How to Set Up a Speed-to-Lead SLA That Sales Actually Follows

A speed-to-lead SLA is a written rule with five clauses: when the clock starts, a first-response target for each lead tier, an automatic escalation when the target is missed, a coverage roster with a legal calling window, and one compliance number reported weekly. Set Tier 1 at five minutes, seven days a week.

At a glance: the five-clause speed-to-lead SLA

• Clause 1 — Clock start. One named event: the form’s submission timestamp, not the CRM sync.
• Clause 2 — Tiers. Four targets by lead source, not one number for everything. Tier 1: five minutes.
• Clause 3 — Escalation. What reassigns automatically at the breach, and who gets named.
• Clause 4 — Coverage. The roster, the after-hours channel rule, and the calling window: US telemarketing calls to residential subscribers run 8 a.m.–9 p.m. in the recipient’s local time.
• Clause 5 — Enforcement. Response-window compliance, weekly, per rep, read by one named person.
• Day one: export 90 days of lead-created and first-attempt timestamps and calculate your median. You cannot set a target without knowing where you start.
• Finish state: the weekly breach report builds itself, and Tier 1 median time to first attempt has sat inside target for four straight weeks.

What actually goes into a speed-to-lead SLA?

Most documents calling themselves a speed-to-lead SLA are one sentence long — “we respond to all leads within five minutes” — which is why they get ignored. A target and nothing else: no clock-start definition, no consequence for missing it, no rule for the hours nobody is rostered.

Write it in five clauses, in this order, because each closes a hole the last one opens: clock start, tiers and targets, escalation, coverage, enforcement. Skip one and the document fails at exactly that seam.

An SLA without clause 3 is an aspiration. An SLA without clause 5 is a memo.

When does the clock start? The clause that decides everything else

Write one event in verbatim: the clock starts at the timestamp written by the form at submission, stored in UTC, in a field the rep cannot edit. Everything else is negotiable; this is not.

The common failure is starting the clock when the record appears in the CRM. A batch sync, a retrying webhook or an hourly aggregator feed can put ten minutes to several hours between the customer pressing submit and the record existing — and in that gap your dashboard shows perfect compliance while the buyer talks to a competitor.

Define two more terms here. First attempt is a dial placed or message sent, logged automatically; first response is a two-way conversation. The target sits on first attempt. Conflating them is how a team hits its SLA by auto-dialing and hanging up — and keeping those timestamps trustworthy is a CRM data hygiene job.

What response target should each lead tier get?

A single five-minute rule for everything loses the room: reps know a whitepaper download is not a callback request. Four tiers is enough. The targets below are judgment calls from running inbound desks, not published findings.

Five minutes on Tier 1 is our target, not a research finding: the published work in this field sets its threshold at an hour, and it is the most miscited material on the web. The March 2011 Harvard Business Review article “The Short Life of Online Sales Leads” (Oldroyd, McElheran and Elkington) audited 2,241 US companies, timing responses to a web-generated test lead: 37% responded within an hour, 23% never responded at all, and the average response time among companies that responded within 30 days was 42 hours — a denominator that by construction excludes the 23% who never replied. That audit is not the source of the “nearly seven times as likely to qualify” multiple, which comes from a separate dataset in the same article: 1.25 million leads across 42 US companies (29 B2C, 13 B2B). That second study compared firms contacting a prospect within an hour of the query against firms that waited even an hour longer, and HBR defined qualifying as “having a meaningful conversation with a key decision maker” — not a sale. Two samples routinely welded into one; we took both apart in what a good speed-to-lead conversion rate actually is. We set Tier 1 tighter than HBR’s hour because a callback request sitting in a queue is a competitor’s head start, not because anyone measured five minutes.

What happens when a lead breaches the window?

This is the clause that makes sales follow the document, and the one almost nobody writes. If the consequence depends on a manager spotting a miss in Friday’s report, the SLA is decoration.

Two details decide whether this survives a sales floor. Reassignment has to move the commission, not just the record — a breach with no consequence is a rounding error to whoever commits it. And publish breaches per rep: a team average lets three good reps carry the one who never picks up.

What do I do with leads that arrive at 11 p.m. or on a Sunday?

Clause 4 is where US SLAs quietly break, because this answer has a legal boundary the others do not. Under the FCC’s TCPA rules at 47 CFR 64.1200(c)(1), no telephone solicitation may reach a residential subscriber “before the hour of 8 a.m. or after 9 p.m. (local time at the called party’s location)” — their clock, not yours. Florida Statute 501.616(6)(a) is tighter still, at 8 p.m. Sell nationally and an SLA written in Eastern time has already broken its own rule for a Pacific lead at 7 a.m., so store each lead’s time zone at capture. Whether a callback to someone who just filled in your form is a solicitation is a question for your counsel; our page on AI cold calling and the TCPA covers where that line sits. General information, not legal advice.

