Managing a Field Sales Team With a Mobile CRM

Ask an owner with six field reps what happened yesterday and you get a version of the same answer everywhere: "Sales were okay. Ahmed had a good day in Madina Town. I think Bilal is struggling."

Ask what specifically happened — which shops were visited, what the owner of each said, which of last week's promising leads have gone quiet — and the answer stops. Not because the owner is careless, but because the information does not exist anywhere. It was said out loud in a shop and then it was gone.

The three things a field manager cannot see

Where reps actually went. The day-book says twelve shops. Nobody can confirm the twelve, and everybody knows nobody can. This corrodes the whole system, because the honest rep who visited eighteen looks the same on paper as the one who visited six and went home.

What was really said. "Interested" covers a shopkeeper who asked the price and a shopkeeper who said "acha, dekhtay hain" to end the conversation politely. Two months later nobody can tell which was which, and the next rep to walk in has no idea what was promised.

Which leads are going cold. This is the expensive one. A lead that never gets a second visit does not appear anywhere as a problem. It is not a complaint, not a loss, not a line in a report — it is an absence, and absences are invisible until somebody counts them deliberately.

The order book will not tell you any of this, and neither will the monthly sales figure. Both only show outcomes, long after the point where a manager could have changed them.

Why day-books and WhatsApp groups fail

Every field team in Pakistan starts with one or both. They fail in predictable ways.

The paper day-book is written at the end of the day, from memory, usually in the twenty minutes before the rep goes home. It is a summary of a summary. It is not searchable, it cannot be filtered by "shops with no visit in thirty days", and it lives in the rep's bag — so when the rep leaves, the territory knowledge leaves with them. That last point is the one owners feel hardest.

The WhatsApp group is worse in a specific way: it feels like it is working. Reps post updates, the manager sees activity, everyone is reassured. But a WhatsApp group is a stream, not a record. Yesterday's updates are already fifty messages up. There is no way to ask "what is the status of Al-Madina Store" without scrolling for ten minutes. Nothing has a next-action date attached. And the updates are unstructured, so one rep writes three lines per shop and another writes "done".

Both systems share the same fatal property: they record activity and not state. What a manager needs is state — which leads are live, which need a visit this week, which have gone cold — and neither a notebook nor a chat stream can produce it.

Spreadsheets are the next attempt and they fail differently; we have written about where the spreadsheet stops working separately.

Assigning territory lists per rep

The first thing a mobile CRM changes is that a rep opens their phone in the morning and sees a list of who to visit, which they did not have to compile.

Two rules make this work.

Assign geographically. A rep with 30 outlets spread across a city visits maybe 12, because the rest of the day is on a motorbike. A rep with 30 outlets in four adjacent streets visits all 30. Split the territory into contiguous areas and give each rep one — never split by outlet size or alphabetically.

Keep the boundaries stable. At least a quarter, preferably longer. Field selling in this market runs on relationships, and a rep who has visited the same shopkeeper six times has an asset that gets thrown away every time you reshuffle territories.

With an organisation account, each rep signs in on their own licence while credits stay pooled at the company level. That matters for attribution: every visit, note and status change is stamped with the person who did it, so a territory has a history that survives the rep leaving. New leads collected from a Maps search go straight into the pipeline and can be assigned to whoever owns that area, rather than arriving as a CSV someone has to split by hand.

GPS punch-in as proof of visit

The rep arrives at the shop, opens the outlet in the app, and taps to record the visit. The app checks the phone is within about 300 metres of the mapped location before it accepts. You get a timestamped, located record that a person was at that shop.

That is the whole mechanism, and it does a specific and limited job: it removes the argument about whether a visit happened.

Why 300 metres and not 50

This is the detail that decides whether the feature works in practice, so it is worth explaining rather than just stating.

Phone GPS is far less precise than the moving blue dot suggests. Outdoors with a clear sky, 5 to 15 metres is realistic. Inside a shop, under a metal roof, in a narrow street with buildings on both sides, in a covered market — which describes most of the places a field rep in Pakistan actually stands — accuracy routinely degrades to 50 to 100 metres, and sometimes worse. Phones fall back to mobile tower and Wi-Fi triangulation when satellites are obscured, and those are much coarser.

Set the radius at 50 metres and you get a stream of false rejections. The rep is standing at the counter of the correct shop, taps punch-in, and is told they are not there. They try again. It fails again. They conclude the app is broken, and within a week they have stopped using it — or they have found a workaround, which is worse, because now you have a record you trust and shouldn't.

Three hundred metres is loose enough to accept every genuine visit and tight enough to be useless for faking one. Nobody punches in from home, from the office, or from a tea stall in a different area. That was the actual concern; the tighter radius does not catch more cheating, it only catches honest reps.

