Monitor and Evaluate

Point-of-Sale Data

A cashier accepting a payment

You may want to use data from point-of-sale systems in cafeterias, snack bars, and other venues to evaluate implementation of your food service guidelines. Point-of-sale systems include hardware such as cash registers, touchscreen monitors with cash drawers, and credit card readers, and may include software that can record purchases made by consumers. If point-of-sale systems are programed to capture the required details of foods and drinks sold, the resulting data can provide the most direct evidence of changes in sales of healthy or less healthy food items.

Using Point-of-Sale Data for Food Service Guidelines Evaluation

When planning your evaluation, meet with food service managers to investigate the ways point-of-sale systems record point-of-sale data in each facility of interest. Based upon the available data, it may be possible to evaluate some specific food items even if it is not possible to measure all the food standards implemented.

Consumer sales data may be already captured at a subset of your facilities. Therefore, you may wish to use those data to represent all consumer sales data in your evaluation, complementing other evaluation data sources collected across a larger number of facilities. Alternatively, it may be possible to work with food service managers across several sites to develop a simple way to record a limited set of important healthy or less healthy food items. For example, food service managers may agree to accurately record healthy entrée sales on each day, healthy side items, or selected ala cart items such as fresh fruit or bottled water. The products selected should:

  • Reflect important food service guidelines requirements that have been implemented.
  • Be sold in sufficient sales volume that they have a meaningful impact on the healthfulness of foods purchased by customers.
  • Be reasonably easy for food service managers across facilities to program into their point-of-sale systems.
  • Be easy for staff to tell the difference between food products and record accurately when checking out customers.

Considerations When Using Point-of-Sale Data

When using point-of-sale data, consider the following issues affecting your evaluation.

  • Menu Cycles. Many cafeterias run on menu cycles where the specific entrées and sides follow a repeating pattern over one or more weeks. Since some food items may be more or less popular than others, total sales and/or sales of healthier or less healthy foods may vary on different days of the menu cycle. Therefore, menu cycles should be considered when collecting point-of-sale data on foods purchased. You should assess at least a full menu cycle, and if feasible, assess multiple cycles. Sales per menu cycle can be a meaningful and relevant time unit by which to measure sales.
  • Total Sales Volume. Total sales volume varies for most facilities during the year; therefore, when you compare the impact on food service guidelines on the sales of healthy entrées between your baseline (pre-intervention) and intervention periods, you will need to adjust for the total sales volume. This is because, in many workplaces, employees may take vacation during certain times of the year such as the weeks surrounding major holidays or school breaks. During these times, the volume of sales may decline. To account for variation in total sales volume, sales of particular food items of interest should always be adjusted according to the total sales volume of all items. An example formula is below. Note that adjusted sales are intended only to evaluate the effect of food service guidelines on the healthfulness of foods sold over time. Adjusted sales volume differs from actual sales volume and should not be used for tracking or ordering inventory or for assessing financial revenue.
  • Amount of data needed. In addition to holidays, other factors may affect sales in ways that cannot be anticipated or predicted. Therefore, it is important to ensure that data are collected over a sufficiently long period of time so that unusual rises or falls in sales on a given day do not overly influence the results of the evaluation. For example, data for your pre- and post-implementation of food service guidelines may be gathered over several months, with multiple full menu cycles represented.
Example Equation to Account for Total Sales Volume

Adjusted healthy entrée sales for a given menu cycle = (Actual sales volume of healthy entrées during menu cycle ÷ Total sales volume during menu cycle) X (Total sales volume for entire study ÷ Number of menu cycles in entire study)

Advantages of Point-of-Sale Data

  • Depending on the type and programming of point-of-sale systems, point-of-sale data can provide the most direct evidence of changes in sales of healthy and less healthy foods because they measure foods actually purchased by consumers.
  • Point-of-sale data includes prepared foods such as entrées, sides, and desserts.
  • Point-of- sale data can provide specific menu items and the date on which they were purchased.

Disadvantages of Point-of-Sale

  • Point-of-sale systems may not be able to differentiate between healthy and less healthy options. They may have the capability but may need to be reprogrammed and some food service managers may be unwilling to do this.
  • Point-of-sale data cannot usually be used to assess the healthfulness of salad bar items and fountain drinks. For example, salad bar items are usually purchased by weight, irrespective of whether consumers purchased lettuce or fried chicken strips.
  • Point-of-sale datasets may be very large and contain a number of different foods and beverages. This may require extensive coding to classify food items as healthy or less healthy.
  • Point-of-sale data from different facilities may vary substantially and complicate multi-site evaluation effort
  • Food service managers may not be familiar with procedures for exporting data in a usable format for evaluation.
Potential Ability of Point-of-Sale Systems to Provide Evaluation Data

Point-of-sale systems vary greatly across facilities in their ability to capture useful data for evaluation. Most modern ones, if properly programmed, can automatically produce highly detailed database records as transactions occur. However, older systems may not record electronic data and some new systems are not programmed to adequately capture the healthfulness of products purchased. You can categorize point-of-sale systems in four ways depending on the information that you are able to get from them.

Does Not Have Usable Sales Data

Some point-of-sale systems may be set up for attendants to simply enter the prices for items purchased without inputting the type of food. This could be due to inherent limitations of the equipment or due to the equipment not being adequately programmed. The point-of-sale system may not be capable of outputting any electronic sales data or the data produced does not identify names for specific foods and beverages. If this is the situation, you will need to use other data sources.

Has Sales Data but Cannot Determine Product Differences

Some point-of-sale systems may be equipped to capture sales data but are not programmed to capture it with enough specificity to identify product categories of interest. Or, the categorization scheme used by these type of point-of-sale systems may not sufficiently differentiate healthier or less healthy selections within a product category. For example, a system may capture that an entrée and side item were sold but not differentiate whether the entrée was a grilled chicken sandwich or bacon cheeseburger or whether the side was steamed vegetables or French fries.

While it may be possible to program some point-of-sale systems to capture data required for evaluation, such efforts may be burdensome to food service managers. This is because reprograming the system and retraining staff takes time. It may be faster and easier for staff to learn and use a point-of-sale system with fewer, broader categories than one with many categories that sufficiently differentiates sales of healthy and less healthy products.

Has Sales Data with Mixed Ability to Determine Product Differences

Some point-of-sale systems can appropriately identify some foods items so that you can classify them as meeting your food service guidelines. However, they may not be able to identify other foods, such as fountain drinks or salad bar items. This can make it impossible to use point-of-sale data for some menu items because you will not know if a sugar-sweetened beverage was purchased or a non-caloric soft drink was purchased, for example. Likewise, most point-of-sale systems will not record what was purchased from the salad bar, which could include healthier foods such as vegetables and unhealthier foods such as fried chicken strips. It should be noted that fountain drinks and salad bar items are consistently difficult food items to record accurately. For this reason, efforts to evaluate food service guidelines using point-of-sale data can be complemented by the procurement and/or production data for products that are not adequately captured at the point of sale.

Has Sales Data with Determination of Product Differences

Point-of-sale systems may be set up to capture data on healthy and less healthy food sales, especially if facilities are operated by large food service companies. However, if evaluation efforts involve multiple facilities, there are likely to be inconsistencies in point-of-sale data between facilities.

Error processing SSI file