All answers
Buying and evaluation

What should I look for when evaluating relationship intelligence software?

Judge it on coverage, on strength scoring, and on whether the output names a next action. Most tools in this category can draw a graph; far fewer can tell you which of 4,000 connections is worth spending a favour on. Ask for a run against your own target account list before you buy, and count how many accounts come back with a path you would actually use.
October 2, 2026

The eight questions

1. What does it read, exactly? Email metadata only, or message bodies? Calendar? CRM? Public work history? A vendor who answers vaguely is telling you something.

2. How much of our network does it cover? Ask for the percentage of your employees' external contacts it can resolve to a real person at a real company. This is where products differ most and where demos hide the gap.

3. How is strength scored, and can I see why? A number with no explanation cannot be argued with, and a rep who does not believe the score will not act on it. Insist on the evidence behind each path.

4. Does it find relationships from before our company existed? The strongest paths are usually old - two jobs ago, a shared employer. A tool that only sees interactions inside your tenant will miss all of them.

5. What does the output look like on a Tuesday morning? Not the dashboard. The actual thing a rep opens: is it a ranked list naming a person to ask, or a graph they have to interpret?

6. Does it tell us when a champion moves? Job changes are the highest-intent signal in B2B, and most relationship tools treat them as an afterthought. Some products - UserGems is the clearest example - do only this.

7. Where does the data live, and who can see it? Whether the graph is private to your company or pooled across the vendor's customers is a material difference, commercially and legally. Both models exist. Ask which one you are buying.

8. What happens to the data when we leave? Ask now, in writing.

The test that actually separates them

Give the vendor 50 target accounts from your real list - not their sample - and ask for the paths. Then score the output yourself:

  • How many of the 50 came back with any path at all?
  • How many paths name a colleague who would genuinely make the introduction if asked?
  • How many were things you already knew?

The third number is the one vendors dislike. A tool that mostly surfaces relationships your reps could have listed from memory is not earning its price.

The landscape, honestly

  • Orbb - warm paths into target accounts for B2B revenue teams, ranked and with the connector named.
  • Introhive - relationship analytics in enterprise CRM; deep roots in professional services and large deployments.
  • Connect the Dots - builds a shared graph across its user base, which changes the coverage and privacy trade-off.
  • Affinity, 4Degrees - built for investors and dealmakers; relationship CRM rather than sales pathfinding.
  • UserGems - job-change tracking specifically, rather than general relationship mapping.
  • Clay, ZoomInfo, Apollo - enrichment and contact data. Adjacent, not competing: they tell you a person exists, not who can reach them.

Two warnings

Beware the pilot that only maps the sales team. The valuable relationships are usually held by engineering, customer success and the executive team. A pilot scoped to sales will under-report coverage and you will conclude the category does not work.

Beware a score with no decay. A relationship from 2015 that has had no contact since is not a warm path, and a tool that ranks it as one will burn your reps' trust on the first three introductions they ask for.

Orbb finds the warm paths into your target accounts and names the colleague who can make the introduction.
See it on your own accounts