The Smallest Thing You Can Verify

A field officer's day ends the same way most days end in Lagos sales: a form, filled in on a phone with eleven percent battery, listing everywhere he went. Twelve visits. Nine conversations. Three follow-ups scheduled. It reads like a good day. Whether it was a good day depends on a question the form never asks, which is what any of that visiting actually produced.

I ran a field team that grew past fifty people and touched more than two thousand physical locations across Lagos: pharmacies, retail counters, gyms, workplaces. The single hardest thing about that job was never recruiting or training or routing, although all three took real work. It was resisting the pull of activity as a proxy for output, because activity is so much easier to count. A visit either happened or it didn't. A call either connected or it didn't. Kilometres travelled show up on a tracker without anyone lifting a finger. Output — a patient who actually booked and attended using the partner code, a pharmacy that actually stocked and moved an item, a referral that actually closed — takes real definitional work to even measure, and most organisations quietly give up on that work and count the easy thing instead.

The failure mode this produces is familiar enough that I'd guess most people reading this have lived through some version of it, in sales or elsewhere. Activity reports come in looking strong. The dashboard is full of green. Commercial results stay flat, or grow at a rate that doesn't match the reported effort. Management, seeing effort and flat revenue side by side, concludes the team isn't trying hard enough, and responds with more pressure and more forms. The team, rational actors responding to what's actually measured, learns to produce more of what's counted — more logged visits, tighter-sounding notes, cleaner-looking calls — without any corresponding change in what those visits produce. Nobody lied. The system just asked for the wrong thing and got exactly that.

What the business can actually verify

The fix starts with a definition most teams skip because it sounds almost too simple to be strategy: decide, in writing, what a unit of completed work looks like at street level, in a form that can be checked against a record somewhere other than the field officer's own report.

"Visited hospital" is not that. It's a claim, and an honest one most of the time, but it can't be checked against anything. "Patient booked and attended using the partner code" is that. It shows up in a booking system. It has a timestamp and an outcome independent of what the field officer says happened. The difference between those two sentences is the entire difference between a metrics system that tells you the truth and one that tells you what your team believes will please you.

This sounds obvious once it's stated and is genuinely difficult to hold onto under pressure, because the verifiable unit is almost always a slower, harder-won thing to produce than the activity that precedes it, and a manager staring at a quarter that isn't going well will always feel the pull toward counting the thing that's easy to make go up.

Five stages, and where each one breaks

The useful move, once you've defined the unit, is to break the distance between an officer walking into a pharmacy and a naira of value being created into stages, because the stages tell you something the aggregate number never will: where in the chain things are actually breaking.

Access is whether the officer reached the person who could actually say yes: not the security guard, not a junior staff member with no authority, the pharmacy owner or the gym manager or whoever holds the decision. Conversation is whether, having reached that person, the need and the offer and the eligibility criteria were actually understood on both sides — stated is easy to log, understood is the harder and more important thing to check. Commitment is whether a specific next action got agreed with a date attached to it (an order, a referral, a slot booked) as opposed to a friendly conversation that trailed off into "let me think about it." Completion is whether that committed action actually happened: the booking was kept, the order was placed, the referral showed up. Value is whether the completed action produced revenue or the proof the business actually needed.

Run a struggling territory through those five stages and the diagnosis becomes almost mechanical, in a way a single blended number never allows. Low access, everything downstream is starved regardless of how good your officers are at conversation, and the fix is territory design or introductions, not more training. High conversation paired with low commitment means the pitch is landing but the offer itself, or the training behind it, isn't closing — that's a script and incentive problem, not an access problem. High commitment paired with low completion is the most maddening failure mode of all, because the field team did its job and the business still didn't get paid, and the fault sits somewhere in the back office — a booking system that drops appointments, a stock-out on the exact item that was promised, a referral that arrives at the clinic and nobody follows up on. Blaming the field officer for that gap is unfair, and worse, it's a wasted diagnosis: you'll retrain someone whose work was already sound and leave the actual leak untouched.

A Tuesday route and why it falls apart

None of this holds up without a route that respects what Lagos traffic actually is, which is not a scheduling inconvenience but a structural constraint on how many genuine conversations one person can physically have in a day.

