Where you land depends on a few numbers this conversation didn't ask for. We don't guess a figure from band answers. Give us the real numbers and your honest range is free.
A higher VQ doesn't mean your business will sell — many never do, and many that do leave real value on the table. It means you're building toward a premium outcome with competitive bidders and your choice of buyer, on your timeline — so a life event or a market shift never forces a discount. A lower read isn't a verdict; it's a map of where your leverage gets built.
Predictable revenue, a team that runs without you, clean reporting — they take time to show up, and they make the business calmer to run today, not just worth more later. Start now and they're real before a transition is ever on the table.
Membership turns this one-time read into a living score: the precise, comp-backed valuation behind your range, sector benchmarking, and the playbooks to move your number — tracked year over year, for $999 / year.
The defensible number behind the honest range above — built on current lower-middle-market deal data, not a rule of thumb.
Re-take the assessment whenever you want and watch your VQ — and your value — move year over year.
How your scores measure against what strong looks like for your industry and size — so you know your number isn't just relative to yourself.
Scores across all seven dimensions and ~50 signals, with written commentary specific to your business.
See which moves lift your number most, and follow step-by-step playbooks to make them.
What buyers in your space are rewarding now, refreshed as the market moves — so your plan never goes stale.
The customer talks to an adaptive interviewer. Your scoring IP lives behind a wall they have no channel to. Five layers — they can only ever reach the top two.
Renders the chat, the chips, the one-line nuance field. Holds no rubric, no weights, no scores. Open the dev tools and there's nothing to find.
Holds session state, decides what to ask next, runs the math. The only place secrets live. Your API keys never touch the browser. On completion it emails the owner their results link and notifies the Agora advisor.
The part that feels human. It asks great questions and adapts to size + industry — but it's an intake specialist, not a judge. It doesn't even hold your 1–5 mapping. Extract its whole prompt and you get questions, not the moat.
A separate, server-to-server call that reads the finished transcript and maps each metric to a score + confidence against your rubric. The customer has no channel to it — prompt injection in the chat can't reach it.
Your Excel scorer as code: weights, band thresholds, category roll-ups, the range. No LLM — it can't be talked into anything. Never leaves your server.
Model choice: Sonnet for the interviewer (conversation quality is the product), Haiku for the cheap structured scorer, human review still gates the paid Insights report. This demo runs on Haiku — production on Sonnet reads even smoother.