How to prioritise field sales visits: which customer should your team see first?

Ask the commercial leader of a company with a field sales team a simple question: how did your reps decide which customers to visit last week? In many organisations the honest answer is some mix of habit, geography, personal relationships and whoever called back. Nobody is doing a bad job. There is just no shared logic behind the calendar.

This article looks at why that happens in larger organisations, what it costs, and how a company can bring priority to the visit calendar step by step, without a big system and without taking away the autonomy that good reps need.

Short answer: To prioritise field sales visits, rank accounts on two things: how well they fit what you sell, and whether there is a reason to visit now. Visit priority works when a small group qualifies and ranks accounts using a few agreed signals, hands the result to reps as a suggested agenda, and feeds the outcome of each visit back into the ranking. Start manually, prove that it changes results, and automate only after that.

The problem leadership recognises but cannot see

The pattern shows up in food and beverage, building materials, technical wholesale, B2B services and consumer brands with an account base: a sizeable group of reps on the road, a CRM, and visits that follow each rep’s own judgement. Many visits are cold. Some go to customers who were never likely to buy more. Some go to a customer who was already about to switch supplier, and the visit comes too late.

For the CEO, CCO or CRO the difficulty is not that the reps are lazy. It is that there is no view. You can see how many visits happened, maybe, if the CRM was filled in. You cannot see whether those were the visits that mattered. So you cannot steer, and you cannot coach on anything but effort.

Published research points the same way. In a case described in Harvard Business Review, account managers at an enterprise software company spent about a third of their week with customers, and 40% of that customer time went to lower-tier accounts, even though the strategy was to grow the top tiers [1]. The company only saw this once it looked at where time actually went, not at what people reported.

What it costs

The cost shows up in three places, and none of them is on a dashboard today.

• Wasted capacity. A visit is one of the most expensive touchpoints a company has. When a sizeable share of visits adds nothing, which is what several teams find when they look back honestly at their own calendars, that is capacity you already pay for and do not use.
• Missed moments. Customers who are changing their ordering pattern, evaluating a competitor or ordering less often do not announce it. The visit that would have saved the account never gets planned.
• A blind forecast. If the pipeline is built from visits nobody can compare, the forecast rests on each rep’s feeling. Leadership ends up correcting numbers instead of improving them.

Why it stays this way

Four things usually keep the situation in place.

There is no qualification step. Leads and accounts go straight from a list or a campaign to a rep. Nobody asks whether this account is worth a visit this month, and by whom.

The CRM is not trusted. If reps do not log visits or outcomes, there is no data to rank anything with. And reps do not log what nobody uses. It is a loop that feeds itself.

Visit data is not linked to customer data. Visits are registered against locations or outlets, orders sit in another system, and contacts in a third. Matching them is partly manual, so most analyses stop before they start.

Autonomy is protected, rightly. Good reps know their territory. A system that dictates the agenda gets ignored within weeks. Any solution has to respect that.

How it can work instead

The route that works is gradual. Each step proves its value before the next one is built.

1. Qualify manually first. A small team, sometimes only a few people, reviews accounts and leads against a short list of criteria and builds the suggested agenda for the reps. This feels low-tech, and that is the point. It shows within weeks whether the criteria predict useful visits, and it costs far less than building a tool on an untested assumption.

2. Pilot with reps who want it. Start with a group of early adopters. Give them a suggested agenda and ask them to follow it for a fixed period, while keeping room for their own choices. Compare outcomes with the rest of the team. The results do the persuading, not a mandate.

3. Turn the criteria into a score. Once you know which signals matter, put them in a simple model. Two questions usually carry most of the weight: how well does this account fit what you want to sell, and is there a reason to visit now? Reasons can be a change in order frequency, an inbound request, a contract renewal, content the account engaged with, or a competitor moving in. The grid works the same way as a lead scoring model for marketing, with one axis for fit and one for timing.

4. Automate the plumbing. Link visit data, order data and contacts, so the score refreshes by itself and the agenda is suggested inside the tool reps already use. Only now does automation pay off, because you know what you are automating.

5. Close the loop. Every visit ends with a short outcome. That outcome goes back into the ranking, so the model learns from what reps see on the ground.

What leadership sees when it works

The main change is that questions which could not be answered before now can be, without asking each rep separately.

• Visit quality: what share of visits led to a next step, an order or a saved account, per segment, per rep and per signal.
• Coverage of what matters: how many of the accounts at risk or with high potential were actually visited in time.
• Adherence and override: how often reps follow the suggestion, and how often they choose something else. Overrides are not a problem. They are information about what the model misses.
• Forecast accuracy: whether the pipeline built from prioritised visits is more predictable than the one before.

These are few numbers, and they give the CCO and CRO something to coach on other than activity.

Where to start

Pick one team and one question: which accounts should this team visit in the next four weeks, and why? Write down five criteria you would use if you had the time to look at every account. Ask two or three people to apply them by hand to a subset of accounts, hand the result to a few willing reps, and record what happens after each visit. In a month you will know whether the criteria hold, where the data is missing, and what is worth automating. That is a far better starting point than a large project built on a guess.

If you want to look at how your field team plans visits today, a short audit of the calendar, the CRM data and the link with orders usually shows quickly where the biggest gain sits.

Frequently asked questions

How do you prioritise field sales visits? Rank accounts on fit (does this account match what you want to sell?) and timing (is there a reason to visit now?). Start with a short list of criteria applied by hand, then turn them into a score once you know which ones predict useful visits.

Who should decide which customers a rep visits? A small qualification team or sales operations should suggest the agenda. The rep keeps the final say and can override the suggestion. Overrides are useful feedback on what the criteria miss.

What data do you need to prioritise visits? At minimum: customer and account data, order history or frequency, recent visits with their outcome, and open leads or requests. Often the hardest part is linking these sources, not the ranking itself.

What should leadership measure? The share of visits that lead to a next step, an order or a saved account, coverage of at-risk and high-potential accounts, how often reps follow or override the suggestion, and forecast accuracy.

Does this work for B2C as well as B2B? Yes. Any company with reps who visit outlets, retailers, dealers or business customers has the same question. The signals differ, the logic does not.

Do you need new software first? No. A spreadsheet and a small team are enough to test whether the criteria work. Automate once you know what to automate.

Sources

[1] Mark Kovac and Jonathan Frick, “It’s 10 AM. Do You Know What Your Sales Reps Are Doing?”, Harvard Business Review, 10 March 2017 (reprinted by Bain & Company). A single company example, not a general benchmark.