| Model | What drives the price | Fits when |
|---|---|---|
| Per seat | Number of licensed users | A defined sales team is the only audience |
| Platform fee | Organisation size, flat | Everyone's inbox feeds the graph, not just sellers |
| Usage or credits | Queries, enrichments, API calls | Programmatic access, variable volume |
| Outcome-based | Pipeline or meetings produced | The vendor is willing to be measured on results |
Relationship intelligence gets better as more people connect their inbox, because the relationship that opens an account is usually held by someone who is not a salesperson - an engineer, a founder, a CSM.
Per-seat pricing creates a direct incentive to connect fewer people, which degrades the product you just bought. If you are quoted per seat, ask what a read-only or contributor seat costs, and whether non-sales staff can feed the graph without a full licence. The answer tells you whether the vendor understands their own product.
A price is meaningless without these:
By agent, because each one does different work and scales differently:
Figures are quoted on a call, because every one of those is scoped to the specific work. That is a deliberate position rather than a reluctance to publish: a seat count does not describe what the Referral Agent is being asked to do, and a flat fee does not describe an events programme.
Two vendors quoting "per user per month" may be counting different users - one counts sellers, the other counts everyone whose inbox is connected. Normalise to total annual cost for the coverage you actually need, then compare that against the coverage test each one produced on your own account list.
The cheapest tool that returns no usable path into your target accounts costs more than the expensive one that does.