Map your customer contact

Why you measure first

“We miss a lot of calls” and “nothing slips through here”: you hear both often, and usually neither is quite right. Five missed calls may come from two customers who keep trying. And a WhatsApp message that arrives in the evening and is only seen two days later does not feel missed to you. To the customer, it does.

That is why you measure first, before you change anything. For one week you record every customer contact in a simple spreadsheet. Afterward you know how many unique requests you receive, how many of them got a response, how quickly, and where requests are left waiting. That is your baseline: the starting point for the next lessons, and the yardstick you use in Lesson 5 to see whether things are improving.

How a request moves through your business

A request can get lost at every step: a call is not answered (respond), nobody writes down that the customer wants a quote (record), or the promised quote is never sent (follow up). The audit shows where things go wrong in your business.

Step 1: list all your channels

Customers choose for themselves how they get in touch. So list every place where a question can come in, including the “unofficial” ones, such as the owner’s private number that regular customers know, or messages via Facebook.

1. Write down every channel

   Note every channel through which customers reached you in the past month. Ask your team as well: they often know better where customers come knocking.

2. Note who checks the channel

   If it says “everyone” or “nobody in particular,” you already have a first clue to a bottleneck.

3. Find out where you can see missed contacts

   For example in the list of missed calls, the unread chats or the inbox. If you cannot see it anywhere, put a tally sheet next to the phone.

4. Decide what you measure

   Include every channel on which customers expect an answer. Record walk-ins only if they have a question, not every customer who comes in to pay or pick something up. If you leave a channel out, write that down as a limitation of your measurement.

Step 2: agree on what counts as a request

Agree with your team beforehand on what you record and how, otherwise you will end up comparing apples to oranges. A request is any question or wish from a customer that requires your business to do something: answer, give a price, make an appointment. The spreadsheet has four contact types.

The difference between unique requests and contact attempts is the heart of an honest measurement. If a customer calls three times without getting through and then sends a WhatsApp message, that is four contact attempts but one request. You record all four, so you can see how often customers have to try: the first as a “New request” row, the rest as “Repeat message” rows. If you count all four as requests, the problem looks four times as big.

Missed calls from unknown numbers

If you cannot reach the caller when you call back, record the call as a “New request” with the subject “unknown (missed call)” and the outcome “Unknown” in the log. Count this group separately in the analysis. It may have been a customer, but it may also have been a wrong number. That is why the opportunity estimate works with an assumption about what share really was a customer.

Step 3: measure for seven days

A week is short enough to keep up and long enough to see busy and quiet days. Choose an ordinary week, without a holiday or a big promotion. Measure for seven days in a row, including days when you are closed: messages and calls still come in.

Preparation: half an hour

1. Choose a target time per channel

   How quickly do you want to give a first response? You choose that yourself; it is not a standard. For example: answer the phone or call back within 2 hours, WhatsApp within 1 hour, email the same business day, and outside business hours before 10:00 AM the next business day. That way everyone judges “late” in the same way.

2. Assign one person per channel to do the recording

   That prevents duplicate rows and gaps. In a small team, one person can do everything.

3. Set up the spreadsheet

   Download the template, delete the fictional example rows and share the file only with the colleagues who are recording. Prefer paper? Use a notebook with the same columns and transfer it at the end of the week.

4. Explain to the team how you record

   Use the team instructions below and agree on who updates the log and when.

What you record per contact

Fill in the columns from Status onward only on the first row of a request, and leave them empty for repeat messages and spam. That way each request counts once in the Summary.

During the measurement week

• Update the log at a fixed time every day, with the missed calls, the unread chats and the inbox alongside.
• If a response comes later, still fill in the status and response time on that request’s row.
• If a question has not been resolved by the end of the week, leave “Open” as the outcome. That is exactly what you want to see.
• Try not to suddenly do everything differently this week: you are measuring how things go now. Improving comes in the next lessons.

Step 4: count and calculate

After seven days, first update the open rows. The Summary tab then works out most of it by itself. This is what each figure means, so you can also work it out on paper:

Average or median?

One message from Friday evening that is not answered until Monday pushes your average up considerably. The median is less affected by that: it is the middle value when you put all response times in order from smallest to largest (with an even number, the average of the two middle values).

Example (fictional): five WhatsApp responses after 5, 10, 20, 35 and 2,880 minutes. The average is (5 + 10 + 20 + 35 + 2,880) ÷ 5 = 590 minutes, almost ten hours. The median is 20 minutes. Both are correct: most customers got a quick answer, but one customer waited two days. So note both, and look separately at contact during and outside business hours.

