A 50-rep team can survive on default task queues. A 200-rep team cannot. Tasks pile, reps cherry-pick the easy ones, and your “high intent” leads sit untouched for 72 hours. Here’s the operating model that works.
One queue per motion, not per rep
Stop creating personal queues. Build queues by motion: Inbound MQL Triage, Demo Follow-up D+1, Closed-Lost Revival, Renewal 90, Whitespace Outreach. Reps pull from the queue assigned to their motion that day. This kills cherry-picking and lets you measure motion-level conversion.
Cap queue size at 40 items
A queue with 200 tasks tells the rep “this doesn’t matter.” Use a workflow that auto-snoozes tasks beyond position 40 to the next business day. The visible queue feels finite; rep completion rates double.
Sort by deal value AND recency, not by date alone
Default sort buries warm leads under stale ones. Use a calculated property:
priority_score = (deal_amount / 1000) + (50 - days_since_last_engagement)Sort the queue descending on priority_score. Reps work the highest-leverage task next, every time.
Use sequences to feed queues, not replace them
A common mistake: enrolling 500 contacts in a sequence and assuming the auto-emails are enough. They aren’t. Configure the sequence to drop a manual call task into the Demo Follow-up D+1 queue at step 3. The queue is where humans add value.
Enforce SLA with a heartbeat workflow
Run a workflow every hour that flags any task in Inbound MQL Triage older than 4 hours and notifies the rep’s manager. SLAs without alerts are wishes.
Report on queue throughput, not task count
The metric that matters is tasks_completed_in_queue / tasks_added_to_queue per week. Below 80% means the queue is choking. Add reps, raise the snooze threshold, or kill the motion.
What to do this week
Audit how many personal queues your org runs. Pick the top three motions and rebuild as shared queues with the priority_score sort. Set a 40-item cap and watch completion rates lift.
