Response time gets reported as a marketing SLA and optimised as an average, which hides exactly the cases that cost money. Measure the tail, fix the routing, and treat the first reply as part of the sales process rather than an acknowledgement of it.
Lead response time is usually owned by marketing, reported as an average, and reviewed monthly. All three of those choices hide the thing that matters.
The buyer who filled in the form was, at that moment, more interested in solving their problem than they will be at any later point. Everything after is decay. Response time is not an operational courtesy — it is the first and cheapest lever in the conversion chain.
The average is the wrong statistic
Most inbound is handled quickly, which drags the mean down and makes the metric look healthy. The revenue loss is concentrated in the tail: the leads that waited eighteen hours, or two days, or until Monday.
Report percentiles instead — median, ninetieth, ninety-ninth — and segment by arrival hour, day of week, source and territory. The pattern that emerges is almost never 'the team is slow'. It is 'this segment, arriving at this time, has no owner'.
Slow responses are usually a routing failure
The common causes, in roughly the order they appear:
- Round-robin assignment that ignores availability, so a lead lands with someone on holiday.
- Territory rules with a gap — a segment that matches no rule and falls into a queue nobody watches.
- Enrichment that runs before routing and blocks it when the provider is slow or the record is unusual.
- Duplicate detection that quietly attaches a hot inbound lead to a dormant existing account.
- Handoff between an SDR queue and an owner with no timer on the transition.
None of these are effort problems, which is why exhortation does not fix them and instrumentation does. Each is visible in the data within a day of looking for it.
Nobody is slow on purpose. They are slow because the lead was somewhere they were not looking.
What the routing logs usually show
Measure to first useful reply
An auto-acknowledgement is not a response, and counting it as one is how a team reports a two-minute response time while buyers wait a day. The clock should stop at the first message that engages with what the person asked — a human reply, or an automated one specific enough to be worth reading.
That definition is harder to game and much closer to what the buyer experiences, which is the only frame in which the metric predicts anything.
Fast and generic beats slow and personalised
Teams often delay the first reply to research the account. The trade is almost always bad: the value of personalisation is smaller than the value of arriving while the buyer still has the tab open.
The resolution is not to choose. Most of the research a rep would do — company context, likely use case, similar customers, what the form said — is retrievable automatically in the seconds after the form is submitted. A drafted, specific first reply waiting for one-click approval collapses the trade-off entirely, and it is one of the few automations where speed and quality are not in tension.
Why this belongs to revenue operations
Owned by marketing, response time is a service level. Owned by revenue operations, it is a conversion input with a measurable relationship to pipeline created — and it becomes possible to say what a fix is worth before doing it, and what it delivered afterwards.
That reframing is what gets routing gaps fixed. A missed SLA is a process complaint. A quantified loss in pipeline creation is a priority.