Is sales per labor hour a good KPI? Why SPLH can hide understaffing

What sales per labor hour measures

Sales per labor hour is one of the most common store productivity numbers, and for good reason: it needs only two inputs every retailer already has.

It answers a narrow question well: how much revenue did each hour of paid labor sit next to? It is a fair way to compare a store with itself over time, to spot a schedule that is clearly heavier than the business needs, and to sanity-check a labor budget.

What it cannot answer is the question most store and district managers actually use it for: was the store staffed right? For that, SPLH has a structural blind spot.

Why SPLH can rise when a store is understaffed

SPLH only sees sales that happened. It never sees the sales that did not happen because no one was available to help, or because the checkout line was long enough that people left.

Rogelio Oliva, a co-author of the traffic-based labor planning research below, made the same point when he argued that tying labor to sales is a weak method: sales reflect what customers bought, not what they might have bought with staff assistance (HR Dive, 2017).

Follow what happens in a busy hour when the store runs short:

1. Labor hours fall immediately. Two people on break, or pulled to receiving, means fewer hours on the floor.
2. Sales fall by less, and with a delay. Some shoppers wait. Some leave. The ones who leave are not recorded anywhere.
3. The ratio improves. Sales divided by fewer hours goes up, so the hour looks more productive on exactly the day it served shoppers worse.

The research on store traffic explains why this matters at peak:

• Perdikaki, Kesavan and Swaminathan (2012) found that conversion falls as store traffic rises, and that labor moderates the effect of traffic on sales.
• Mani, Kesavan and Swaminathan (2015) used hourly traffic, sales and labor data from 41 stores of a large retail chain and found that all 41 stores were systematically understaffed during a 3-hour peak period, and that eliminating that understaffing could significantly increase sales and profitability.
• In a field implementation described by Knowledge at Wharton (2019), Marshall Fisher, Santiago Gallino and Serguei Netessine reported that right-sizing staffing in 168 stores of one retailer over six months produced a 4.5% revenue increase and a nearly $7.4 million annual profit increase, after paying for the added labor.

If peak understaffing is common and SPLH rewards it, a store team managed to an SPLH target has a steady incentive to run lean in the hours that matter most.

A worked example

This is a hypothetical store, built to show the method. It is not customer data.

A specialty store looks at its Saturday peak, 12:00 to 15:00, on two Saturdays with the same traffic: 300 shoppers enter in those three hours. Average transaction value is $60 both days.

On Saturday B, sales per labor hour went up 11%. Sales went down $900, or 17%.

Now put both sides in money. Six fewer labor hours at a hypothetical loaded cost of $18 an hour saved $108. At a hypothetical 45% gross margin, $900 of lost sales cost about $405 of gross profit. A weekly SPLH report would show Saturday B as the better day. It was the more expensive one.

You can run this comparison on your own stores in a few minutes: pick two days with similar traffic, split them by hour, and line up labor hours on the floor, conversion and SPLH side by side. Our free Retail Staffing and Checkout Calculator (Google Sheets) does the hourly coverage math from a door count and a schedule.

What is a good sales per labor hour?

There is no universal benchmark worth managing to. SPLH moves with:

• Basket size and price level. A cashier who scans five items does the same work whether the basket is $10 or $100.
• Wage rates and labor mix. Markets and roles with different pay produce different ratios for the same service.
• Format. High-touch formats (furniture, eyewear, beauty consultations) need more labor per sale than self-service formats.
• Non-selling work. Replenishment, receiving and online order picking add hours that do not touch the sales floor.

A better definition: a good SPLH is the one a store sustains while conversion and service hold up in its busiest hours. Find it from your own data, store by store, and compare stores only within the same format and region.

What to read next to SPLH

Keep SPLH. Stop using it alone to judge staffing. Put these four measures beside it, by hour:

1. Conversion by hour (transactions ÷ entries). If SPLH rises while conversion falls in the same hours, the store is short, not productive.
2. Shoppers per associate on the floor (estimated shoppers present ÷ associates actually on the sales floor). See the customer-to-staff ratio guide for the method.
3. Minutes the checkout line ran above target. Long lines are where unrecorded walkaways come from. See how to measure checkout wait time.
4. Scheduled versus deployed labor. Who was on the rota, and who was actually on the floor in the busy hour. Breaks, deliveries and online order picking landing at peak are the most common gaps.

A simple weekly review: list each store's three highest-SPLH hours of the week and check conversion and line length in those same hours. Any hour that is high on SPLH and low on conversion is a staffing problem hiding as a productivity win.

How to set labor from traffic instead

The research above points to a different starting point: plan labor against expected traffic, then check it against outcomes.

• Chuang, Oliva and Perdikaki (2016) developed a traffic-based labor planning approach for retail stores. In Oliva's summary of the work, stores attained between 85% and 95% of their sales potential with their existing staffing, and tying labor to traffic could raise that to 99% (HR Dive, 2017).
• The Wharton team's three-step method is practical for any chain with historical data: estimate how revenue varies with staffing at each store from past traffic, sales and planned versus actual staffing (unplanned absences are useful natural experiments); validate the estimate by changing staffing in a few test stores; then roll out and measure.

Neither approach requires abandoning SPLH. They change which number leads the decision.

Common mistakes

• Setting a daily SPLH target. It invites cutting peak hours to rescue the day's number.
• Using hours scheduled instead of hours worked. The difference is the gap you are trying to find.
• Comparing SPLH across formats or regions. Different baskets and wages make the comparison meaningless.
• Reading a rising SPLH as good news by default. Check conversion and line length in the same hours first.

How Super Vision approaches this

Super Vision turns the cameras a store already has into an operational record of what happened on the floor. For this problem, that means seeing the hours when visitor traffic rose but service coverage did not: when checkout lines formed and how long they stayed above target, and staff presence at registers and service points where cameras cover the zone. Paired with POS data, that context explains why a store's sales per labor hour moved, not just that it moved.

It does not score individual associates, and it does not infer purchases from video. Revenue and transaction metrics come from your POS. Start with one location and a defined question, such as whether Saturday peaks are covered.