Most distribution dashboards carry more tiles than anyone can act on. Sales against target, growth on last month, a map of pins, a leaderboard, a ring showing attendance. By the third week the field managers have stopped opening it.
The fault is rarely the software. The measures were chosen because they were easy to compute, not because a decision hangs on them. A number nobody can change by Thursday is a report; a KPI tells a named person to do a specific thing. There are fewer of those than most dashboards carry.
What follows is the set that earns its place on a Kenyan distributor's dashboard, and how to read each one when it moves. It is opinionated about the rest, because a decorative metric costs attention, not screen space.
What a Distribution Dashboard Is Supposed to Do
A dashboard has one job: to shorten the distance between something going wrong at a duka in Kisumu and somebody in Nairobi knowing while it can still be fixed. Everything else is history, and history belongs in a monthly pack.
That splits every measure in two. Outcome measures — value sold, cash collected — tell you the score, honestly and late. Behaviour measures — coverage, productive calls, strike rate, lines per call, range — tell you what the field did to produce it, and they move first. That is the only reason to watch them.
Coverage and the Productive Call
Coverage asks how much of the planned outlet universe was actually visited in the cycle. In Kenya, where most trade runs through dukas, kiosks, roadside stalls and open-air markets rather than chains, it says whether the business is present in its own market at all.
A productive call is narrower: a visit that produced an order. The gap between calls made and calls that sold something is where the honest conversation lives, because coverage without productivity is tourism with a phone.
Read together, they interpret themselves. Coverage rising while productive calls stay flat usually means routes have been padded with easy, nearby outlets. Productive calls rising while coverage falls means the team has retreated to the outlets that always buy, and the tail of the territory is being abandoned. Neither number survives a sloppy definition, so settle these first:
- What counts as a visit — a geo-verified arrival at the recorded outlet with an outcome captured on the spot, not a row typed up that evening.
- What counts as productive — an order line captured at the outlet. A stock check logged is real work, but it is not a productive call.
- Which outlets sit in the denominator — the active universe for the cycle, with closed shops and duplicates removed before the count.
- Who owns it — the field manager who can change a beat, not the analyst who assembles the pack.
Strike Rate: Whether the Call Turned Into an Order
Strike rate is orders divided by calls attempted, over a stated period for a stated group. Compute it for a rep, a route, a channel or a single line, and each cut answers a different question.
Its value is diagnostic. Falling while coverage holds steady, it says reps are still turning up and outlets have stopped buying — which has few likely causes. The distributor is out of the fast movers, a price has moved, a competitor has put a scheme on the shelf, or the team has lost interest. Each is checkable within a day.
Two disciplines keep it honest. State the denominator, because strike rate on planned calls and on all calls made are different numbers. And never compare across channels: a duka round in Nakuru, an agrovet route towards Naivasha and a wholesaler desk in Nairobi are three different jobs, and ranking them measures territory, not effort.
Lines Per Call and Range Selling
Lines per call counts the distinct products on the average order. It measures the depth of a transaction and checks the commonest failure in field selling: the rep who sells the one line the outlet always takes and moves on.
Range selling asks the same question across the portfolio — how much of what an outlet ought to stock is actually stocked, cycle after cycle, against a must-stock list for that class of outlet. Depth without breadth rests the business on two fast movers, comfortable until a competitor targets them.
Read movement carefully. Lines per call rising while order value stays flat can mean the mix has broadened, or that orders are being split to flatter the metric. Range failing in one channel while holding in another usually indicts the list, not the reps: a list written for a mini-mart and applied to a kiosk will fail forever.
- Duka and kiosk — a short list built around pack sizes the shopkeeper can afford to hold.
- Mini-mart and supermarket — broader and planogram-led, with sight of what the buyer already listed centrally.
- Agrovet — driven by season and crop cycle in that county, so the list is revised on a calendar.
- Wholesaler — measured on breadth held, because the onward buyer decides what moves.
Fill Rate: The Order You Could Not Serve
Fill rate is the proportion of what was ordered that was actually delivered. It can be computed on lines, units or value, and the three will not agree — value flatters, because one large line delivered in full hides several small ones that were cut. Line fill rate is the honest basis.
