Lead Routing and First-Response SLA Automation

Most first-response SLA figures we are shown are wrong by roughly a factor of two. The clock starts when the record was created and stops when a rep marked a task complete, which measures CRM housekeeping rather than the customer's experience of waiting. Measured properly, from form submission or inbound email to the first message that actually left your domain, median response time in the teams we audit lands between four and nine working hours against a stated target of one. What the clock has to know: Working hours per person, not per company, because a rule that hands a Friday evening enquiry to someone on parental leave produces a clean dashboard and no reply. Czech public holidays, and the August shutdown weeks manufacturers actually observe, which no CRM's default calendar contains. Out-of-office state read from the calendar rather than from a checkbox somebody forgot to tick. And a written definition of first response agreed before anything is built: an automated acknowledgement is not one, and if you decide it counts, we will build it that way and label the metric so nobody is confused later. Routing that reflects how the team is actually shaped: Round-robin distributes work evenly, which is only correct when reps are interchangeable, and they usually are not. We route on the dimension that matters in your business: region by postal prefix rather than by country, since CZ and SK territories rarely split on the border; product line where the technical knowledge is not transferable; existing account ownership where a relationship outranks the queue. Capacity limits stop the fastest responder silently absorbing everything until they burn out. Every rule states its fallback, and the fallback is a named person, never a queue nobody owns. Escalation that moves work: Untouched past the first threshold, the record notifies its owner. Untouched past the second, it moves, to a specific person, with the original owner informed, and the reassignment written on the record with the reason. Escalation that only notifies is decoration. A lead sitting for six hours needs a different pair of hands, not a second alert to the same one. What you receive: Routing and SLA workflows in HubSpot, Salesforce or n8n, depending on where your logic already lives. The working-hours and holiday calendar model, including per-person exceptions. A response-time report built on send events rather than activity logging, broken down by source, owner and hour of arrival, which is usually the report that ends the internal argument about where the time goes. And a rules document your ops lead can change without calling us. Not included: We do not implement dialers or speed-to-lead call automation. We do not build routing on a lead score we have not validated ourselves, because routing on an unexamined score only distributes a bad decision faster. And we do not write your response templates. Who this is not for: Teams whose sales headcount is moving by more than a couple of people a quarter. Routing built on a shape that will not exist in three months is waste, and the honest sequence is to settle the team first and call us after. And anyone who wants the SLA report to look better rather than be right; the first true measurement is usually uncomfortable, and that discomfort is what you are paying for. The retainer covers what changes: people joining and leaving, territories being redrawn, capacity rebalanced when the report shows one person carrying a third of the volume.