Break down your figures

• By channel: where are the most requests left waiting?
• During and outside business hours: how much comes in when you are closed, and how long does it wait?
• By time of day: are there regular busy times, such as around noon or just before closing time?
• By owner: are there requests without an owner, or does everything end up with one person?

A filter on the columns in Google Sheets or Excel is enough for this.

Step 5: keep facts and assumptions apart

Your audit produces facts: things you have counted. As soon as you explain what those facts mean, you are making assumptions. That is fine, as long as you label them as assumptions. The danger lies in an assumption that gets passed on as a fact, such as “we are losing SRD 36,000 every week.”

Put two headings under your figures: Measured and Assumed. Write each assumption underneath, with the figure you use. Then someone else, or you yourself a month from now, can see what the calculation rests on.

Step 6: carefully estimate what you may be missing

What do those unanswered requests cost you? The honest answer: you do not know exactly. But you can make a hypothetical calculation, with assumptions you choose and write down yourself.

• Missed unique requests (measured): new requests without a response, or answered so late that the customer had reasonably already gone elsewhere. No repeat messages, spam or questions from existing customers.
• % genuinely qualified (assumption): what share was really looking for something you provide?
• % reachable (assumption): what share would you still have spoken to if you had responded in time?
• % that becomes a customer (assumption): look at how many answered new requests in your audit ended with “Appointment/order” as the outcome. For the missed group it remains an assumption, but a well-founded one.
• Average order value: from your invoices or sales records, not from memory.

Why the result is not profit and not a promise

Even with honest assumptions, the calculation remains a model. Keep four things in mind:

• Repeat callers. People who keep trying are often helped in the end. Someone who was answered after three attempts is a service problem (the customer had to make an effort), but not a missed request. So count only requests that really got no response.
• Capacity. If your calendar or your shop is already full, you may not even be able to take on extra work. Then it is more about answering customers properly, referring them elsewhere or offering a later time.
• Differences between types of business. A restaurant gets many requests with a low order value, a hotel few with a high booking value, and regular customers of a salon often simply call back later. For a practice or clinic, a missed call is mainly about being reachable and about care; it is better not to think in terms of revenue there. Do not copy percentages from another business, not even from this example.
• Unknown attribution. You do not know whether a customer you called back would have come anyway without that call. And if your revenue rises after you have improved something, that may also be due to the season, prices or the competition.

Use the result to decide whether a bottleneck is worth tackling, and roughly what a solution may cost. Not as a promise to yourself, your team or the bank.

Calculate with your own figures

Enter your own numbers and assumptions in the calculator; the formulas are shown alongside and nothing is stored. Among the requests answered late, count only those where the customer had reasonably already gone elsewhere. The calculator has one field for the share that was a real opportunity and one for the share that would become a customer with timely follow-up. If you want to include reachability separately, as in the example, multiply first: 60% reachable × 40% book = 24%. Enter that 24% as the share that would become a customer.

The calculator also has its own page, which is handy when you calculate again later with new figures:

Worksheet: your top 3 communication bottlenecks

Finish the audit with the three bottlenecks that leave the most requests waiting or keep customers waiting longest. Three is enough: if you tackle everything at once, you finish nothing.

1. Gather the problem cases

   Filter by the statuses “Unanswered” and “Answered late” first, then by requests with a follow-up date that have not been resolved. Promises that are not kept often weigh more heavily than missed calls.

2. Look for patterns

   One channel? Fixed times? Outside business hours? No owner? Whatever recurs across several requests is a bottleneck; a single chance outlier is not.

3. Choose the three biggest

   Weigh how many requests it affects and how annoying it is for the customer. A quote that never arrives weighs more than a price question answered an hour later.

4. Fill in the worksheet

   For each bottleneck, write down the evidence from your audit (a figure, not a feeling), the effect on the customer, a first small measure, and who carries it out from when.

Where to go next

• Phone not answered at busy times, or nobody calling back? Continue with Lesson 2: phone and callback system.
• WhatsApp messages left waiting, everything on one phone, the same questions over and over? Then Lesson 3 on WhatsApp Business will help.
• Promised quotes and callbacks without an owner or a date? You tackle that in Lesson 4 on appointments and follow-up.
• Keep the totals from the Summary: in Lesson 5 you use them as a point of comparison.

Checklist for this lesson

Downloads for this lesson

Test your knowledge

What to do now

• Choose a start date, share the team instructions and keep the log for seven days in a row.
• Right after the measurement week, schedule an hour for the calculations, the opportunity estimate and the worksheet with your top 3 bottlenecks.
• Then continue with Lesson 2, or with the lesson that fits your biggest bottleneck.
• Keep the totals from the Summary: in Lesson 5 you compare your new weekly figures with them.