Most firms don't have a lead problem.
They have a response-time problem.

Every founder I talk to who says they need more leads has, sitting in an inbox somewhere, a list of people who already raised their hand and never heard back fast enough to matter. That is not a marketing problem. It is a plumbing problem, and it is the cheapest thing in the business to fix.

The reason it goes unfixed is that it doesn't feel like a problem. Nobody's dashboard has a red light for it. The lead came in, someone eventually called, the call didn't convert, and everyone concluded the lead was bad. The lead was fine. The clock beat you.

What the research actually says

The most rigorous public work on this is "The Short Life of Online Sales Leads," published in Harvard Business Review in March 2011 by James Oldroyd, Kristina McElheran and David Elkington. They audited how fast 2,241 US companies responded to a web-generated lead, and separately analysed roughly 1.25 million leads across dozens of B2C and B2B firms.

Two findings from that paper are worth memorising:

And the distribution of behaviour, which is the part that should interest you commercially:

Read those two tables together. The odds of a real conversation fall off a cliff inside the first hour, and roughly six in ten firms are on the wrong side of that cliff. Nearly a quarter never reply at all. That is not a competitive market. That is a market where showing up on time is a strategy.

There is an older and even more widely quoted study — Oldroyd's Lead Response Management work with InsideSales.com, around 2007, covering roughly 15,000 leads and 100,000 call attempts — which is the origin of the famous "five minute" rule. I'd treat the specific multipliers from that one carefully: they have been retold so many times in so many blog posts that the numbers have drifted from the source. The directional finding is sound and consistent with the HBR paper. The exact figure you saw on a LinkedIn carousel probably isn't.

Why the curve is that steep

The multipliers look implausible until you understand what you are actually competing for. You are not competing for a person. You are competing for a state that person was briefly in.

Someone filling in a form or picking up the phone is at the top of a very short window in which the problem is present in their mind, they have set aside attention for it, and they have not yet talked to anyone else. That window closes on its own. It closes faster if a competitor gets there first — and the research says a competitor probably will, because the same buyer typically contacted several firms in the same sitting.

So the question is never "will this person still be interested tomorrow." It is "will this person still be in the mode they were in when they wrote to you." Usually not. Tomorrow they are in a meeting, the problem has receded, and your call is an interruption rather than an answer. Same lead, different person.

The touches nobody counts

Response time gets all the attention because it is one number and it is easy to shame people with. The larger leak is downstream, and almost nobody instruments it: what happens on touch two, three and five.

A single fast response is not a system. Most deals that die don't die because the first call was slow — they die because the first call happened, went fine, and then nothing happened for eleven days because the person who owned the follow-up was busy doing the work that the last deal generated. That is the structural trap of a founder-led firm: your capacity to follow up is inversely correlated with how well the business is doing.

If you want one diagnostic to run this week, it isn't average response time. It's this: pick the last twenty inbound conversations that didn't close and count how many simply stopped, with no decision either way. In most firms I have looked at, it's more than half. Those are not lost deals. They are unattended ones, and they are the cheapest pipeline you will ever have access to, because you already paid to acquire them.

Why the fix is not hiring

The instinctive answer is to put a person on it. That mostly fails, for three reasons.

• A person has a shift. Leads don't. A meaningful share of inbound arrives outside working hours, and a human-only process guarantees those start the next morning at best.
• A person degrades under load. Response discipline is perfect in a slow week and collapses in a busy one — which is precisely when the pipeline is most valuable.
• A person doesn't leave a trace. If follow-up lives in someone's memory and their sent folder, you cannot see the leak, so you cannot fix it, so you conclude you need more leads.

What actually holds is boring: an intake path that logs every inbound the moment it lands, an acknowledgement that goes out immediately and honestly, a cadence of follow-ups that fires whether or not anyone remembers, and a single view of which conversations are open and which are stalled. None of that is sophisticated. All of it is infrastructure, and infrastructure does not have a bad week.

The firms that win inbound are rarely the ones with the best answer. They are the ones whose answer arrives while the question is still warm.

What to measure, starting Monday

1. Time to first human response, measured from arrival — not from when someone opened the email. Track the median and the worst case. The worst case is the one that tells you whether you have a system or a habit.
2. Share of inbound answered within one hour during business hours, and the same figure outside them. The gap between those two numbers is the size of your automation opportunity.
3. Stall rate: of conversations opened in the last 90 days, what share have had no contact in 14 days and no explicit outcome. This is the number that usually shocks people.
4. Touches before close, for the deals that did close. Then compare it to the touch count on the ones that stalled. The stalled ones almost always stopped short of the number that works.

If you cannot produce those four numbers today, that is the finding. You are running the most important part of the revenue engine without instrumentation, and every conversation about lead volume is happening on top of an unknown.