How do you know if your sales team is replying?

What a phone shows you

Almost nothing that survives contact with a second question.

Pick up a company handset and you see conversations. You can read them, scroll back, and come away with an impression. What you cannot do is answer the question you started with.

A phone shows threads, not intervals. Every message carries a timestamp, so the gaps are technically there. Reading 40 threads and subtracting timestamps by hand is not something anyone does twice.

It also shows one number. If you have 12 reps on 12 company numbers, the handset in your hand accounts for 1 of them. The other 11 are wherever those reps are.

So the phone answers “is this rep replying to this customer”, once, for as long as you are willing to keep scrolling. It does not answer “is my team replying”, and it never answers it on a Tuesday when you have 20 minutes.

What a rep reports

Something closer to the truth than you might expect, and still not usable.

Ask a rep how fast they reply and most will say something honest. They will also be wrong, in a specific and predictable direction.

People remember the conversations they handled. The enquiry that arrived at 7pm on Friday and got answered Monday morning is not a memory, it is an absence. Nobody recalls the message they never opened.

Self-reported numbers also arrive in the wrong shape. A rep says “usually within the hour”, and that is an average of their good days weighted by what they can bring to mind. You wanted a distribution and you got a feeling.

There is a fairness problem underneath this too. If the only record is what each rep tells you, then the rep who reports carefully looks worse than the rep who rounds down. You end up measuring candor.

What an archive shows

The thing a phone and a self-report both lack is a record that exists whether or not anyone chose to create it.

Link your company numbers as companion devices and every message on them is written to an archive as it arrives. Same mechanism as WhatsApp Web: 1 QR scan, about a minute per number, nothing installed on anyone’s phone. Reps are told the archive exists before their number is linked.

From that point the intervals are arithmetic rather than recollection. The clock starts at the customer’s message and stops at the first reply from anyone on your side. Deleted and edited messages arrive as new events with flags, so a recalled message does not quietly remove itself from the count.

Two honest limits. Capture starts at linking, so the weeks before that are not in the archive and no amount of reporting will conjure them. And speed is not quality: a reply of “noted” in 30 seconds scores better than a careful answer at 20 minutes, which is why nobody should run a team on response time alone.

Every look at the archive is itself recorded. Opening a conversation, running a search and downloading an export are each written to an access log before anything appears on screen, and rows cannot be edited, deleted or switched off. That log is read by your workspace owner and any auditor you bring in, not by managers browsing each other.

The four numbers

Once the record exists, the question stops being philosophical. These are the 4 worth putting on a wall, and they are the same 4 covered in what to measure first.

Median first response. How long a customer waits before anyone answers. Median rather than average, because one weekend drags an average somewhere useless.

Waiting right now. A live count, not a report. This is the only one of the 4 you can act on today. Set a reply SLA and the same list becomes the customers waiting past it; SLA tracking keeps both.

Unique customers engaged. Distinct humans spoken to in the period. It separates a busy week from a productive one.

Conversations gone quiet. Threads past whatever threshold would embarrass you. Set the threshold yourself; two days is a common starting point and a deliberately unambitious one. It is 1 of the 4 follow-ups in how to track sales follow-ups on WhatsApp.

Four numbers is not a dashboard. It is the smallest set that answers the question, and the reason to stop at 4 is that a fifth invites a sixth.

Where this is the wrong tool

If your reps sell on their personal WhatsApp accounts, this does not apply and should not be made to. An archive covers the company-owned numbers you link, and nothing else.

If you need archiving that satisfies a securities regulator, buy something built for that. WAOversight is not FINRA or SEC-grade archiving and we will not describe it as though it were. Smarsh and its peers exist for that requirement.

And if you have 3 reps in 1 room, you probably already know the answer. The measurement starts paying for itself somewhere around the point where you can no longer see everyone.

What to do this week

Pick 1 number and link it. Watch the median first response for 5 working days and resist changing anything.

You are looking for the shape of the week, not a verdict on a person. Most teams find the median is tolerable and the tail is not, which is a different problem from the one they thought they had.

For the same question answered rep by rep, with the 3 ways to get the numbers side by side, see are my reps replying on WhatsApp. Then read the response tracking detail for how the clock is defined, and the metrics guide for the thresholds. If you want to see it against your own numbers first, the demo runs on sample data.