> ## Documentation Index
> Fetch the complete documentation index at: https://docs.guhan.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# How Guhan scores prospects
> Every landed prospect gets a fit chip — Strong, Moderate, or Weak — from Guhan's research agent. Here's what it means and how the two layers of filtering work.
**Every landed prospect gets a fit chip: Strong fit, Moderate fit, or Weak fit. The chip comes from Guhan's research agent — which reads each prospect's profile and rates them against what you sell.** Prospects who don't clear your hard-required filters (title, location) never land in the first place.
Two layers do the work:
**Title and location** are required by default. A prospect who misses either is rejected outright — they never land in your List.
Every prospect who clears the hard gates gets a Strong / Moderate / Weak chip from Guhan's research agent, which reads their profile and rates them against what you sell.
## The fit chip
Guhan's research agent read this prospect's profile, their company, and the triggering signal (if any) — and decided they tightly match at least one of your products or your workspace-level positioning. Right role, right scale, clear signal or pain alignment. **Reach out today.**
General audience fit but weaker on one dimension — no signal evidence, wrong scale, or partial role match. **Worth reaching out if you have pipeline capacity.**
Audience alignment is thin — the seller's offering doesn't obviously solve a problem this prospect faces. **Deprioritize.** Guhan still lets them land so you can see the assessment, but the messaging agent will de-prioritize them in the queue.
The chip shows on every prospect table + the Prospect detail page. Sort a List by fit tier to work your Strong prospects first.
### The per-product breakdown
If you have multiple products in Brand & Voice, the Prospect detail page shows a per-product fit breakdown — each product gets its own Strong / Moderate / Weak chip plus a one-sentence reason.
Example, if you sell an audit tool AND a policy-monitoring tool:
* **AuditPath** — Strong. Multi-branch expansion + RBI regulation makes the audit-trail consolidation angle a natural fit; the caveat is that they may already have a legacy tool in place.
* **PolicyStream** — Moderate. Related pain but a secondary priority — they'll evaluate this after AuditPath is live.
The message agent uses this at send time — a prospect enrolled in an campaign for Product A gets messaging anchored on Product A's fit reason.
## The hard gates
By default, **title** and **location** are required. A prospect who fails either is rejected — they never land in your List, and you never spend a credit on them.
You can change which fields are required on the source config panel — look for the **Required / Nice-to-have** chip next to each rule. Common patterns:
* **Everything required** — tightest filter. Best when your audience is very narrow and mis-fires are expensive.
* **Title + location required** (default) — good balance. Industry and headcount contribute to the fit chip but don't block landing.
* **Title only** — widest filter. Best when you want to see anyone in a role and let Guhan's fit chip do the sorting.
## Why prospects get filtered out
The List detail page has a **Filtered** accordion showing every prospect that didn't make it, with a plain-English reason:
* **Wrong job title** — their title didn't match any of yours
* **Wrong location** — they're in a country you didn't target
* **Company / Page (not a person)** — LinkedIn returned a Page or Company row instead of a real profile (auto-handled)
* **Competitor employee** — they work at one of your competitors (auto-block from your Brand & Voice competitors list)
* **Do Not Contact** — they're on your DNC list
* **Already in your workspace** — you found them via a different source already
This is useful for debugging. If Guhan keeps rejecting people you think should qualify, the Filtered accordion tells you which rule is being strict.
## What the research agent reads
To produce the fit chip, Guhan's research agent reads:
* The prospect's **LinkedIn profile** — role, tenure, current company, headline, past positions
* The prospect's **company** — industry, headcount, HQ location, description, and any signal evidence (funding, hiring, exec change) attached to it
* The **triggering signal** (if any) — matched keyword, post text, engagement type
* Your **Brand & Voice** — what you sell, who you sell to, customer pain points, competitor context
* Every **Product** you've added — with per-product ICP and value props
It produces a rich research brief (10-14 sentences of actual account research) plus the fit chip. Both surface on the Prospect detail page.
## Regenerating the assessment
Every prospect gets a fresh assessment on landing (fire-and-forget). If a prospect's context has changed — new signal, new role, new company — hit **Regenerate brief** on the Prospect detail page. Costs 1 credit against your monthly budget, then falls back to your top-up wallet if the monthly is exhausted.
## Related
The full flow from search to landed — every stage a prospect passes through.
What to check after your first landing.