Coaching 03
Profound,
Run
Properly.
Citation tracking, share of voice, and prompt-level visibility across AI search. Powerful once someone attaches a decision to it — an expensive dashboard until they do.
What Profound Does
Profound measures how your brand appears inside AI search — which prompts surface you, how often you're cited, who gets named instead, and how that moves over time across ChatGPT, Perplexity, Gemini and Google's AI Overviews.
That's genuinely hard to do by hand at any scale. Running thirty prompts across four engines once is an afternoon. Running them weekly, tracking competitor share of voice, and spotting a drift in how you're described is not something a person does reliably.
What the tool cannot do is tell you why. It reports that you dropped out of a prompt's answer. It doesn't tell you whether that's because a competitor earned a placement in the source being cited, because your schema broke, or because the model changed how it weights recency. That interpretive step is where the value is, and it's the part teams skip.
What We Set Up
Prompt Design
The single highest-leverage decision. Track the prompts your buyers actually ask, weighted toward unbranded discovery. Most default prompt sets over-index on branded terms and produce a flattering, useless picture.
Reading The Data Honestly
High branded visibility is not good news on its own — it means the model can read your website. The gap between branded and unbranded performance is the actual finding.
Competitor Share Of Voice
Who gets named when you don't, and which sources keep producing them. This is where the placement target list comes from.
Source Tracking
Which pages the engines cite for your category. Frequently competitor-authored listicles — meaning a competitor is effectively writing the answer about you.
Alerting Worth Having
Alerts on material change only. A tool that pings on every fluctuation gets muted in a fortnight, and then it may as well not exist.
The Monthly Read
A written interpretation, not a dashboard screenshot. Three things that changed, why, and what to do. Reporting formats covered in SEO coaching.
Coaching Or Done For You
Setup And Prompt Architecture
Buyer definition, prompt set design across intent tiers, competitor list, and baseline run. Done once, properly, because everything downstream inherits these choices.
Team Training On Your Account
Your data, your competitors, your prompts. How to read the visibility graphs without over-reacting to noise, and which movements are worth escalating.
Diagnosis Workflow
The step tooling doesn't cover: turning “we dropped on this prompt” into a specific, testable cause. This is the skill that makes the subscription worth paying for.
Ongoing Monitoring (Done For You)
I run the account, watch the movement, and deliver a monthly written read with a ranked action list. Useful when nobody internal has the hours yet.
Handover
The goal in coaching mode. Your team runs the cadence, I stay available for the ambiguous months. Rhythm design is covered in the published operating cadence.
Fit
Good fit
- checkYou've bought Profound and it's being checked but not acted on.
- checkYou're evaluating whether AI visibility tooling is worth the spend.
- checkYou need competitor share-of-voice data to make a case internally.
- checkAn agency wanting to offer AI visibility reporting across a client book.
Poor fit
- closeNo baseline understanding of your category's prompts — start with a GEO audit instead.
- closeExpecting the tool to explain causes. It reports what changed; diagnosis is human work.
- closeWanting daily alerting on every fluctuation. That gets muted within a fortnight.
- closeA brand nobody searches for yet. Fix demand first; measurement won't create it.
Engagement Shapes
Setup & Prompt Architecture
Fixed scope · one-time
Buyer definition, prompt set design across intent tiers, competitor configuration, baseline run. Done once, properly — everything downstream inherits these choices.
Team Training
Half day or ongoing
How to read the data without over-reacting to noise, which movements are worth escalating, and how to turn a drop into a testable cause.
Managed Monitoring
Monthly · done for you
I run the account and deliver a written monthly read with a ranked action list. Not a dashboard screenshot.
Pricing is set per engagement after a scoping call — scope, account count and cadence move it too much for a single published number to be honest. You will have a fixed figure before any work starts, and there are no open-ended hourly arrangements.
Common Questions
Do we need Profound, or can we do this manually?
You can baseline manually — a structured prompt set run by hand across four engines gives you a defensible starting point. What you can't do manually is track it weekly across hundreds of prompts. Tooling earns its place at monitoring, not at diagnosis.
We bought it and nobody uses it. Is that fixable?
It's the most common situation and yes. Unused tooling is almost never a tool problem — it's that no decision was ever attached to the dashboard. We start by defining what you'd actually do differently based on what it shows.
Can you run it for us instead?
Yes. Done-for-you covers setup, prompt design, weekly monitoring and a monthly read with recommendations. Most clients start there and move to coaching once someone internal can own it.
Are you certified?
Profound University certification is in progress. In the meantime the work stands on the audits — the methodology is published openly and you can read a full worked example before hiring me.
Measure it,
then move it.
Profound set up around the prompts your buyers actually ask — and a workflow that turns the data into decisions.
BOOK A PROFOUND SESSION