Speed to lead is the single operational metric that most reliably predicts whether an enquiry becomes a booked job. It is also the one most trades and service businesses have never measured.
This page sets out what the published research actually found, why the problem is structural rather than a matter of effort, and what can realistically be done about it.
What "speed to lead" means
Speed to lead is the elapsed time between a prospective customer making contact, calling, texting, submitting a form, and your business responding with something the customer can act on. Not a voicemail greeting, not an auto-acknowledgement, but a real response.
The shorter that window, the higher the probability the enquiry converts. The relationship is not gentle: it falls off a cliff rather than sloping away.
What the research found
Two studies do most of the work here, and they are commonly conflated. They are separate pieces of work with different methods.
The Harvard Business Review audit (2011) submitted test enquiries to 2,241 companies and measured the response. Average first response among those who replied at all was 42 hours. Almost a quarter 23%, never replied. Firms that responded inside the first hour were roughly seven times more likely to reach a meaningful conversation with a decision-maker than firms responding in the second hour.
The MIT Sloan / InsideSales response study (2007), earlier work, and the source of the five-minute rule, analysed three years of data across more than 15,000 leads and 100,000 call attempts. It found the odds of qualifying a lead were about 21 times greater at a five-minute response than at thirty minutes, and the odds of making contact at all about 100 times greater.
Both measured web-form enquiries rather than inbound phone calls. That limitation matters and is set out in full on the lead response time statistics page. There is no equivalent peer-reviewed research on missed calls to Australian trades and service businesses, and the missed-call figures circulating online generally trace to vendor surveys rather than published studies.
The five-minute window
Five minutes is the threshold the MIT data points to, and it is a useful planning number even though it came from web enquiries. Inside five minutes, the customer is still engaged with the problem that made them reach out. At thirty minutes they have usually contacted someone else. Past an hour you are frequently ringing back to be told the job is already booked.
For a phone call the window is likely tighter still, for a reason that has nothing to do with the research: a caller who hits voicemail has the next business one tap away, and the attempt leaves no record for you to follow up.
Why this is structural, not a discipline problem
The instinct is to treat slow response as a failure of organisation, be more on top of the phone, check messages more often. That misreads the problem.
Enquiries arrive precisely when you are least able to take them. Mid-job. On a customer's site. Driving between appointments. In the evening, when a homeowner has just noticed the problem. On a weekend or public holiday, when urgency peaks and coverage is thinnest. The distribution of incoming enquiries is close to the inverse of your availability to answer them.
No amount of diligence closes that gap, because the constraint is that you cannot be doing the work and answering the phone at the same time. Which is why the fix has to be structural too.
What actually closes the gap
Two mechanisms, usually run together:
- The call gets answered. An AI voice agent picks up, qualifies the job against your criteria, and books it into your calendar while the caller is still on the line. Response time is effectively zero, at any hour.
- A call that rings out becomes a conversation. Missed-call text-back sends an SMS within seconds, so the customer is in a conversation with you before they dial the next number.
Both remove the human bottleneck from first contact. You still close the job, the automation handles the response and the qualification.
Measuring it in your own business
Two numbers are worth knowing. First, how many inbound calls actually go unanswered each week, take this from your phone records rather than from impression, because unanswered calls leave no trace in the diary and almost everyone guesses low. Second, how long a typical callback takes from missed call to conversation.
Multiply the missed calls by your average job value and your normal close rate and you have the weekly cost. The lead leak calculator does the arithmetic, and the worked example on the cost of a missed call walks through it with real figures.