Last updated: August 7, 2026
Two vendors get the same demo request at 9:14 on a Tuesday morning. One replies at 9:16. The other replies the following afternoon. By then the buyer has already had a conversation, seen pricing, and formed an opinion. The second vendor is not competing on product anymore — it is competing against a relationship that already exists.
That gap has a name: speed to lead. It is the elapsed time between a prospect raising their hand and a human or AI agent making meaningful contact. It is one of the few revenue levers that requires no new budget, no new headcount, and no change to your product — and it is the one most B2B teams quietly fail at.
Speed to lead is the interval between an inbound signal — a form fill, a pricing-page chat, a WhatsApp message, a missed call, a demo request — and the first substantive response the prospect receives. The word substantive is doing real work in that sentence. An automated "thanks, we got your message" receipt is not a response. A generic nurture email that lands three hours later is not a response. A response is an interaction the buyer can act on: an answer to their question, a qualifying question back, or a booked meeting.
The two terms measure the same clock. Teams get into trouble over what starts and stops it. Many CRMs record the first logged activity, so an auto-reply, a sequence enrollment or a rep marking a record "attempted" all count as a response. The number on the dashboard is not the number the buyer experienced.
A cleaner definition: start the clock when the lead is created, stop it when a two-way conversation begins. Anything else flatters you.
The "five-minute rule" gets repeated so often in sales content that it has started to sound like folklore. It is not. The original academic work behind it is a Harvard Business Review study, "The Short Life of Online Sales Leads" by James Oldroyd, Kristina McElheran and David Elkington. The researchers audited 2,241 U.S. companies by submitting a web-generated test lead to each and timing the reply. Among firms that responded at all within 30 days, the average response time was 42 hours. Twenty-three percent never responded. The same Harvard Business School research record documents the companion analysis of 1.25 million leads, which found that firms making contact within an hour were roughly seven times more likely to qualify the lead than those responding in the second hour.
A later field test produced a similarly bleak picture. Drift mystery-shopped 433 B2B software companies and published the results: only 7% replied within five minutes, and 55% had not responded at all after five business days.
Two things follow from this. First, the advantage of being fast is enormous. Second — and more usefully — the advantage is available, because almost nobody is taking it. Speed to lead is one of the last competitive edges in B2B that most of your competitors have simply declined to pick up.
| Response window | What the buyer is doing | Practical outcome |
|---|---|---|
| Under 5 minutes | Still on your site, context fresh | Live conversation; you set the evaluation criteria |
| 5–60 minutes | Moved on, reachable | Callback or reply likely; you are one of several |
| 1–24 hours | Has spoken to a competitor | You are responding to a framed opinion, not a blank slate |
| Over 24 hours | Intent has decayed | Effectively a cold lead requiring re-nurture |
Nobody sets out to answer a demo request two days late. Slowness is almost always structural, and it tends to come from one of three places.
A lead arrives, waits for enrichment, gets scored, gets assigned to a territory, then lands in a rep's task list — which the rep checks between meetings. Each step is defensible alone; the sum is a two-hour delay before a human sees the record. Teams that adopt AI lead routing that assigns inbound leads in minutes usually find routing was never the bottleneck — the handoff to human attention was.
Inbound demand does not respect working hours. A meaningful share of B2B form fills arrive in the evening, at weekends, or from a time zone where your team is asleep. If your average response time looks fine on weekdays and catastrophic overall, you do not have a speed problem — you have a coverage problem, and hiring more reps in the same time zone will not fix it.
Buyers raise their hand on WhatsApp, on a chat widget, or by calling and hanging up when nobody picks up. Sales teams are still organised around email and scheduled calls, which produces systematic delay on exactly the channels where expectations are highest. It also explains why unanswered inbound calls quietly cost so much revenue.
Before you can improve the number, three rules make it trustworthy.
Use the median, not the mean. A handful of leads answered three weeks later will drag an average into uselessness. The median tells you what a typical buyer experienced. Report the 90th percentile alongside it so the tail stays visible.