What it does not prove

Be clear about this internally, because overselling it causes problems.

A punch-in proves presence. It does not prove a conversation happened, that the owner was in, or that anything useful was said. Pair it with a required outcome on every visit — ordered, no requirement, owner absent, refused, revisit — so the record carries content and not just a coordinate. Two seconds of tapping per visit, and the manager's report becomes readable.

Offline-first is not a nice-to-have

A CRM that needs a connection to save something is a CRM that field reps stop using.

The places field selling happens are the places signal is worst: inside a covered wholesale market, a basement showroom, an industrial estate on the edge of the city, a shop with three floors of concrete above it. A rep who loses a visit note because the sync failed will not write the second one.

The Android app holds the assigned list on the device and records visits, notes, statuses and punch-ins locally, syncing when signal returns. From the rep's side nothing appears to happen — the app just works, which is the point. It also means the day does not depend on how much mobile data is left on the rep's package, which for teams paying reps a fixed allowance is not a small consideration.

Tap-to-call and tap-to-WhatsApp matter for the same reason: any step that requires retyping a phone number is a step reps skip.

The daily report the owner should actually read

Most CRM dashboards drown a manager in charts. A field manager needs three things, and reading them should take four minutes.

Visits made, per rep. Count and locations. This is where you see that one rep does 14 a day and another does 31, which is a coaching conversation and not an accusation — the 14 might be a spread-out territory, and now you can check.

Leads that moved. Which shops changed status yesterday, and to what. New to contacted, contacted to interested, interested to won or dead. Movement is the leading indicator; sales are the trailing one. A day with 30 visits and zero status changes usually means the visits were courtesy calls.

Tomorrow's plan. Which leads have a next-touch date of tomorrow, for each rep. If a rep's tomorrow is empty, they will improvise, and improvised days are the ones where nothing moves.

What to leave out of the daily read: revenue charts, conversion percentages, anything comparing months. Look at those monthly. Daily, you want the three questions above, because those are the only ones you can still do something about.

Private notes and per-rep performance

Two more things worth having.

Private notes on a lead — visible to the owner, not to other reps. This is where the useful, awkward context goes. "Owner's brother handles purchasing, deal with him." "Paid late twice last year." "Do not send Bilal, they had a disagreement." Reps will not write this in a shared field and they will not write it at all if they think the shopkeeper might somehow see it.

Per-rep performance over a month, not a day. One bad day is noise. What is worth reading is the pattern: visits per day, the ratio of visits to status changes, and the share of a rep's assigned list they have touched at all in thirty days. That last figure is often the revealing one — a rep can be busy every day and still be visiting the same comfortable forty shops out of a hundred and twenty.

The lead CRM keeps all of this attached to the lead rather than to a report, which is the difference that matters: when someone opens Al-Madina Store in two years, the whole history is there, including who visited and what they wrote.

The trust question, honestly

We should be straight about this, because vendors usually are not.

Location tracking is intrusive. Being asked to prove where you are standing is not a neutral experience, and a rep who feels monitored rather than supported will work worse, not better. Anyone selling this as an unambiguous good is not being honest about it.

Three positions we hold, and which are built into how the product works:

Visit-time punch-in, not continuous tracking. The app records a location when the rep chooses to record a visit. It is not following them through the day, not recording where they had lunch, and not building a movement trail. That distinction is the whole difference between proof of work and surveillance, and it is worth insisting on whichever tool you use.

Explain it before you switch it on. Tell the team what is recorded, when it is recorded, who can see it, and what it will and will not be used for. Reps who understand that punch-in exists to make coverage countable — and to protect the honest rep from being lumped in with the one who does six visits a day — mostly accept it. Reps who discover it by accident never do, and you have lost more than the feature.

It is optional and admin-toggleable. Some teams run with it, some do not, some enable it only for particular territories or for a period while a coverage problem is being fixed. It is not compulsory for the rest of the CRM to work, and we would rather a team used the visit records without punch-in than not use the tool at all.

The practical test we would apply: could you show a rep exactly what their manager sees about them, and would they think it was fair? If yes, you are on the right side of it. If you would rather they did not know, that is the answer.

Legal obligations to employees around monitoring vary and we are not lawyers — if you have doubts about your position as an employer, take advice.

Where to start

Do not roll this out to twelve reps at once. Take one rep and one territory for a month.

Assign their list, turn on punch-in after explaining it, and read the three-line daily report every morning. At the end of the month you will know two things: whether the visit count matches what the day-book was claiming, and how many assigned outlets never got visited at all. In our experience the second number surprises people more than the first.

Then expand. The free trial includes 25 leads, which is enough to build one small territory and see whether the reporting tells you anything you did not already know. Start there, or see how the pieces fit together.