Three things I'd get before anything else. A cohort table by start month for the solo accounts, not an aggregate churn number, because "lose three, add four" hides whether accounts die at month four or month fourteen. Hours by account for a normal month and a heavy month, which is what harrow is pointing at. And the actual contracts for the two teams, specifically the notice period and whether they survive a change of control, which is a question for counsel in the relevant state and not something to eyeball.
On the team concentration: annual agreements with 30-day outs are 30-day agreements. If the notice is genuinely 30 days, your visible backlog on 60% of revenue is one month long. Team spend also tracks listing count, since most of it gets budgeted per listing against gross commission. A team doing 90 sides a year that drops to 60 doesn't cancel, they cut the per-listing package and the retainer holds while the volume line collapses.
The 22 solos at $400 come to about $8,800. Work out the labor against that. If each one is five or six hours a month, the owner is selling his contractors' time at a rate that only works while he does sales for free. That segment is also where the self-service pressure lands first, because a solo agent with a template library and an AI drafting tool gets to "good enough" on social posts and listing copy without you.
The failure I'd model is the team's own marketing coordinator. Teams hire one at some point, keep the vendor for the listing presentation work, and pull the volume flyers and email in house. That's a 50% haircut on your biggest account with no cancellation notice attached to it.
