The problem in numbers
Count three things in your account this week. How many leads currently have no owner, or an owner who left the company. How long the median lead waited for its first activity last month. And how unevenly last quarter's leads spread across reps. Most teams find orphaned leads in double digits, median waits measured in days, and one rep quietly receiving twice the flow of another.
Every study of lead response points the same direction: speed to first contact is one of the strongest levers on conversion, and it decays by the hour. Routing is the unglamorous machinery behind that speed. A lead that reaches the right person in one minute can be worked in five. A lead that sits unowned is a coin flip by tomorrow.
The design
Everything routes through the Leads inbox, from every source: web forms, chat, imports and manually created leads. One entry point, or the rules fork per source and rot separately. Two fields carry the routing logic. A Lead source field, single option, set automatically per channel. And a Territory or Segment field, single option, only if your routing actually branches on it. Resist adding more. Routing rules multiply by every field they read.
The assignment itself uses Pipedrive's automatic assignment, which distributes leads or deals round robin across a chosen group, with conditions read from fields. The working pattern: one rule per branch, a rotation of the reps who genuinely work that branch, and a named fallback owner, usually the sales manager, who catches everything the conditions miss. The fallback is not decoration. It is the difference between a routing gap you see tomorrow and one you find at quarter end.
Around the assignment, two workflow automations complete the system. One notifies the new owner the moment a lead lands, by email or a Slack message, with the source and the next expected action in the text. The other creates an instant follow-up activity on assignment, due the same day, so the lead enters the rep's task list rather than their good intentions. The general craft of keeping such automations few and maintainable is covered in automations that matter.
Implementation steps
Day one: clean the inputs. Merge duplicate sources into one Lead source option set, archive dead leads, and confirm every intake channel actually creates a lead in the inbox. Day two: build one assignment rule for your biggest branch, add the notification and the instant activity, and run ten test leads through it. Then, and only then, add the remaining branches. Teams that build all rules first and test later find the overlaps in production.
Put the rotation list under change control. New hire: added to the rotation in their second week, at half flow if your volume allows a warm-up. Departure or holiday: removed before the last working day, open leads bulk-reassigned the same morning. Write these two checklists down. Routing systems die by absence, not by design error.
How you measure that it works
Three numbers, monthly. Unowned leads: should be zero, and every exception should trace to a rule gap you then close. Time from lead creation to first activity done, from the Insights activity reports: watch the median and the worst decile, because averages hide the leads that wait days. And distribution per rep against the rotation's intent. If one rep receives materially more, a condition is leaking. These same numbers feed the response time targets in SLA design, which is this playbook's natural sequel.
Common failure modes
The ghost in the rotation: a departed rep still receiving leads. Remove departures from assignment rules as part of offboarding, the same day access is revoked. The over-branched ruleset: six conditions deep, nobody can predict where a lead lands, and fixing one branch breaks another. Collapse back to the fewest branches that match real specialisation. The silent source: a new form or campaign that never got wired into the inbox, discovered weeks later. Audit intake channels quarterly against your marketing calendar. And the fairness drift: territories change, one patch grows fat, and the rotation stops matching the market. Review distribution when territories move, not annually.
Questions
Which Pipedrive plan do we need for automatic assignment?
Pipedrive's built-in automatic assignment for leads and deals is available on the higher plans, Professional and above at the time of writing. On lower plans you can approximate routing with workflow automations and a shared filter discipline, but the round robin itself is the feature worth upgrading for.
Should leads route to people or to a shared inbox?
To people, always, with a shared fallback view for safety. Shared ownership means no ownership, and response times prove it within a week. The routing rule can be simple, but the output must be one name per lead.
How do we handle holidays and sick days?
Remove the absent rep from the assignment rotation before they leave, and reassign their open leads with a bulk edit on day one of the absence. Build the checklist into your handover routine. Automation distributes work. Only a human decision removes someone from it cleanly.
What is a reasonable routing speed target?
Assignment itself should be instant, which automation delivers. The meaningful target is first response. Inside an hour is a solid SMB standard, and same business day is the floor. Set the target you can staff, then measure it honestly.
Do we route by territory, segment or round robin?
Start with the simplest rule that matches how you sell. Pure round robin suits interchangeable reps. Territory fields route when language or geography matters. Segment routing pays when deal sizes differ tenfold. Layer rules only when a plain rotation demonstrably misroutes.