Measuring for Business Success
Tip for May 2008 - Volume 2
Checking if your strategy is working: Why POS Reports are not Enough
When you manage a restaurant, you regularly try new or different things to try and build sales, or to reverse a sales slump. For example, you could change some items on the menu. How will you know if that is accomplishing your objective?
You want to know if your actions have increased sales, or had no effect on sales at all, or even caused a decline in sales! In other words, you want to know: after I changed the menu, did sales go up? By how many percent? If the change continues, how much will sales of the new items be, and how much will total sales increase in the next 3 (or more) months?
If you have a chain of restaurants, you could test the change on a few restaurants, and compare their performance with the performance of the others that did not change their menus. That kind of “split-testing” is accurate, and relatively easy to do with appropriate reporting software.
But, how do you do this if you only have one restaurant? Let’s see why simply reviewing standard POS sales reports could cause you to reach the wrong conclusions.
Let’s say you use standard POS reports to find out. First, consider the “daily sales recap”. Let's say it's the end of Friday, and the report says you sold $2,000. Is that good or bad? How can you tell, unless you compare it to something else. But what should you compare it against?
Let's say you compare it against the previous day's sales, and that was $1,500. Is Friday’s sales of $2,000 good? You still won't know, because your Friday sales are expected to be higher than your Thursday sales, anyway. So, comparing against the previous day wasn't any good. In fact, for most restaurants, comparing one day against another (or looking at a daily sales trend chart) is useless, because for most restaurants, sales are always lower on Mondays, then build up to a climax on Saturdays.
To solve that, you would have to compare one day's sales against sales for the same day last week. That usually means you need to run a second report (for last Friday's sales) to do that.
But that's still not enough. What if the previous week was the end of the summer vacation? Then, sales may be up (or down) because business is affected by the seasons (e.g., people leaving or arriving for vacation affect sales). You have to look at still another report to find out what last year's sales at around this time were, so you can compare them against the current sales.
But then, the restaurant's daily sales are highly erratic. Maybe someone had a birthday, and brings a whole bunch of friends and family to the restaurant, and sales are up 20that night. Or maybe it was a stormy night, and very few people went out. A single day's results are too random to indicate whether the new menu has had an impact on sales. So, you have to run reports for a cumulative period (e.g., a week, a month) after the change in menu, and then compare that against the same period before the menu. And against the same period the year before (to remove seasonality). Still more reports.
Finally, you want to know the trend over a period of time, not just an average. For example, let's say that over the 5-week period after the menu change, the total weekly sales (adjusted for seasonality) were: $10,000; $11,000; $12,000; $13,000; $14,000. Compare that to a weekly sales trend of: $14,000; $13,000; $12,000; $11,000; $10,000. Those 2 opposite patterns produce the same total sales and average sales for the 5 weeks. But, wouldn't your conclusion about the effectiveness of the menu change be totally different?
All in all, you have to run maybe a dozen POS sales reports, then try to figure out what all the numbers indicated. That will likely be too tedious and too time-consuming for you. As the big public corporations have already realized, reports are great for financial controls. But they are not useful for strategic decision-making.
What to do? You either get business-intelligence software that digests the raw data into information for you, or you rely on gut-feel.
