The short answer: Relationship strength is not one number. It is a blend of recency, frequency, two-way contact and channel range. Score it in a way a seller can explain, confirm it with the introducer before asking, and pair it with timing before you act.
The most common mistake in warm introduction work is treating every connection as equally warm. A LinkedIn connection from six years ago is not the same as a colleague who emailed the buyer last week. Asking someone to introduce you through a relationship that has gone quiet is a fast way to spend goodwill and get politely ignored.
Relationship strength scoring is the layer that fixes this. It turns the fuzzy question "how well does anyone here actually know this person" into something a seller can act on. Done well, it stops cold asks dressed up as warm ones. Done badly, it creates false confidence. This guide covers what to measure, how to tier it, where scoring goes wrong, and how to use a score without breaking trust.
Why raw connection counts mislead
Without some measure of strength, a team's network is just a long list of names. A sales team of a few dozen people can easily have tens of thousands of connections between them, and only a small share of those are relationships anyone could lean on this quarter.
The question that matters is not "does anyone here know this person." It is "does anyone here know this person well enough to help, right now." Those are very different questions, and confusing them is behind most bad introduction requests. If you want the mechanics of the ask itself, our guide to warm introductions covers it step by step.
The four signals that matter
Strength is a blend of several partial signals. Each one tells part of the story, and none is enough alone.
- Recency: when the two people last interacted. A relationship that was close two years ago and silent since is not close now.
- Frequency: how often they interact. One meeting a year is a thin signal. A steady back and forth over a project is a much stronger one.
- Two-way contact: whether both people start conversations and reply. A newsletter that one person sends and the other never answers is not a relationship.
- Channel range: whether contact spans email, meetings and LinkedIn. Someone you email and also meet with is easier to approach than someone you only email.

Most teams give recency and two-way contact the most weight, because they are the hardest to fake. The exact weights matter less than one property: the score has to be explainable. A seller should be able to see why a relationship landed where it did.
Turning signals into tiers
Numbers between 1 and 100 invite false precision. Most teams get more value from a handful of plain-language tiers. The tiers below are illustrative. They are a starting point to adapt, not a benchmark.

The line that matters most is between warm and room temperature. That is where most bad asks happen. A seller sees that a colleague is connected to the buyer on LinkedIn and reads it as warm, when the two have not spoken in years.
Where the signals come from
A useful strength score is built from activity a team already generates in the course of normal work:
- Email metadata: who emailed whom, when, and whether they replied. Not what they wrote.
- Meeting history: who met together, how often, and how recently.
- CRM activity: logged calls, notes and stage changes tied to a contact.
- Public professional context: shared employers, job changes and LinkedIn connections. These help explain a relationship more than they measure its strength.
Notice what is missing: the content of anyone's email. Who wrote to whom, and when, carries most of the information a strength score needs. Leaving message bodies out makes the privacy conversation with sellers and legal much simpler.
The failure modes to design around
Overconfidence in a single number. A score that looks precise gets treated as fact. Always show the evidence behind it. "Emailed back and forth several times this month, last met in the spring" is something a seller can judge. A bare number is not.
Mistaking volume for strength. A burst of transactional email, such as invoices or scheduling, can look like a strong relationship. Good scoring looks at the shape of contact, not just the count.
One-way traffic. Newsletters, ignored invitations and your own team's cold outreach can all register as activity. Requiring two-way contact filters most of this out.
Stale data. A score that refreshes rarely drifts out of date as people change roles and go quiet. Strength should decay when contact stops.
Privacy leakage. Pooling personal contacts into a shared view is a fast way to lose sellers' trust. Keep personal relationships private, and let people see that a path exists without exposing everything behind it.
How to use a score without breaking trust
The strongest teams treat a strength score as a suggestion, not an instruction:
- The score tells the seller which paths to explore first.
- The seller checks with the possible introducer before making the ask.
- The introducer confirms the relationship is what the score suggests.
- The request goes out with enough context that the introducer can describe it honestly.
This does not have to be slow. The loop can often close in a day. What it prevents is the request that lands on a colleague who barely remembers the buyer and now has to decide whether to fake familiarity.
Where strength meets timing
A strong relationship at the wrong moment is often worth less than a room-temperature one at exactly the right moment. Strength is one dimension. Timing is another.
Teams that get this right combine the two:
- Strong relationship plus a fresh timing signal, such as a new role: act now.
- Strong relationship and no timing signal: a light, useful check-in, not a pitch.
- Room-temperature relationship plus a strong timing signal: worth a careful reactivation.
This matters more as buying groups grow. Forrester's research finds that on average 13 internal stakeholders and nine external participants influence business buying decisions. With that many people involved, you will rarely have a strong tie to all of them, so knowing which ties are real tells you where to start. Our guide to combining relationship signals with timing signals shows one way to weigh both.
What to ask any tool
If you are evaluating software that scores relationships, ask:
- Explainability: can a seller see the evidence behind each score?
- Freshness: how quickly does the score change after new contact, or after a long silence?
- Two-way handling: does it discount one-way traffic?
- Channel range: does it account for contact across email, meetings and LinkedIn?
- Privacy: are message bodies left out, and do personal contacts stay private?
Vendors that cannot answer clearly deserve some skepticism. For the broader category question, see our piece on relationship intelligence versus sales intelligence.
A hypothetical example
The following is a hypothetical scenario for illustration. It does not describe a real customer or company.
A seller at a streaming platform wants a meeting with the head of video at a national snack brand. Three colleagues are connected to that buyer. On the surface, all three look like warm paths. Looking at the signals tells a different story. The first colleague connected on LinkedIn years ago and has never exchanged an email with the buyer: room temperature at best. The second emailed the buyer often last year during a campaign, but nothing since: warm, fading. The third worked with the buyer's manager until recently and still trades notes every few weeks: strong, though one step removed.
The seller asks the third colleague first, with a two-line note on why the timing makes sense. The colleague confirms the relationship is current and makes the introduction to the manager, who brings in the head of video. The first colleague is never asked, which saves an awkward request.
Checklist
- Define strength using recency, frequency, two-way contact and channel range.
- Translate scores into a few plain-language tiers and label them illustrative.
- Show the evidence behind every score, not only the number.
- Discount one-way traffic and transactional bursts.
- Let strength decay when contact stops.
- Confirm with the introducer before making any ask.
- Pair every strong relationship with a timing check before you act.
Common mistakes
- Treating a LinkedIn connection as a warm path. A connection is a fact. Warmth needs evidence.
- Trusting a bare number. If you cannot explain the score, do not act on it.
- Skipping the introducer check. The score can be wrong about a relationship. The introducer usually is not.
- Asking through room-temperature ties. Mention the shared name instead.
- Ignoring timing. A strong relationship does not mean today is the right day.
From one seller to another
Strength scoring is not about turning people into numbers. It is about knowing which relationships are real before you ask a colleague to vouch for you. Get that right and warm introductions start to feel less like favors and more like a habit.
If you want a shared system for fit, timing and access, the free Media Territory System course is a good place to start. Elo works inside Gmail and LinkedIn and shows warm paths, meaning who on your team knows someone and how recently, using who emailed whom and when rather than email bodies. It adds dated signals that flag what changed at an account. On the Team plan, managers can see coverage and signals across the book.
Sources: Forrester, The State Of Business Buying, 2026, Jan 21, 2026 (checked Nov 7, 2026).



