Speed to lead: what the numbers say

Most businesses trying to get more from their website reach for traffic, then for conversion rate, then for the copy on the page. Reply time is rarely on the list, despite costing nothing and being entirely within their control.

Part of the reason is that it is invisible. Nobody sees the enquiry that went cold, because the person simply contacted somebody else and never mentioned it.

The other part is that it is not measured. Almost no small business can say what its median response time was last month, which makes it impossible to notice it getting worse.

It is worth measuring for that reason alone, before any attempt to improve it.

It is also the rare improvement that costs nothing and needs nobody's approval. No budget, no new tool, no negotiation with anyone: the same team replying to the same enquiries in a different order gets a different result. That is unusual enough to be worth taking seriously even before the attribution data tells you which sources deserve the attention most.

The best-known work here is the Harvard Business Review study of online sales leads published in 2011 by James Oldroyd, Kristina McElheran and David Elkington, which examined a large volume of lead response data.

Two findings are quoted constantly. Firms that responded within an hour were around seven times more likely to have a meaningful conversation with a decision maker than those that responded an hour later, and more than sixty times more likely than firms that waited a day.

The second finding was that most companies were nowhere near that. Average response times ran to many hours, and a substantial share of enquiries were never responded to at all.

The study is over a decade old and the underlying behaviour has, if anything, moved further in the same direction as people have become accustomed to immediate replies from everything else.

The mechanism is not mysterious and it is worth understanding, because it tells you where to spend effort.

A person who submits an enquiry is at peak intent at that moment. They are on your site, thinking about the problem, with your page open. An hour later they are doing something else and the context is gone.

They are also usually contacting more than one supplier. The first useful reply frames the comparison, and often ends it, because a decision made under mild time pressure tends to favour whoever made it easy.

And there is a signalling effect. A fast reply reads as competence and interest before anything has been said about the work itself.

The number you want is the gap between the submission timestamp and your first substantive reply, measured across a month, reported as a median rather than an average.

The median matters because averages are wrecked by the enquiry somebody found three weeks late. One such case moves the mean enormously and tells you nothing about the typical experience.

If submissions land in a spreadsheet the calculation is a subtraction between two columns, provided somebody records the reply time. That is the only new habit required, and it is the one that makes everything else visible.

Look at the distribution too. Most businesses find a fast cluster and a long tail, and the tail is where the losses are.

Very few businesses are uniformly slow. The common pattern is that most enquiries are handled quickly and a minority sit for days.

The causes are mundane. It arrived at the weekend. It went to somebody on leave. It landed in spam. It came in during a busy week and slipped down the inbox. Nobody was sure whose it was.

Because the median looks fine, the tail is invisible in the headline number, which is why the distribution is worth looking at rather than the summary.

Fixing the tail is also the cheapest improvement available, since it requires no change to how the fast cases are handled.

Three things, in order of impact.

Knowing the enquiry exists. This is where a missing notification costs real money, and it is why a record that does not depend on email delivery matters more than it appears to.

Knowing whose it is. An enquiry with no owner waits for somebody to feel responsible, which takes longer than any assignment rule would.

Having something to send. Most first replies are the same handful of shapes, and having them written in advance turns a twenty-minute task into a two-minute one. That is usually the difference between replying now and replying later.

The ownership part is worth dwelling on, because it is where most of the delay hides. An enquiry everybody can see and nobody owns waits longer than one addressed to a specific person, even when the specific person is busier. A single owner column in the sheet removes the ambiguity entirely.

An instant autoresponder is not a fast reply, and treating it as one is how businesses convince themselves the problem is solved.

The confirmation message does useful work: it reassures, sets expectations and buys patience. It does not start a conversation, and the research is about conversations.

The reasonable arrangement is both. An immediate confirmation with a realistic timeframe, followed by a genuine human reply well inside that timeframe.

What you should not do is let the autoresponder become the reason nobody feels urgency about the real one.

A meaningful share of enquiries arrive outside working hours, and few small businesses can answer them at midnight.

The answer is not pretending otherwise. It is setting an expectation that is true and then beating it: a confirmation that says you will hear from us by the end of the next working day, followed by a reply first thing.

What loses the lead is silence of indeterminate length, not the delay itself. People are reasonable about waiting when they know how long.

It is also worth knowing when your enquiries actually arrive. If a third of them come in on Sunday evening, a Monday morning routine is worth more than any amount of weekday tuning.

1. Alert on arrival through a channel you genuinely watch.
2. Assign an owner immediately, even if it is always the same person.
3. Reply with something real, however brief, rather than waiting to compose a full answer.
4. Record the reply time in a column beside the submission.
5. Look at the median and the tail once a month.

The third point does most of the work. A short reply acknowledging the specifics and saying when a full answer will follow beats a thorough response two days later, on both the research and on common sense.

If none of this is measured today, start with the simplest version: for the next twenty enquiries, note the arrival time and the first reply time in two columns.

That sample is enough to tell you whether you have a speed problem, a tail problem, or neither, and it costs nothing beyond the discipline of filling in a cell.

Most businesses that do this discover the tail they did not know about, and fixing it is usually a routine rather than a purchase.

The two columns live beside the enquiry in the same sheet, which is what makes the habit stick. A measurement kept somewhere separate from the work stops being recorded within a fortnight, and then the number quietly reverts to being an opinion.

How fast is fast enough?

Does an automatic confirmation count as a response?

What should I measure, mean or median?

How do I record reply time without a CRM?

What about enquiries that arrive at the weekend?

Is a short reply better than a thorough one?

Does this apply to low-value enquiries too?

What is the most common cause of a slow reply?