Deal Retail Traffic and Retail Transactions differently

Sales performance is relative. How does anyone know what good or great performance is without comparison to what was possible? You can’t. But every day, retailers ask, “How are sales today?” How are sales compared to what? The answer: opportunity.

Putting traffic and customer conversion together enables us to understand performance relative to opportunity. The chart below shows ten stores ranked from highest traffic to lowest, with the corresponding conversion rate for that store indicated by the dot above the bar corresponding to the conversion rate indicated on the secondary vertical axis on the right. Again, the sales rank is embedded in the bar, so we have a clear view of opportunity, performance versus the opportunity, and sales rank.

One of the first things you notice is that traffic and conversion rate tend to be inversely related. The higher-traffic stores tend to have lower customer conversion rates, and the lower-traffic stores tend to have higher customer conversion rates.

If you think about it, it makes perfect sense. When stores get busy, customers don’t get served, line-ups get long, and prospects leave without making a purchase. In low-traffic stores, there’s a better chance for prospects to be successfully converted into buyers, and the conversion rates reflect this.

By representing the data in this way, you can understand what’s behind the sales ranks. Take the Philadelphia store, which ranked second in overall sales, despite having only the fourth most traffic. But it had the second-highest conversion rate, and it’s this combination of traffic and conversion rate that explains the high sales rank.

Now let’s compare results from sets of stores. As the chart below shows, Chicago ranked first in sales, but it didn’t have the most traffic; New York did, but it managed only a third-place sales rank. In fact, if we compare these two stores, we learn that Chicago generated 23% higher overall sales, but it achieved this with 9% less traffic. Chicago had a higher average ticket, which helped, but the bigger driver was the conversion rate — Chicago had a 19% higher conversion rate than New York.

Here’s another interesting set in the chart below. Philadelphia is ranked second in sales, while Houston ranks fourth. Philadelphia generated 5% more sales than Houston, but what’s remarkable is that it did so with 35% less traffic than Houston. Based on sales transaction data and sales results, these stores looked evenly matched.

But traffic and conversion completely change how we feel about the stores’ performance. Philadelphia does an excellent job with the traffic it gets; Houston is under-performing versus a bigger traffic opportunity. Without traffic and conversion data, this retailer could never understand these differences.

Each store in a chain is unique. Each store is located in a unique geography. It has different staff, different demographics of the customers in its immediate market, different competitors, different climate, and different micro-economic conditions.

In the final analysis it’s less important how stores compare to other stores as it is to how it performs versus itself; however, comparing stores does serve a purpose: it helps us more precisely target resource requirements and gain insight into what’s possible. How does Philadelphia deliver these results? What’s the staffing level? Who’s the manager? Is he/she doing something that other stores should be emulating?

Let’s look at another example of comparative store performance, first with sales results and transaction data, then again with traffic and conversion added into the mix.

The chart below shows a district of twelve stores from a different chain, ranked from highest average daily sales to lowest.

When we take the stores that had the highest and lowest average daily sales and compare these results to the number of sales transactions each produced, as in the chart below, we see that the transaction rank correlates to sales results fairly closely. The top three sales stores also had the top three transaction counts, and the lowest selling stores had the lowest number of transactions.

Given that so many retailers rely on transaction count data as a proxy for traffic, it’s worth exploring these data further.

While the sales results and transaction count results are directionally consistent, they don’t match perfectly. Washington ranked first in sales but trailed Boston in transaction counts by 64%. If we think of transaction counts as a proxy for traffic, we would conclude that Boston had 64% higher traffic than Washington and that Denver had 10% less.

At the other end of the scale, the bottom three stores’ transaction counts correlate very highly to overall sales results. The Reno store had the lowest number of transactions and the lowest sales rank. Using transactions as a proxy for traffic, we would conclude that Las Vegas had 12% more customer traffic than Reno.

Before we get too carried away with thinking that transactions are a good proxy for traffic, I want to dispel that idea right now: transactions are not a good proxy for traffic. The chart below shows average daily traffic compared to the average daily transaction count by store. As you can see, the variances between transaction counts and traffic counts are very significant.

Let’s look at the top three stores first. Boston had 64% more transactions than Washington, but it actually had 227% more prospect traffic. And on a transaction count basis, Denver had 10% fewer counts than Washington — but 60% more prospect traffic. At the low end of the scale, we see the same thing. Las Vegas didn’t have 12% more traffic than Reno, as the transaction count implied; it had 83% more prospect traffic.

These are not small variances. They’re significant differences that have wide-ranging consequences. They completely change how management needs to think about and manage performance in these stores.

Today, many retailers still use transactions as a proxy for traffic. These transaction counts are used for, among other things, scheduling staff. As the above examples illustrate, if you scheduled staff based on transaction counts instead of prospect traffic, you would be seriously off the mark, and you would miss sales opportunities that you didn’t even realize the stores had.