A 90-Day Frontline Sales Improvement Plan for Retail Chains

What the plan is for, and what it is not

The plan is for a chain or dealer network where footfall is adequate and conversion, ticket or both sit below what the best stores in the same band achieve. It changes the floor: approach, demonstration, objection answers, the close, the number and the follow-up. It is not a pricing review, a range review or a marketing plan, and it will not rescue a store whose location or stock is the problem. The diagnosis that tells the two apart, before a rupee is spent, is at /learn/why-store-sales-are-down-when-footfall-is-not/.

It is also not a training programme with a launch date. Chains have run those, and the classroom day is remembered fondly and changes little, for the reasons at /learn/why-retail-sales-training-does-not-stick/. In this plan, training is one week of the twelve, and coaching by the store manager on what was actually said to customers is nine of them.

Before day one: the baseline

Nothing in the plan can be judged without two to three weeks of measurement before anything changes. The baseline needs, per pilot store and per advisor where possible: door count, conversations held, buyers, average bill, non-buyers whose number and date were taken, follow-ups sent, and sales recovered within 30 days. If the chain has no way to count conversations, the baseline is the door count, the bills and a manual tally of numbers taken, and the first fortnight of the plan is spent putting a measurement in place. The KPI set and how to read it is at /learn/frontline-sales-kpis-store-manager/.

The baseline also names the lever. A store with 60 percent coverage on weekends has a rota problem before it has a coaching problem. A store where a third of non-buyers leave without a number has the cheapest fix in retail waiting. A store with good coverage and numbers but low conversion on conversations has a pitch or objection problem, and the objection mix tells which.

The plan, week by week

Owners are named by role. "Manager" is the store manager; "Area" is the area or cluster manager who owns four to twelve stores; "HO" is the head office sponsor, usually the head of retail or sales. The number to read is the one the week's work should move; if it does not, the following week repeats rather than advances.

Choosing the pilot stores

1. Pick two bands, not two stores. A band is four to six stores of similar footfall and category mix, so that a store's result can be compared with its peers rather than with the network average.
2. Include a below-median and an at-median store in each band. The plan should show that it lifts the weak store and does no harm to the adequate one.
3. Keep two matched bands as controls that get the measurement and nothing else. Without a control, the festive week or a competitor's stock-out will be credited to the plan, or blamed on it.
4. Choose managers who will do the routine, not the stores with the best numbers. The plan tests whether manager-led coaching moves conversion; a manager who will not stand on the floor at peak tests nothing.
5. Tell the floor what is being measured and why, and get consent in the way the law and the chain's own policy require. Advisors who believe the measurement is for punishment record fewer conversations and the baseline is a fiction.

The four things that break these plans

• Everything at once. A new app, a new incentive, a new training module and new KPIs in the same fortnight, across all stores. Nobody can say what worked, managers are overwhelmed, and the floor waits for it to pass. One lever at a time, two bands at a time.
• The incentive first. Paying for numbers taken before anyone can count numbers taken produces fake numbers. Behaviour components go in at week 10, after seven weeks of measurement have made the behaviour visible and normal.
• Head office coaching. A trainer visiting each store once a month cannot change what is said on Tuesday. The store manager can, and the plan stands or falls on whether managers do the routine at /learn/store-manager-daily-routine-to-increase-sales/.
• The festive exception. The plan is paused for the festive weeks because the floor is busy, and never restarts. Peak weeks are when the approach rule and the number matter most; the rota guide at /learn/peak-hour-staffing-retail-stores/ covers how to hold the routine when the floor is full.

What to put in front of the board on day 90

Three numbers, pilot against control against baseline: conversion including recovery, average transaction value, and sales per employee. Behind them, the behaviour numbers that explain the change: coverage, numbers taken, follow-ups sent, attach rate. Then the roll-out order by band, the manager routine written as the standard, and the incentive change with its cost. A pilot that moved conversion by three to five points in the below-median stores and held ATV or raised it is a normal result for a plan that was actually done; a pilot that shows twelve points has a measurement problem or a control problem.

Chains that measure conversations rather than only bills can also show the board what changed on the floor, in the advisors' own words: the objection that stopped ending conversations, the close that started being asked. Borentis produces that reading from consented conversations, in Hindi, English or Hinglish, with the scorecard by store and advisor that the area manager coaches from; the follow-up is drafted for a person to send and nothing goes to a customer automatically. The plan above works with manual tallies and a manager who listens; software makes the week 12 comparison something a board can trust.