The annual network review: a December ritual for your relationships

An annual network review is ninety minutes, once a year, spent comparing what you believe about your relationships against what your calendar and message history actually show. The gap between the two is where friendships quietly end — and the review exists to close it while closing it is still cheap.

Why a yearly ritual, and why December

Most relationship maintenance is reactive: a birthday notification, a chance encounter, a crisis. Those triggers are real, but they share a flaw — they only fire for relationships that still produce events. The friendship that’s been thinning for two years generates no birthday surprise, no run-in, no crisis. It generates silence, and silence doesn’t interrupt anything.

A scheduled review is the only mechanism that catches what produces no signal. Once a year is the right frequency for the big picture: often enough that drift can’t compound for long, rare enough to stay a ritual rather than a chore. (Weekly and monthly habits exist too, but they answer “who do I contact next,” not “is my social life pointed where I want it” — the broader practice is covered in our guide to maintaining your network.)

December and January earn their reputation here. The year boundary hands you a clean data range. The holiday-card and greetings season has just forced a rough pass over who you think of warmly — if you ran a holiday card system, that list is sitting right there as input. And the people you’ll want to reconnect with are unusually receptive to “new year, long time, let’s fix that” messages.

Put it in the calendar like a dentist appointment: same week every year, ninety minutes, door closed.

Who mattered this year: evidence over memory

Start with data, not introspection, because introspection has known bugs: it over-weights the last six weeks, the loudest people, and whoever you feel guilty about. The record corrects for all three.

Three sources, fifteen minutes: your calendar (who did you choose to see more than once — choose, not “was in a meeting with”), your message threads (who did you reach for when something was hard, or wonderful, at 11pm), and your photo roll (who is standing next to you in the moments you’d call highlights). Write the names down as they accumulate. Most people land on fifteen to thirty.

Then the first honest question of the review: does this list match the people I’d have named? There are usually two kinds of surprise. Someone you’d call a close friend who appears nowhere — that’s drift, and it goes on the next section’s list. And someone who’s quietly everywhere in your year but absent from your mental “inner circle” — a relationship that grew faster than your self-image updated. Both surprises are the entire payoff of using evidence; neither shows up if you work from memory.

If you want a sharper lens for the same exercise, a contact priorities matrix — important on one axis, energy-giving on the other — turns the name list into four quadrants you can act on.

Who’s drifting that shouldn’t be

Now invert the question. Don’t ask who you saw — ask who’s missing. Name your closest people from feeling alone, then check each against the record. The painful pattern to look for: someone you’d rank in your top ten, with two contacts logged all year, both initiated by them.

That pattern is relationship drift in its purest form — no conflict, no decision, just attention flowing elsewhere while mutual fondness stays fully intact. It’s worth being precise about what this is not: it’s not a friendship problem, it’s an accounting problem. You didn’t choose distance; you just never chose closeness on any particular Tuesday.

For ties that have gone fully dormant — years, not months — the case for reconnecting is stronger than intuition suggests. Levin, Walter & Murnighan (2011) asked executives to reconnect with dormant contacts and found those conversations delivered more novel value than equivalent conversations with active contacts: the trust was largely still there, but the years apart meant both sides had genuinely new things to say. The awkwardness people fear is mostly imagined, and the message that works needs no apology — open with what reminded you of them, not with “sorry I’ve been terrible.”

Out of this section comes a short list: three to seven names with a gap between felt closeness and actual contact. Not all of them will become intentions. That selection is the final step.

Pruning without guilt

The review will also surface the opposite list: relationships you maintain out of inertia. The group chat that stopped being fun in 2023. The former colleague you grab a quarterly lunch with out of pure momentum. The “we should catch up” exchange you’ve both been performing for three years.

Pruning these is the step people skip, because it feels like a verdict on the person. It isn’t. Pruning means one thing: I’m releasing myself from the obligation to proactively invest here. The person stays in your contacts, stays greeted warmly at the reunion, stays one message away if life re-converges. What ends is the pretending — and the thin, constant guilt that pretending generates.

The arithmetic behind this is unforgiving. Dunbar’s (1992) work on network layers keeps finding the same shape: roughly five people in the closest circle, fifteen good friends, 150 meaningful ties — bounded by time, which doesn’t scale. Every inertia-lunch is funded by hours the drifting-but-loved list never received. Pruning isn’t coldness; it’s where the budget for warmth comes from.

The 90-minute review, step by step

1. Gather the evidence (15 min)

   Calendar, messages, photos for the year. Write every name that appears as a chosen, repeated presence. No judging yet — just the list.

2. Mark who mattered (15 min)

   Star the people who made the year better, steadier, or more honest. Note the surprises in both directions: present-but-unranked, ranked-but-absent.

3. List the drifters (15 min)

   Three to seven names where felt closeness exceeds actual contact. For each, note the last real exchange — you’ll want it for the reconnect message.

4. Prune the inertia (10 min)

   Name what you’re maintaining out of momentum or guilt. Release the obligation explicitly — write the sentence down. Nothing else changes.

5. Choose 3–5 intentions (20 min)

   Pick the few relationships that get deliberate attention next year, and define the repeatable behavior for each. Person plus behavior plus rhythm — nothing vaguer survives.

6. Schedule the first move (15 min)

   Before standing up: first walk in the calendar, reconnect message drafted, recurring reminder set. An intention without a scheduled first action is a wish.

Intentions, not resolutions

The output format matters more than it looks. “Be a better friend” is a resolution — an outcome with no person, no calendar slot, and no way to tell on any given Wednesday whether you’re doing it. It will be dead by February, like most of its species.

An intention has three parts: a name, a behavior, a rhythm. “Walk with Anna, first Saturday monthly.” “Call Dad on the drive home each Sunday.” “Quarterly dinner with the old team, I organize it.” Each is small enough to actually happen, scheduled enough to recur without willpower, and specific enough that you’d notice it being skipped. Three is plenty; five is the ceiling. The discipline of choosing few is what makes the chosen ones real — an intention list with twelve entries is a guilt list with a new name.

Then make the system remember, because you won’t: recurring calendar entries, or reminders in a personal CRM like Endearist that resurface each intention on its rhythm — which also makes next December’s review trivial, since the year’s contacts are already logged. The ritual’s worth compounds with repetition. The first year teaches you how much drift you’d accumulated; every year after, it catches drift while it’s still one good message from repaired.