Skip to content

How Many Stores Do You Use for a Pilot Project?

As we begin discussing an Alert Tech pilot project, one of the questions we’re frequently asked is, “How many stores do you use for a pilot project?” We don’t have a fixed number, because there is a different answer for each retailer, and that’s something we work though with you. This post will help begin that conversation.

The Purpose of a Pilot Project

If you’re considering a pilot project with Alert Tech, then you’ve begun to realize the value of Alert’s technology, but you want to test how well it works with your brand before you begin a large scale deployment. You’d like to answer questions such as:

  • Will it work with our level of service?
  • Will it work with our existing processes?
  • Are there additional factors we can’t simulate or predict that could affect how well this will work for us?

In order to answer those questions, we need to define what it means for a pilot to “work.”

Begin by Defining Success

Before starting a pilot project, we want to be clear on what success looks like for your brand. If – like most retailers – you’re not currently collecting data like this in your stores, you may not have a good idea of your starting point. So let’s see where you are first and then adjust accordingly.

Start With a Baseline

Based on industry standards, we could set one preliminary goal of, say, driving 30% of door traffic to the fitting rooms. When we begin a pilot, we’ll have an actual baseline for your stores, and at this point we might realize that 30% is not a realistic number for you. Once we know where your stores are performing, we can set realistic goals together, and then see if we can achieve those targets during the pilot.

Decide Which Stores to Use for a Pilot

Obviously, as a part of a pilot you want to see how well your brand can assimilate new technology and processes. But when you’re looking at driving ROI, one of the biggest markers of success is how far you can move the needle from your starting baseline. Let’s keep this in mind as we evaluate which stores to use for a pilot.

Testing at Top Performing Stores

You probably have a great set of stores that you like to use as your “test” stores – they have excellent managers and staff who are ready and willing to try something new and then give very good candid feedback. These “A” level stores have good processes and really deliver on your brand promise.

It can also be very interesting to assess how top performing stores respond when we look at various metrics. In some cases, we’ll find that an “A” level store is “sandbagging it”.. excelling in many areas, but with room to improve in one area. Fortunately, top players often have the ability to make a quick adjustment. With many of our clients, we see a very direct correlation between this adjustment and an increase in sales.

Busy-Store-EntranceOverall, these “A” stores are great places to start with a pilot, but are also limiting. The problem is that not every store in your chain is operating at this level. Most, of course, are not.

Testing at Average Stores

Depending on your brand focus, you may measure average (what we’ll call “B” level) in terms of different sales or service metrics, including:

  • Sales
  • Net Promoter Score
  • Refer to Friend
  • etc

If you would like to only run a pilot at only one type of store, you’ll get the best approximation of your chain-wide ROI at stores that perform closer to average.

Testing at Other Stores

Of course, you will also have “C” level stores. These stores may be a “C” for a variety of reasons: location, team, particular assortment, time of year, or some other reason. These stores may not be ones you would typically use to test something new, but there is one reason you want to consider including these in your pilot – there’s room for serious growth at this level.

Your “C” stores may have management who have had a taste of greatness and are hungry to improve. In addition, these stores have the most room to improve from the baseline – with the right tweaks, a “C” can become a “B” or even an “A.”

Other Considerations

When you decide to roll out new technology across your entire brand, you’ll do so either during new construction or as a part of a remodel. These two types of installations have significant differences – there’s a different energy and a different type of team in place – so it’s good to have some of each type of store in a pilot project.

Selecting a Group of Stores For a Pilot

We recommend pilot projects that contain “A,” “B,” and “C” stores. This will enable you to answer a number of questions:

  • Can you turn “B” stores into “A” stores?
  • Can you turn “C” stores into “B” or even “A”  stores?
  • Can you further increase the performance of “A” stores?
  • How much can you improve above the baseline at each level?
  • What would that look like across the entire brand?

We recommend a cohort with more than one “A,” “B,” and “C” store. This mix of stores will provide the clearest idea of the results you would get if you were to roll out Alert Tech technology across your entire chain.

Due to budget constraints, pilot projects will rarely contain statistically significant sample size, but multiple stores will provide you with the best data to be able to make predictions and to decide whether to roll out the technology full scale. Ideally, there will be about 3 of each type of store in the pilot, with at least one or two new stores vs remodels.


How many stores do you use for a pilot project? There’s no one-size-fits-all answer, but there is a right answer for your brand and budget, and we’re happy to help you find it.


Retail Analytics: How to Collect Retail Data, and What to Do With It

[xyz-ihs snippet=”TOFU3—QM1-CTA”]

Recent Blogs