Beat Planning Software: How Top Distributors Actually Design Weekly Routes

Last month I sat with a distributor in Sharjah who was running 14 reps across 1,200 outlets. His weekly beat plan was a printed Excel sheet from 2019. Literally the same routes, same days, same sequence — for four years.

And his secondary sales were flat. Shocking, right?

Here's the thing about beat planning. Most distributors treat it like a one-time setup. You map your outlets, you assign your reps, you print the schedule, you forget about it. Then you wonder why coverage drops, why your A-class outlets get visited every 11 days instead of every 7, and why your top rep keeps "forgetting" to visit that one grocery in Al Quoz.

The distributors who actually grow — the ones doing 18-24% YoY in categories that the rest of the market says are saturated — they treat their beat plan like a living document. It changes monthly. Sometimes weekly. And almost always, software is doing the heavy lifting.

What a good weekly beat actually looks like

Let me describe what I see at the better operations we work with.

A rep starts the day at 8:47 AM (not 9, because the first outlet opens at 8:30 and the second one's owner takes tea at 9:15). The route is sequenced not alphabetically, not by zone code, but by drive time between outlets factoring in actual traffic on that day of the week. Tuesday morning traffic on Sheikh Zayed Road is not Wednesday morning traffic. Good beat planning software knows this.

The rep visits 32 outlets that day. Of those, 19 are A-class (visited weekly), 9 are B-class (visited fortnightly, and today happens to be their week), and 4 are C-class catch-ups. Each outlet has a target time window — 8 to 14 minutes depending on order complexity. The rep's phone tells him if he's running long.

Lunch break is at the 17th call, near a spot he actually likes, because the system learned his pattern and stopped fighting it.

By 5:30 PM he's done. Not because he ran out of time. Because the route was designed to finish on time.

Compare that to the Excel-and-WhatsApp approach where a rep does 22 visits on Monday, 41 on Tuesday (because his manager called him out), then 14 on Wednesday because he's exhausted. Coverage is theatre. Numbers look fine on the monthly report. The shelf tells a different story.

The five inputs that actually matter

I've watched a lot of teams build beat plans. The ones that work pay attention to five things, in this order:

Outlet class and visit frequency. Not every shop deserves a weekly visit. An A-class outlet doing AED 8,000/month in your category gets 4 visits a month. A C-class doing AED 600 gets one. Most distributors I meet are over-servicing the bottom 40% of their outlets and under-servicing the top 15%. The math is brutal once you see it.

Geographic clustering. This sounds obvious. It isn't. I've seen reps zigzag across three emirates in a single day because the beat was built around outlet codes, not coordinates. Good sales route planning groups outlets into tight clusters where the rep is never driving more than 6-8 minutes between calls.

Day-of-week patterns. A wholesale market in Karachi behaves differently on Friday than Tuesday. Grocery owners in Riyadh have predictable prayer-time gaps. Hypermarket buyers in Manchester only see reps on specific mornings. Your beat needs to respect this, not fight it.

Rep skill and outlet match. Your senior rep should be on key accounts and negotiation-heavy outlets. Your junior shouldn't be cold-calling a regional chain buyer. Sounds basic. Half the distributors I talk to assign by postcode and hope for the best.

Productive time per call. Honestly, this is where most software falls short and where we spent a lot of time on Zivni. A 12-minute call should produce an order, a stock check, a competitor note, and ideally a planogram photo. If your average call is 4 minutes, the rep is order-taking. If it's 22 minutes, somebody's drinking too much chai.

Where I got it wrong

When we first built beat planning into Zivni, I thought the answer was full automation. Press a button, get the perfect route, send it to the rep. Done.

It didn't work. Or rather — the routes were mathematically optimal and operationally useless. The algorithm didn't know that the rep had a relationship with one shopkeeper who'd been buying from him for nine years and refused to deal with anyone else. It didn't know that one neighborhood in Lahore is genuinely unsafe after 4 PM. It didn't know that Tuesday is the day a particular chain's buyer is in town from Jeddah.

So we changed the model. The software proposes. The supervisor disposes. The rep flags. Over six to eight weeks, the system learns the local truth and the routes get sharper. After three months, the beat plan is something neither a human nor an algorithm could have built alone.

That's the unglamorous version. But it's how it actually works in production with the 200+ distributors using us now.

A small test you can run this week

Pull last month's GPS data for your top three reps. Look at two numbers: average distance between consecutive outlets, and average time between consecutive outlets. Then look at planned versus actual visits per day.

If consecutive outlets are more than 2 km apart on average in an urban market, your beat is clustered badly. If actual visits are below 80% of planned, your day is too aggressive or your sequence is broken. If they're above 110%, you're probably double-counting drive-bys as visits — which is its own problem.

This isn't a software pitch. You can do this in Excel if you have the GPS logs. Most teams don't, which is the real reason they end up looking at beat planning software in the first place.

The distributors who win their categories aren't smarter than everyone else. They just look at their beat plan more often than once a year. And they're willing to change it when the numbers say so.

Which brings me back to that distributor in Sharjah. Is he still on the 2019 Excel sheet? What do you think.