DICK’s Sporting Goods ran an in-store staffing test with MarketDial to find how much added labor pays for itself. Of three staffing models tested against matched control stores, a high increase in engagement specialists lifted sales, conversion, and customer satisfaction enough to offset its added labor cost.
What problems did Dick’s Sporting Goods need to solve?
DICK’s needed to find the store staffing level that balances customer engagement with store profitability. Labor costs also had to fit operating expenses. To find that balance, DICK’s set up several staffing scenarios and measured each on three outcomes:
- Sales
- Conversion (the share of store visitors who buy)
- Customer satisfaction
The goal behind all three was a store where every customer feels welcomed, engaged, listened to, and that DICK’s was servicing their needs.
The test was only possible because DICK’s already considers testing as part of how it makes decisions. A change as expensive as adding labor gets tested in a sample of stores before it reaches the whole fleet.
Why does in-store customer experience matter?
Customer experience ranks just behind price and product quality in how people decide what to buy, according to PwC’s Future of Customer Experience survey. Dick’s used that research as a starting point for the test.
How did DICK’s Sporting Goods design the staffing test?
DICK’s tested three levels of added engagement staffing against a control group, with the test groups chosen by a cross-functional team.
They tested three different models:
- High level of increased staffing focused on customer engagement
- Moderate level of increased staffing focused on customer engagement
- Minor level of increased staffing focused on engagement
- Control Group
DICK’s built the in-store engagement test with a large cross-functional group so every interested party had a say. Bringing together central operations, strategy and analytics, athlete insights, finance, and loss prevention to decide the best use of labor, what was worth testing, and which models the data supported.
How were test and control stores selected?
MarketDial picked a representative mix of treatment stores and matched each one to a control store with similar historical sales patterns.
Adding labor to all ~850 stores to see what happened would have been a waste of time and money. DICK’s started instead with a smaller pool of stores that already met a baseline staffing level, and MarketDial selected treatment stores from that pool so every store reflected the wider fleet as a whole.
The matched control stores are what allow a lift to be credited to the staffing change rather than to the store itself. Once the design was set, DICK’s put the plan in place in stores, and the team watched the data come in to see whether each model drove a significant lift.
What were the results of the staffing test?
Model A, a high increase in engagement specialists, was the clear winner. It lifted sales and conversion enough to offset the added labor cost.
- The test reached statistical significance when DICK’s expected it to, which is hard to do in store testing.
- Two of the three treatment groups significantly improved the athlete experience.
- Only one of those two, Model A, was profitable.
Statistical significance here means the team could be confident the lift came from the staffing change, not from random noise.
Did more store staff improve customer satisfaction?
Yes, the test produced a statistically significant lift in customer satisfaction scores. DICK’s customer satisfaction data feeds straight into MarketDial, so the team can see whether any test moves its core customer KPIs.
Satisfaction scores are usually too noisy for a confident read. Few customers fill out the surveys, and results swing from week to week. In this test, the lift was large enough to reach statistical significance regardless of the noise.
To understand why more engagement specialists led to more sales, DICK’s paired the data with feedback from the floor:
- Calls with store teams
- Shop-alongs and secret shopper visits
- In-person store observation
Why did the DICK’s staffing test succeed?
The test worked because of three things:
- Good test design. Every team with a stake shaped the test groups; treatment stores represented the fleet, and each one had a matched control.
- Execution in stores. Test stores staffed to plan and carried it out as designed.
- The right length and store count. The test ran long enough, in enough stores, to reach significance on schedule.
DICK’s came away knowing which staffing level pays for itself, backed by sales, conversion, and customer satisfaction data.
Frequently asked questions?
What is an in-store staffing test?
An in-store staffing test changes labor levels in a sample of stores and compares results to similar stores that keep current staffing levels. It is done to see whether added labor lifts sales and conversion enough to cover its cost.
Why didn’t DICK’s test add staff in all of its stores?
Adding labor to all 850 DICK’s Sporting Goods stores would have been expensive and more difficult to interpret. A representative sample with matched control stores answers the same question at a fraction of the cost. MarketDial helps retailers run these tests to save time and money rolling out new ideas.
What is a matched control store?
A matched control store is a store with historical sales patterns similar to a test store, but with no changes made. Comparing the two isolates the effect of the change being tested. Retail test-and-learn uses this method to determine what changes are statistically significant.
Can a retail test measure customer satisfaction?
Yes, when satisfaction data is connected to MarketDial’s testing platform. DICK’s feeds its customer satisfaction scores into MarketDial, and this test produced a statistically significant lift in satisfaction.
What is MarketDial?
MarketDial is a retail test-and-learn platform. Retailers use it for predictive retail analytics by picking test and control stores, measuring the lift from a change, and deciding whether to roll it out. MarketDial saves retailers
