Working with multi-channel retail organizations, we’ve experienced the frequent concern that online is competing with, or “cannibalizing” retail sales. It seems like a reasonable problem for those responsible for the P&L of the retail business to consider, same for the general managers responsible for the store level P&L.
I like to do something that we “digital natives” (professionals whose career has only been digitally driven) miss all too often. We talk to retail people and customers in the stores, store managers, general managers, sales and service staff. Imagine that… left-brain dominant Data Athletes that want to talk to people! Actually, a true Data Athlete will always engage the stakeholders to inform their analysis with tacit knowledge.
Every time we do this, we learn something about the customer that we quite frankly could not have gleaned from website analytics, transactional data, or third party data alone. We learn about how different kinds of customers engage with the product and their experience in an environment that to this day is far more immersive than we can create online. It’s nothing short of fascinating for the left-brainers. Moreover, access and connection with the field interaction does something powerful when we turn back to mining the data mass that grows daily. It creates context that inspires better analysis and greater performance.
This best practice may seem obvious but is missed so often. It is just too easy to get “sucked into the data” first for a right brain dominant analyst. The same thing happens in an online only environment. I can’t count how many times I’ve sat with and coached truly brilliant web analysts inside of organization who are talking through a data backed hypothesis they are working through from web analytics data, observing and measuring behaviors and drawing inferences… and they haven’t looked at the specific screens and treatments on the website or mobile app where those experiences are happening. They are disconnected from the consumer experience. If you look in your organization, odds are you’ll find examples of this kind of disconnect.
So Does The Web Compete with Retail Stores? Well that depends.
While many businesses are seeing the same shift to digital consumption and engagement, especially on mobile devices. The evidence is clear that it would be a mistake to assume that you have a definitive answer. In fact, it is virtually always a nuanced answer, that informs strategy and can help better focus your investments in online, and omni-channel marketing approaches.
In order to answer this question you need a singular view of a Customer. Sounds easy, I know. So here’s the first test if you are ready to answer that question:
How many customers do you have?
If you don’t know with precision, you’re not ready to determine if the web is competing, or “cannibalizing” retail sales.
More often than not what you’ll hear is the number of transactions, the number of visitors (from web analytics) or the number of email addresses or postal addresses on file –or some other “proxy” that’s considered relevant.
The challenge is these proxy values for Customer count belie a greater challenge, without a well thought data blending approach that converts transaction files into an actionable customer profile, we can’t begin to tell who bought, and how many times.
Once we have this covered, we’re now able to begin constructing metrics and developing counts of orders by customer, over time periods.
Summarization is Key
If you want to act on the data, you’ll likely need to develop a summarization routine –that is, that does the breakout of order counts and order values. This isn’t trivial, leaving this step out creates a material amount of work slicing the data.
A few good examples of how you would summarize the data to answer the question, by channel includes runs the totals by channel:
- by month
- by quarter
- by year
- last year
- prior quarter
- by customer lifetime
- and many more
Here’s The Key Takeaway, It’s not just one or the other.
Your customers buy across multiple channels. Across many brands and many datasets, we’ve always seen different pictures of the breakout between and across online and retail store transactions.
But you’re actually measuring the overlap, and should focus your analysis on that overlap population. To go further, you’ll require summarization “snapshots” of the data so you can determine if the channel preference has changed over time.
The Bottom Line
While no one can say that the web does, or doesn’t definitively “cannibalize sales” the evidence is overwhelming that buyers want to use the channel that is best for them for the specific product or service, at the time that works for them.
This being the case, it is almost inevitable that you will see Omni-Channel behaviors when your data is prepared and organized effectively to begin to see that shift in behavior.
Often times, that shift can effectively equate to buyers spending more across channels, as specific products may sell better in person –it’s hard to feel the silky qualities of a cashmere scarf online, but you might reorder razor blades only online.
The analysis should hardly stop at channel shift and channel preference. Layering in promotion consumption can tell you how a buyer waits for the promotion online, or is more likely to buy “full price” in a retail store –we’ve seen both of these frequently, but not always ― every data set is different.
Start by creating the most actionable customer file you can, integrating the transactions, behavioral, and lifestyle data, and the depth that you can understand how customers choose between the channels you deliver becomes increasingly rich, and actionable. Most of all –remember, it’s better to shift the sale to an alternative channel the customer prefers, than to lose it to a competitor who did a better job.