rank position.
Other tools worth a look
Nightwatch turns on white-label reports and unlimited seats from its Professional plan. It suits agencies that want search and AI tracked together.
Profound targets enterprise brands. Agency Growth is listed at $99/month plus add-ons, with client workspaces as paid extras and 10 pitch workspaces a month. Agency pricing isn't fully transparent, so get a quote.
Scrunch lists Agency Core at about $500/month with white-label reporting and a pitch environment, though its official plan details have conflicted across sources.
Rank Prompt
Rank Prompt lists white-label reports on a $149/month Agency plan and client logins and prospecting on a $299/month tier. Those claims come from its own comparison page, so run the free first report before committing.
For broader browsing, the GEO software directory at bestgeosoftware.com lists more options.
Delivering the first report: a one-week plan
| Day | What you do |
|---|---|
| 1 | Create three client projects, load about ten prompts each, add competitors |
| 2-3 | Run the baseline, read the actual answers, remove weak prompts |
| 4 | Add branding, pick the reporting period, draft your own executive summary |
| 5 | Review each report as if you were the client's CEO |
| 6-7 | Send reports, invite clients to read-only access if the tool supports it |
Your pitch to the client is one sentence: we now monitor how ChatGPT, Google AI Overviews and other assistants describe you, here's your first report.
Pricing the service
Typical agency platform spend runs roughly $100 to $800 per month before enterprise contracts, extra domains, added credits or publisher costs, according to SearchEye. Third-party sources report agencies reselling monitoring, reporting and remediation at $500 to $1,500 per client per month.
There's a fair counterargument from BrandLight: the reporting layer is commoditized, because a client can buy the tool directly for $99. So don't sell the dashboard. Sell the interpretation and the fixes. A sensible structure is a monitoring retainer anchored against the client's existing SEO retainer, with GEO remediation (content, technical fixes, citations) as project work.
A worked example. A $399 Growth-style plan covering 15 clients costs about $27 per client. A per-domain tool at $99 each costs $99 per client. Same clients, roughly a 3.7x difference in cost of goods before anything else.
Common mistakes
- Sending a blended score across engines. Show each one.
- Reporting a screenshot instead of a rate. AI answers vary run to run, so report a rate across repeated samples with the evidence behind it.
- Choosing the tool from the demo. Test with a real client brand and open the exported file the way the client will.
- Forgetting the custom-domain question. If clients will log in, ask whether it runs at insights.youragency.com.
- Skipping the next step. A report without a recommended action is a thermometer with no doctor attached.
Which setup to pick
| Agency type | Sensible starting point |
|---|---|
| Solo consultant, up to 5 clients | Otterly Lite or LLMrefs, branded by hand |
| Growing agency, 10-30 clients | An agency plan with pooled responses and isolated projects, such as Promptwatch Growth |
| Already on Semrush or SE Ranking | Their add-ons, with the client report led by the AI answer |
| Few enterprise clients with in-house teams | Profound or Scrunch, seat-based |
Whatever you choose, run a sample report in the trial first and check three things: per-brand pricing, the branding on the exported file, and what the client portal actually looks like. If you'd rather have a team build the whole service line with you, 1001 SEO Media works with agencies on AI search and GEO.