How teams split website leads

Teams split website leads four ways: rotation, race, rules, or some hybrid of the three. Each model works, each fails in a characteristic direction, and the difference between a routing system agents trust and one they quietly route around is rarely the model — it’s whether assignment is tracked, visible, and enforced by software rather than memory. We build lead routing for a living, so here’s the honest map.

Round robin: fairness first

Rotation assigns each new lead to the next agent in line. Its virtue is that nobody can argue with it: over a quarter, everyone gets the same flow, and the team lead never stands accused of feeding favourites. Its failure mode is indifference to fit and effort — the rotation hands a French-speaking buyer to an agent who can’t serve them, and the teammate who never calls back draws the same share as the closer. Rotation without accountability subsidizes the weakest follow-up on the team. It works when paired with capacity caps and a reassignment rule, which is where most teams that start here end up.

First-to-claim: speed first

The race model pushes each new lead to every eligible agent at once — push notification, claim button, fastest finger wins. Its virtue is built-in speed to lead: the system rewards exactly the behaviour that converts internet leads. Its failure modes are hoarding — the agent who claims everything and works half — and quiet hours, when nobody’s finger is on the phone. The fixes are mechanical: claim timers that return an unworked lead to the pool, an escalation ladder when no one claims within minutes, and a shared pond so the unclaimed lead lands somewhere visible instead of dying in a notification tray.

Criteria routing: fit first

Rules-based routing assigns by attributes before fairness ever enters: geography and postal code, price band, language, source, buyer versus seller. It produces the best consumer experience — the condo lead reaches the condo specialist — and it’s how larger teams stop routing from feeling random. The failure mode is complexity debt: rule stacks accrete until nobody remembers why the west-side seller leads skip two agents. Keep rules few and legible, and let the routing-rules playbook handle the recipes; the goal is a system any team member can predict from the outside.

Hybrids, and the money side of the split

Mature teams almost always land on a layered model: criteria first to find the eligible pool, then rotation or claim within it, then a reassignment rule behind everything — no contact in a set window and the lead moves on. Two supporting decisions matter as much as the model. Write down what fair distribution means before anyone’s pipeline depends on the answer. And settle the commission side explicitly: house-generated leads normally carry a larger team share than self-generated deals, and the team split calculator will show you what any proposed split actually pays an agent per closing — run the numbers before you publish the policy, not after the first dispute.

Whatever you choose, make the system keep the score

The model matters less than the bookkeeping. Assignment should stamp who received each lead and when; response time should be measured, not recalled; claims, timers, and reassignments should execute themselves. That’s the shape lead routing takes in BrokerFold’s CRM — round robin and claim mechanics with push notifications, so a website lead is routed and buzzing in someone’s pocket in seconds, and the record of what happened next writes itself. Teams don’t fall out over routing models. They fall out over untracked exceptions — and tracked routing is what makes both the leads and the splits defensible.