The clause is three lines. Outside the calling window: respond in writing on the channel the lead used, inside the Tier 1 target, offering times inside the window. Inside the window, outside your roster — evenings, Saturdays, federal holidays: the full Tier 1 target applies, or you write down that it does not. Never let an after-hours lead inherit a business-hours target silently — that is how a 98% dashboard sits on a lead flow where a third of inquiries wait until Monday.

How do I know sales is actually following it?

One number, defined once: response-window compliance = leads first attempted inside their tier target ÷ all leads created in that tier. Weekly, per rep, per tier, business hours and after hours on separate lines. Report the median time to first attempt beside it, never the mean — one lead answered three days late drags a mean far enough to hide a bad week.

Below 80% compliance on Tier 1, stop buying leads and fix the process: you are paying full price for inventory you are not working. Split the after-hours line out because the two failures have different fixes. Business-hours breaches are a routing problem a decent CRM solves. After-hours breaches are a rostering problem, solved with shift coverage, automation, or not at all.

What does a five-minute SLA cost — to miss, and to hold?

Only the first input below is a published figure. LocaliQ’s 2026 Search Advertising Benchmarks put the US average cost per lead for search advertising at $66.69 across all industries and $90.92 for Home & Home Improvement.

• A contractor buys 300 search leads a month at $90.92 — $27,276 in media.
• 28% arrive outside the roster — 84 leads, $7,637 of spend queued until next business day. Measure your own share; this is the input you must not borrow.
• Its CRM shows lead-to-deal of 6% in-window against 2% next day, at $4,000 a deal. Illustrative inputs, not our data and not a benchmark — pull yours from closed-won records.
• 84 × (6% − 2%) × $4,000 = 3.4 deals, $13,440 a month.

That figure is a ceiling, not a promise: the most closing the after-hours gap could be worth, and the number any fix has to come in under. If your cycle is long, run it on cost per booked call — our US cost per booked call benchmarks carry that formula.

The cost of holding it splits the same way. Business hours is a configuration job: lock the timestamp field, build the round robin and its reassignment rule, wire the breach alert, schedule the report. Two to four days of admin in a mainstream CRM, plus 20 minutes a week to read it. After hours is a staffing commitment — five minutes across evenings, weekends and federal holidays is roughly 128 hours a week outside a 40-hour roster, more than three full-time equivalents before leave. Below roughly 20% of leads arriving outside your roster, tightening business-hours routing is the whole job. Above it you are choosing between shift work and speed-to-lead automation.

Kept separate from the research above, because it is a different kind of evidence: in our own client work we typically see speed to lead, fixed alone, roughly triple conversion for a business previously answering inbound in hours. That is an operator claim from campaigns we run — no published sample, no window, not a guarantee — and it does not stack with other fixes, since faster response is part of how contact rate improves.

Frequently asked questions

What should a speed-to-lead SLA target be?

Five minutes on live-intent inbound — paid form fills, callback requests, missed inbound calls — seven days a week, with longer targets for lower tiers. Five minutes is our recommendation, not a published finding: the March 2011 Harvard Business Review article “The Short Life of Online Sales Leads” sets its threshold at an hour, and its audit of 2,241 US companies found 23% never responded to a web-generated test lead at all, with an average response time of 42 hours among those that responded within 30 days. Publish the target you can actually hold and report the breaches.

When does the speed-to-lead clock start?

At the form’s own submission timestamp, in a field reps cannot edit. Starting it when the record lands in your CRM hides every delay from a batch sync, a webhook retry or an hourly aggregator feed — exactly the delay the lead notices and your dashboard does not.

Can I call a lead at 10 p.m. if they just filled in my form?

The FCC’s telemarketing rules at 47 CFR 64.1200(c)(1) bar any telephone solicitation to a residential subscriber “before the hour of 8 a.m. or after 9 p.m. (local time at the called party’s location)”, and Florida stops commercial solicitation calls an hour earlier at 8 p.m. Whether a same-minute callback to someone who asked to be contacted falls inside the definition of a solicitation is a question for your own counsel, not a template. The safe clause: respond in writing on the channel they used, offer times inside the window, and store each lead’s time zone at capture so the window is judged against theirs. General information, not legal advice.

Why does my sales team ignore the SLA we already wrote?

Almost always because it has a target and no clause 3. If missing the window costs nothing — no reassignment, no movement of commission, no name in a shared channel — the SLA competes with everything else on the rep’s screen and loses. Make the breach reassign the lead automatically, move the credit with it, and publish breaches per rep.

What is a realistic response-window compliance rate?

There is no published benchmark, so report against yourself. The threshold we apply when auditing an inbound desk is 80% on the top tier in business hours, after-hours compliance stated separately. At the US average search cost per lead of $66.69 in LocaliQ’s 2026 benchmarks, every 100 breached leads is $6,669 of media you paid for and did not work.