This was a practical growth model, not a research project. The goal was simple: score leads based on likely quality, write those scores back into HubSpot, and use them to lower CAC and make campaign decisions faster.
The model can be explained in four steps: collect the right signals, score lead quality, group people into clear tiers, and let marketing act on those tiers.
Pull CRM, campaign, inquiry, and behavior data into one usable marketing view.
Rank leads by likely fit, likely value, and how strongly they deserve follow-up.
Write an A-D tier back to HubSpot so every team sees the same priority level.
Scale good-fit acquisition, nurture the middle, and suppress low-value spend.
This page used to over-explain the math. The business questions were much simpler.
Find the leads and audiences most likely to turn into profitable customers, not just cheap clicks.
Identify the middle tier that needs stronger follow-up before they are ready to buy.
Spot the audiences that create activity without enough value so spend can be pulled back sooner.
The important part was not the model label. It was giving marketing one clear way to prioritize people.
Best-fit leads and customers. Higher confidence on conversion and longer-term value.
Promising but not yet ready. Better follow-up and better messaging matter here.
Shows some interest but weaker fit. Better used for lighter nurture than aggressive spend.
Low fit or low value. Do not keep funding acquisition here just because volume looks nice.
The model mattered because it changed campaign behavior. These were the kinds of decisions it supported.
Bid more aggressively and route better-fit audiences into stronger offers because the model showed they were worth it.
Use a lighter, more educational sequence instead of forcing every lead straight into the same hard-conversion path.
Stop treating low-fit leads like growth. That alone helped lower wasted spend and tighten CAC.
A simple output table like this is what made the model useful to the marketing team.
| Segment | Primary Move | Reason |
|---|---|---|
| Tier A acquisition | Scale budget | Higher predicted value justified more aggressive spend. |
| Tier B nurture | Improve follow-up | Good fit, but needed better sequencing before conversion. |
| Tier C leads | Use lighter nurture | Interest was present, but the fit and expected value were weaker. |
| Tier D traffic | Suppress or exclude | Low-value volume was inflating activity without producing enough margin. |
A clearer way to spend, route, and nurture. The win was not mathematical elegance. It was cleaner marketing decisions tied to better unit economics.