An enquiry lands at 4:10pm on a Thursday. You are on a job, then the school run, then dinner. You reply at 8:30 the next morning with a good answer and a fair price. The customer thanks you politely and says they have already booked someone.
Everyone who runs a small business has a version of that story, and plenty of blog posts will attach a dramatic percentage to it. The trouble is that those numbers come from other businesses, other sectors and often nowhere traceable at all. The cost of slow response time in your business depends on your customers, and your inbox already holds the evidence.
You can build the model yourself in an evening with a spreadsheet.
What you need before you start
Gather these. You do not need all of them to be perfect.
- An enquiry list. The last 60 to 150 enquiries from email, contact form and website chat, with the date and time each arrived.
- First reply times. When someone from your business first responded. Your sent folder has this.
- Outcomes. Whether each enquiry became paid work. Your invoicing tool or job book has this, even if matching takes some patience.
- Average margin per job. Gross margin, not revenue. A rough figure is fine.
If you use a website assistant already, its analytics give you conversations, leads and response time, which saves a lot of matching work for that channel.
Building the model
- Calculate each gap. For every enquiry, subtract arrival time from first reply time. Record it in hours.
- Mark the non-replies. Some enquiries never got an answer at all. Give them their own label rather than an enormous number, because they distort averages.
- Bucket the gaps. Under 1 hour, 1 to 4 hours, 4 to 24 hours, over 24 hours, and never replied. Use different buckets if your trade works on a different rhythm.
- Count wins per bucket. How many enquiries in each bucket became paid work?
- Calculate a win rate per bucket. Wins divided by enquiries.
- Compare the fast bucket with the others. The gap between them is your cost of delay, per enquiry.
- Scale it. Multiply that gap by the number of enquiries currently landing in the slow buckets each month, then by average margin.
A worked example
These figures are invented for illustration. Put yours in their place.
Say a garden landscaping business pulls 120 enquiries from the last six months and buckets them:
| Reply time | Enquiries | Became paid work | Win rate |
|---|---|---|---|
| Under 1 hour | 22 | 8 | 36% |
| 1 to 4 hours | 31 | 9 | 29% |
| 4 to 24 hours | 44 | 9 | 20% |
| Over 24 hours | 15 | 2 | 13% |
| Never replied | 8 | 0 | 0% |
Here, 67 of the 120 enquiries (the 4-to-24-hour, over-24-hour and never-replied rows) sat in slow buckets. That is roughly 11 a month.
If those 11 had converted at the under-1-hour rate of 36%, that would be about 4 jobs a month. At their actual combined rate (11 wins from 67, about 16%) they produce under 2. The difference is roughly 2.2 jobs a month.
With an average margin of $800 a job, slow replies are costing this business in the region of $1,760 a month. Not a borrowed statistic: its own history.
That figure is a ceiling, not a promise. It assumes every slow enquiry could have been answered within the hour and would then have behaved like the fast ones. Some could not; a few of the slow ones arrived at 3am on a Sunday. A more cautious version assumes you could only move half of them into the fast bucket, which halves the figure to roughly $880 a month. Put both numbers in front of yourself. If even the cautious one is larger than what faster replies would cost you, you have your answer.
Reading the result without fooling yourself
A table like that invites overconfidence, so check a few things before acting on it.
Look for hidden causes. Enquiries that arrive late on a Friday may be both slower to answer and naturally lower intent. If your slow buckets are full of vague "just browsing" messages, speed may not be the only story. Skim a sample of each bucket.
Mind small samples. Fifteen enquiries in a bucket is thin. Treat the percentages as a direction, not a precise figure.
Split slow from never. A never-replied enquiry is a process failure, usually a form that went to a shared inbox nobody watches. Fixing that costs nothing. Slow replies are a capacity problem, and the fixes cost time or money. What missed enquiries actually cost your business goes further on counting the never-replied group.
Check whether speed matters at all. If your under-1-hour and 4-to-24-hour win rates are nearly identical, your customers are deliberating rather than racing. That is useful to know too, and it means your money belongs somewhere other than faster replies.
Where the slow replies come from
Once you know the size of the cost, look at when slow enquiries arrive. Sort the slow buckets by arrival hour and day. In many service businesses they cluster in evenings, weekends and the hours when everyone is out on jobs.
Then look at channel. A slow email reply and a slow website form reply are often slow for different reasons. Emails land in a personal inbox that someone eventually checks. Form submissions often go to an address nobody reads until a customer complains. If your never-replied row is mostly form submissions, you may have a plumbing problem rather than a people problem, and it can be fixed in five minutes by changing where the form sends.
The timing pattern points to the fix. If slow replies cluster during working hours, you need a better inbox routine. If they cluster after hours, you need something answering while nobody is at a desk, whether that is a rota or a website assistant that answers from your content and captures details for the morning. Reducing lead response time from hours to seconds looks at the options.
Your next hour
Open your sent folder and export your last 60 enquiries into a spreadsheet with arrival time, first reply time and outcome. Build the five-bucket table above before doing anything else. Then run your contact page through the website response time checker to see what a visitor experiences today, and repeat the bucket table in three months to see whether the gap has closed.