It is a supply measure, but the field pays for it. A run of cut orders teaches reps to stop offering the lines that keep failing, and a cycle later strike rate and range fall for reasons that look like field performance and are nothing of the sort. When strike rate drops, check fill rate before you check the rep.
One rule protects it: compute against the order as taken in the field, never the order after the office trimmed it to what was available. An edited order always fills perfectly, which is why it must not be the baseline.
Secondary Sales Against Primary Despatch
Primary sales are what you despatched to the distributor. Secondary sales are what the distributor sold onward into the trade. A business watching primary alone is measuring its own loading discipline and calling it market performance.
The gap is the measure, read over a rolling window rather than a single month, because month-end loading moves stock without moving demand. Primary ahead of secondary means inventory is accumulating in the distributor's warehouse, and presents later as returns, expiry or a distributor who cannot pay. Secondary ahead of primary means cover is being drawn down. Days of stock belongs beside both.
Kenya adds two wrinkles. Volume sold to a wholesaler is secondary sales but not retail offtake, so bulk-breaking throughput counted as coverage overstates shelf presence. And along the Northern Corridor, goods invoiced here can leave the market entirely towards Malaba and Uganda — a territory can post healthy secondary numbers while its own dukas run empty.
Outstanding and Days Sales Outstanding
Outstanding is the shilling value the trade owes you, aged into buckets. Days sales outstanding converts that balance into time: how long, on average, an invoice takes to become money in the bank. Both belong on the dashboard: a balance can look stable while the age inside it deteriorates.
Read the movement against sales. DSO rising while sales rise means you are financing your own growth, which should at least be a conscious choice. DSO rising while sales are flat is a collection problem or a customer in distress, and the ageing says which. The oldest bucket matters; the newest always looks reassuring. Credit control belongs upstream of both: a limit enforced at order capture is a control, the same limit reviewed monthly a description.
Two Kenyan realities distort this. M-Pesa is how much of the trade pays, and a till or paybill receipt not yet matched to an invoice looks exactly like an unpaid one — unallocated receipts inflate DSO and send collectors after customers who have already paid. And an invoice only starts its clock properly if it is a valid electronic tax invoice under the eTIMS regime KRA administers, because a customer who cannot claim the expense will hold the paperwork back.
A Monday Morning in Nakuru
A distributor's office off the A104 on the edge of Nakuru town, a little after seven. Reps check bags and phones before heading out — some into the town's dukas and kiosks, some towards the agrovets in trading centres along the Naivasha road. The area sales manager has a screen open on his phone, not a printed pack.
He asks three questions and the review is over in the time it takes to drink tea. Which outlets on last week's beat did not order, and does anyone know why. Which lines were cut on Friday's deliveries. Which accounts moved into the oldest ageing bucket, and who is collecting today. Each question has a name attached before the vehicles leave.
The version that achieves nothing is more common. The manager reads the sales-against-target tile aloud, observes the team is behind, and asks everyone to push harder. The tile was accurate and unactionable, because nobody could name the outlet, the line or the invoice behind it. One screen shows exceptions; the other shows a total.
The Measures That Are Decoration
Some numbers survive on dashboards for years because removing them feels like admitting the dashboard was wrong. Remove them anyway. The usual offenders:
- Sales against target, alone. It is the score, and never contains its own explanation, so a leader who sees only this can only ask people to try harder.
- Visits made, without productivity or geo-verification. Anything a rep can improve without selling is something a rep will improve without selling.
- App logins and active users. Useful during rollout, uninformative once the app is compulsory.
- Photo counts. They measure compliance with a rule about photos, not whether a shelf was corrected.
- Volume leaderboards across unlike territories. A wholesaler-heavy Nairobi route and a rural round towards Machakos rank in the same order regardless of who works harder.
- Attendance as a headline. A payroll input and a management exception, not a sales measure.
The blunt test: if a tile moved sharply tomorrow and nobody would do anything differently, delete it. A dashboard is a short argument about what matters this quarter, and every addition weakens it.
The Cadence That Makes People Act
Measures fail more often on timing than on definition. A weekly number delivered fortnightly is a historical document. A daily number sent to someone with no authority to change anything is a nuisance. Write down cadence, owner and the decision each report feeds.
