FlowsBook a callSign inCreate free accountSign up
Workflows by industry

How mystery shopping firms plan a week of store visits across their shoppers

A week's stores, shared fairly between shoppers and put in visiting order. Mystery shopping route planning that takes minutes on Sunday instead of an afternoon of maps.

Part of the guide Workflow automation

By the xMap team · · 4 min read

Mystery shopping route planning comes down to two jobs done in the right order: first share the week's stores between your shoppers so each gets a compact patch of the city, then put each shopper's stores in the order that wastes the least travel. Do it that way round and nobody spends half a day crossing town for one visit, every store is covered, and the plan can be sent out on Sunday evening with directions for each stop.

This is for project managers and schedulers at mystery shopping, retail audit and customer experience agencies. You will find the logic of a good visit week, a worked example with made-up stores, the traps that make shoppers drop jobs, and an AI employee that produces the week's plan from your client list.

Share first, then order

It is tempting to give each shopper a list and let them sort out the order. The trouble is that a list chosen by store name or by client usually scatters across the city. Sharing by area first fixes that:

  1. Put every store due this week in one table, with its address and the client brief.
  2. Split them between the shoppers you have, by area, so each gets about the same number of stores close together.
  3. For each shopper, order their stores to make one efficient loop.
  4. Send each shopper only their own stops, with the brief for each visit.

The Split by territory block does step two. It makes compact areas of about equal size and never gives one person more than their share: with 25 stores and 3 shoppers, nobody gets more than 9. Each store comes out with the shopper's name in a field called owner. The Route planner then reads that field and makes one route for each shopper.

What a week looks like in practice

Here is an invented week for an agency in Madrid. A telecom client wants 30 of its shops visited and a coffee chain wants 12 of its cafes checked at lunchtime. The agency has four shoppers available, named Inés, Pablo, Lucía and Diego, all working on foot in the centre and the inner districts.

ShopperVisitsWalking timeArea
Inés113 h 10Centre
Pablo102 h 40North
Lucía113 h 30South-east
Diego102 h 55West

Made-up stores and times, for illustration only.

Each shopper's email lists their visits in order, with the distance and time from the stop before and a directions link. The lunchtime cafe checks still need to fall between 12:00 and 14:00, so the scheduler moves those to the middle of each day by hand. A route planner orders stops for the shortest trip; it does not know the client's time windows.

Where visit weeks go wrong

  • The same shopper at the same store every month. Staff start to recognise them. Territories shift a little from week to week as the store list changes, which helps, but keep a rule in the brief: no shopper visits the same store two cycles running.
  • Overloaded days. Travel time is not visit time. A thorough audit with a purchase and a return can take 40 minutes, so cap the stores per day, not just the kilometres.
  • Stores without a usable address. A store that cannot be placed is listed after the routes without a visit order. Check the client's store file before the week starts.
  • Sending the whole plan to everyone. Shoppers only need their own stops. A shared master list invites swaps that break the routes.

Telling clients where visits happened

Clients often want proof that the right stores were covered. Because every stop carries its shopper, order and position, the same run can produce a map of the week's visits and a simple PDF summary for the account manager. That is useful in quarterly reviews, when a client asks why one region had fewer visits than another.

Have an AI employee plan the week

In xMap Flows the whole Sunday task becomes one AI employee:

  1. Schedule: every Sunday at 18:00.
  2. My data: this week's store list, uploaded from the client file or kept in a Google Sheets file.
  3. Split by territory: between the shoppers you name.
  4. Route planner: one route for each shopper, by the owner field, walking or by car.
  5. xMap LLM: a short brief for each visit from the client's notes.
  6. Email: one message per stop, sent to the shopper's address from the table, or a row per stop in a Google Sheets file each shopper can filter to their name.
  7. Show on a map for the account manager.

On Sunday evening the scheduler checks a map and four routes, makes the time-window changes, and the week is planned. The merchandiser visit plan for the week template is the closest ready-made version, and store visit route for a due diligence trip shows the same idea for one person. The bigger picture of field teams and their AI helpers is in the AI employee for field operations.

Questions

Can the plan respect store opening hours or client time windows?

Not by itself. The route planner orders stops for the shortest trip. Note fixed times in your store list and adjust those visits after the plan is made.

What happens when a shopper drops out midweek?

Remove them from the names in Split by territory and run the flow again. Their stores are shared out between the others and routed afresh.

How many stores can one shopper's route hold?

Up to 50 stops in a route. That is more than most shoppers can do in a week, so cap it lower to match your visit length.

Does each shopper need an account?

No. They only receive their route by email or see their rows in a shared sheet.

Templates that do this

Read next

More on workflows by industry · Every article · Templates

Put it to work

An AI employee in xMap Flows does this on live data, on its own timetable. Start from a template or describe it in plain words.