How LinkedIn Relationship Half-Life Scoring Works (And How to Use It)

Every professional relationship on LinkedIn has a hidden clock. The moment you last interacted with a connection, the relationship begins losing warmth — not linearly, but exponentially. Thirty days after a conversation, the relationship is still mostly intact. Ninety days later, it has measurably cooled. Two years later, the connection exists in name only.

The half-life model makes that decay explicit and, crucially, predictable. It's the same mathematics used in physics — a quantity that halves on a fixed schedule — applied to professional relationships. Here's how the scoring works, what the three zones mean, and how to act before relationships go cold instead of after.

The Formula

Each connection is scored from 0 to 100 based on days since your last interaction:

Score = 100 × 0.5^(days since last interaction / 90)

Ninety days is the half-life: the period after which a relationship retains half of its warmth. That produces a decay curve with familiar milestones:

• 0 days — score 100. The relationship is at full warmth.
• 90 days — score 50. Half the warmth is gone.
• 180 days — score 25. A quarter remains.
• 270 days — score 12.5. Effectively cold.

The exponential shape is the important part. The relationship loses ground fast early on and then asymptotically approaches zero. Waiting until a connection "feels" cold means letting the most recoverable part of the decay — the first 50 points — slip away.

The Three Zones

The raw score maps to three zones that tell you what to do:

• Green (50–100). The relationship is warm. You've interacted within roughly the last three months. No action required — the goal is to keep it here through periodic touchpoints.
• Yellow (25–50). The relationship is cooling. You last interacted roughly 3–6 months ago. This is the action zone — the connection is still cheap to revive with a single thoughtful message, but that window is closing.
• Red (0–25). The relationship has gone cold. Reviving it now costs substantially more effort — and the longer it stays red, the more the revival message needs to explain itself.

The zone boundaries aren't arbitrary. They reflect where the marginal cost of revival changes. Pulling a Yellow connection back to Green costs one message. Pulling a Red connection back costs real effort — and some Red connections, frankly, aren't worth the effort. That's fine. Triage is the point.

What Counts as an Interaction

Any meaningful contact resets the clock: a meeting, a call, a substantive message exchange, a referral you actually made good on. The model works best when the "last interaction" field holds real data rather than the date you connected. A connection you added in 2022 and never spoke to isn't at day 0 — it's been decaying the whole time.

When no interaction has been logged, the model falls back to the connection date as a conservative starting point. That means an imported network gets an honest snapshot on day one: cold connections score cold, warm ones score warm, and the yellow zone shows you exactly where the easy wins are.

How to Use the Score (and Not Abuse It)

The score is a triage tool, not a to-do list:

• Weekly, act on Yellow. A ten-minute pass over your yellow zone each week — reaching out to one or two connections before they slip to red — is the entire game.
• Respect the trigger. A revival message that references something current and specific about the person (a role change, a company milestone) is dramatically more effective than a generic one. Scoring tells you who to contact; context research tells you what to say.
• Track the balance too. Recency is half the picture. Relationships also carry a reciprocity balance — favors given versus received — that tells you who you can reasonably ask for something and whose trust you still need to earn back.

Where the Scoring Runs

The half-life score is recalculated daily from newest known interaction data — an explicit log, otherwise the last interaction date, otherwise the connection date. Zone changes are tracked over time, so you can see relationships moving in both directions: the ones you're actively maintaining (staying Green) and the ones quietly slipping (Green to Yellow, Yellow to Red) before a revival becomes genuinely expensive.

When a connection does drop zones, SocialCraft AI's reconnection workflow takes over the research half: it checks the connection's company for recent news and drafts a personalized message referencing a real event. The human 10-minute weekly pass stays yours.

This formula is a SocialCraft product heuristic, not a validated law of relationship psychology. The score measures recorded recency, not affection, trust, importance, or consent to contact. Use it to decide what deserves review, then make the outreach decision from context.

Sources and further reading

• Google Search Central: Creating helpful, reliable, people-first content

Related SocialCraft guides

• LinkedIn relationship management: context over scores
• How to reconnect with old LinkedIn connections

See which of your connections are half-decayed — score your network in SocialCraft AI →