Exclude automated receipts. If an autoresponder stops your clock, your dashboard will show a response time of eight seconds and your pipeline will not improve. Count only two-way interactions.
Segment by source and by hour. Aggregate speed to lead hides the two failures that matter most: the channel you are slow on, and the eight hours of the day when you are absent.
Pair the timing data with quality data. Fast contact on leads that were never going to buy is just faster waste, which is why speed work pays off most when it sits on top of reliable lead scoring that identifies real buyers rather than everyone who downloaded a PDF.
Assignment introduces waiting. Let an automated first response open the conversation and gather context, then hand a warm, partially qualified thread to a rep. This inverts the usual order — qualify first, then respond — and that inversion is where most of the time saving comes from.
Evenings, weekends, and holidays are where median response times go to die. Whether you solve this with a follow-the-sun team or with automation, the requirement is the same: no arrival window should have a different service level than any other.
If someone messages on WhatsApp, replying by email adds hours and signals that your process matters more than their preference. Meeting buyers on the messaging channels they already use removes an entire round trip.
A five-minute target that nobody reports on is a wish. Publish median and 90th-percentile response time by source, weekly, in the same place the team reviews pipeline. Metrics that are visible get managed.
Speed to lead has stayed broken for fifteen years because the honest fix used to be unaffordable. Sub-five-minute coverage across every channel, every hour, in every language your market speaks is not a staffing problem you can solve by hiring harder. It is an arithmetic problem.
AI agents change that arithmetic. An AI worker can pick up a WhatsApp message, a web chat, or an inbound form at any hour, ask the two or three questions that separate a real opportunity from a browser, check availability, and book the meeting into the right rep's calendar — all inside the window where the buyer is still paying attention. The rep no longer spends their morning chasing; they walk into calls that are already qualified.
This is the specific job Alba, Darwin AI's inbound sales worker, was built for: responding to every inbound lead in seconds, qualifying in natural conversation, and handing off to a human at the point where a human actually adds value. The measurable change is usually not "we answer faster" in the abstract — it is that the weekend and after-hours cohorts, which used to convert at a fraction of the weekday rate, start converting like everything else.
One caveat worth stating plainly. Instant does not mean thoughtless. An AI first response that ignores what the buyer wrote is worse than a slower human one. The value comes from pairing speed with a genuinely useful first message, which depends on how well the agent handles qualification and prioritisation mid-conversation.
It is also worth setting expectations about the ceiling. Faster response improves the odds of engaging a lead; it will not make a bad-fit prospect a good one. What it does reliably is stop you losing deals you had already earned. With the Ebsta and Pavilion 2025 GTM Benchmarks recording average B2B win rates falling to around 19%, those are the cheapest deals available to you.
Under five minutes for high-intent inbound (demo requests, pricing enquiries, chat) and under one hour for lower-intent content leads. Context for how hard that is: in Drift's test of 433 B2B companies, only 7% replied within five minutes, so hitting the target puts you in a small minority.
The underlying mechanism — buyer attention decays fast and the first responder frames the evaluation — has held up. The Harvard Business Review study that popularised it found an average response time of 42 hours across 2,241 audited companies, and later field tests have found similarly slow behaviour. Treat five minutes as a design constraint rather than a magic threshold.
Yes, though the mechanism differs. In a six-month evaluation, being first does not close the deal on day one — it gets you into the shortlist and lets you shape the requirements document everyone else responds to. Slow first contact often means you are never evaluated at all.
Not if it is useful and honest about what it is. Buyers object to irrelevance and to being trapped, not to automation. An agent that answers the actual question, offers a real time slot, and escalates to a human on request generally outperforms a delayed human reply.
Take the timestamp of lead creation and the timestamp of the first two-way interaction, exclude automated receipts, and report the median and 90th percentile by lead source and by hour of arrival. Averages hide the failures you most need to see.
Alba qualifies inbound leads in natural conversation across WhatsApp, chat and email — around the clock — and books the meeting before your competitor has opened their inbox.
See how Alba works