Picture a Tuesday route built the tidy way, on a map, with eight stops spread across a spread of neighbourhoods because each one looked winnable on paper. By ten in the morning the officer is stuck on a corridor that seemed fine in planning and is not fine in practice, and every stop after the second is now a rushed five minutes instead of the fifteen a real conversation needs. The fix isn't a better app. It's grouping visits by corridor rather than by target list, building in slack for the rain that will close a road or the diversion that will eat forty minutes without warning, and being honest in advance about which relationships genuinely need a face in the room and which ones can be sustained by a phone call or a WhatsApp message instead. A route that assumes ideal conditions will be wrong by nine every morning, and an officer improvising against a broken plan produces exactly the kind of activity that looks fine on a form and means nothing.

What you pay for is what you get

The last piece is compensation, because whatever you reward is what a rational field officer will optimise for, full stop, regardless of what the job description says.

Pay purely on completed revenue and officers learn quickly to avoid the accounts that take longer to close (the larger pharmacy chain that needs three conversations before it signs, the workplace programme that has to clear a procurement process) in favour of quick, small, easy wins that hit the number faster. That's a rational response to the incentive, and it will quietly starve your pipeline of exactly the accounts that matter most a year out. Pay purely on visits and you get noise: activity for its own sake, disconnected from anything the business needed. The workable middle is a staged model that pays something real at commitment, more at completion, and the largest share at value. Enough at each earlier stage that a long-cycle account still feels worth pursuing, weighted enough toward the end that nobody gets rich stacking meetings that go nowhere.

What coaching actually looks like once you can see the stages

Managers under pressure default to a single lever, which is more pressure — a stern meeting, a tighter deadline, a reminder that targets exist. The five-stage breakdown is worth the trouble precisely because it replaces that single lever with several different, specific ones, and specific levers are the only kind that actually move a struggling officer's numbers rather than just his stress level.

An officer weak on access usually isn't lacking effort; he's lacking the right introduction, and no amount of pushing him to visit more locations fixes a territory where he's structurally stuck talking to gatekeepers. The fix there is pairing him with someone who already has the relationship, or restructuring the territory so he isn't cold-calling his way into every account. An officer strong on access and conversation but weak on commitment is usually working from an offer or a script that doesn't close, which is a training problem and sometimes a product problem, and no amount of coaching on his people skills will fix an offer that genuinely doesn't answer the question a pharmacy owner is actually asking. And an officer who's landing commitments that don't complete needs, more than anything, to be defended from blame while someone fixes the actual leak upstream — because burning out your best closer by holding him responsible for a stock-out he had no control over is how you lose the one person on the team who was actually doing the hard part of the job well.

This is also where the five stages earn their keep at the portfolio level, above and beyond the individual one. Look across fifty officers and the stage-by-stage breakdown tells you something a blended activity number never can — whether the whole team is stalling at the same point, which usually means the problem isn't the people at all. Fifty officers all weak on completion is not a training failure distributed across fifty individuals. It's one back-office failure wearing fifty different names on a report.

Back to the form

That officer filling in his form at the end of a Tuesday with eleven percent battery left is responding, precisely and rationally, to whatever the form asks him to prove. Ask him to prove he moved. Count kilometres and visits, and you'll get kilometres and visits, in impressive quantities, forever, alongside a commercial result that never quite catches up. Ask him to prove something happened that a system somewhere else can verify (a booking made and kept, a code redeemed, a shelf actually restocked) and the form stops being a performance and starts being evidence. The whole discipline of running a field team at scale in this city comes down to that one design choice, made early, and mostly invisible to everyone except the officer who has to live inside whatever you decided to count.

Notes on sources

• Informal employment at 93% and self-employment at roughly 85.6% of the employed population, Q2 2024: Nigeria Labour Force Survey, Q2 2024, National Bureau of Statistics.
• Broadband and internet subscriber figures underpinning why field reporting tools need to tolerate patchy connectivity: NCC Industry Statistics — Internet Subscriber Data, Nigerian Communications Commission, June 2026 (approx. 156.9 million total internet subscribers).