- Daily, pushed not pulled — exceptions only. Outlets missed on yesterday's beat, orders cut, accounts blocked on credit, devices that have not synced.
- Weekly — behaviour. Coverage, productive calls, strike rate and lines per call by rep and route, each carrying a named decision: change the beat, retrain, re-sequence, escalate supply.
- Monthly — structure. Range against must-stock, secondary against primary with days of stock, DSO with ageing, fill rate by line.
- Quarterly — the measures themselves. Outlet master hygiene, must-stock lists, definitions and denominators. Nothing corrodes trust faster than a definition that changed silently mid-quarter.
One thing belongs on every screen and is almost always missing: data freshness. Field capture in Kenya has to survive patchy rural connectivity and the occasional power interruption, so the app works offline and syncs when a connection returns. That has a reporting consequence — a route that has not synced looks identical to one that sold nothing. Show the last sync time beside the number.
Where These Programmes Go Wrong
The commonest failure is measuring before the outlet master is trustworthy. Coverage computed against a list full of closed shops and duplicates is precisely wrong, and once a leadership team has argued about that number twice, the dashboard loses authority. Clean the universe first.
The second is letting the distributor self-report secondary sales in a spreadsheet. What arrives is a tidy version of the month, delivered after anyone could act on it. Secondary data has to be captured where the transaction happens — order booking, van sales, distributor billing — or it is narrative, not measurement.
The third is attaching incentives to gameable measures. Pay on visits and visits rise; pay on photos and photos rise; pay on order lines and orders get split. That is not dishonesty, it is people responding rationally to what you rewarded. Incentives belong on outcomes that are hard to fake — value sold, cash banked, range verified. Underneath all three failures sits one habit: building the dashboard for the leadership meeting rather than the field manager.
How 1Channel Supports This
1Channel is a sales force automation and distributor management platform. Its relevance here is that the numbers come from transactions captured at the point of work rather than typed into a return afterwards. Visits are geo-verified against the outlet record, orders are captured on mobile, and the applications work offline and sync when a connection returns.
The outlet master carries channel classification — duka, kiosk, mini-mart, supermarket, agrovet, wholesaler, petrol-station forecourt — with territories built on county, sub-county, ward, route or beat. That is what makes coverage, strike rate, lines per call and range against a must-stock list computable per channel rather than as one total. Secondary sales can be captured from order booking, van sales and distributor billing and reconciled against primary despatch and stock on hand, so the primary-to-secondary gap and days of stock are visible without a spreadsheet.
Fill rate can be measured against the order as taken. KES credit limits and ageing apply at order capture, and collections, including reconciliation of M-Pesa till and paybill receipts, run through the same ledger the DSO number is built from. Reporting is configurable — exception alerts, scheduled reports, dashboards by role, approvals routed through the reporting hierarchy. On invoicing, the platform supports the electronic tax invoice obligations KRA administers through eTIMS; 1Channel is not a certifying body and makes no certification claim.
Key Takeaways
A dashboard is not an inventory of everything the system can calculate. It is a short statement about what the business is trying to change this quarter, and what you left off decides its quality.
- Behaviour measures move before outcome measures. Coverage, productive calls, strike rate, lines per call and range tell you what the field did; sales and collections tell you the score too late to change it.
- Fix the definition and the denominator before you report the number. What counts as a visit, what counts as productive, which outlets sit in the universe — settle these in writing, then leave them alone.
- Check fill rate before you blame the rep. Cut orders teach reps to stop offering the lines that keep failing, and the damage surfaces a cycle later as a strike rate problem.
- Primary despatch is not market performance. Read secondary against primary over a rolling window with days of stock beside it, and remember that wholesaler throughput and cross-border movement are not retail offtake.
- Delete any measure nobody would act on. Bare visit counts, logins, photo counts and cross-territory leaderboards cost more in attention than they return in insight.
None of this needs a large programme to begin. Take one territory, clean its outlet universe, agree four definitions, and run a daily exception list and a weekly behaviour review for a quarter. The measures that survive are the ones worth building the rest of the dashboard around.
