For a certain brand marketer of frozen entrees, it used to be that no matter how the company loaded the cases at the beginning of a promotion week, retail stores would be out of stock on its best sellers before the weekend.
Sales were being missed, shoppers disappointed. If ever there was a crying need for real-time information, this was it.
In a pilot program with a Midwestern retail partner earlier this year, the frozen entree maker started looking at daily store data each night. It forced out pallets loaded with precisely the right items to replenish each store at the midpoints of promotional events.
The result: out-of-stocks dropped 12% from the prior year. The retailer's frozen food category grew more than 7% compared with market growth of 4%, and the manufacturer enjoyed a net sales increase of 19%.
This kind of success story is becoming commonplace as more supermarkets are linked into real-time networks, which gather "census-quality" -- that is, store-by-store -- sales data for entire chains or markets on a daily basis.
Brand marketers and brokers describe such data as "tactical" and "actionable," meaning that subscribers can access sales and movement information on a particular item in a particular store on the next business day, in some cases early the next morning. Decisions can be made while a promotion is still in progress.
Currently, real-time information is available via two distinct systems: one offered by Catalina Information Resources, a unit of Information Resources Inc., Chicago, and the other from Efficient Market Services, Deerfield, Ill.
"As of Jan. 10, we have 2,663 stores on line," says Robert Billings, general manager of CIR. IRI recently acquired controlling interest in the company, which was originally a joint venture and Catalina Marketing, Corp., St. Petersburg, Fla.
He explains, "We are attached to the point-of-sale loop. Catalina Marketing continues to provide daily data from its Checkout Coupon system. It views and collects our data. It sees everything that crosses every scanner in the store. The client tells us what items to track in a particular chain.
"Every night at 11 p.m. the data is transferred from each store to a central processing facility in Chicago, where it is processed and cleaned in our mainframe computer. The data is accessible at 5 a.m. the next day. Our data turnaround time is six hours," Billings said.
Using IRI Data Server proprietary software, clients can access movement and pricing records via modem to their own data base, she said.
Penny Baron, executive vice president of strategic business development at EMS, says her company's system connects an electronic "platform" (a small computer) to each store's checkout system, which captures the sales of every item each day. EMS also tracks features by ad zone and displays by store each week and blends this information in the in-store platform, which then calculates and analyzes unit sales, base or normal sales, price, display, feature and expected sales.
The information is transmitted from the in-store platform to the EMS information center, which cleans and processes the data, transforming it into information that meets the user's requirements within 24 hours, said Mike Spindler, executive vice president of sales and client services at EMS.
EMS information can be accessed through Quick Response Mailbox, a service also marketed by Nielsen North America, which is a strategic partner with EMS. Quick Response Mailbox offers more than 9,000 different reporting capabilities to manage new product distribution, store servicing, assortment, shelf management and promotion planning, Baron said.
The CIR and EMS systems are almost equally positioned in terms of in-store installations and seem to be staying neck and neck, says Carlene A. Thissen, president of Retail Systems Consulting, Naples, Fla.
"It really does matter how many stores you have on line. Data is collected at night and the manufacturer has it by early the next morning. Brand managers need it that quickly to evaluate their overall promotion strategy or new product performance," Thissen says.
"Scanning data was strategic, but real-time information is more tactical. The manufacturer can see where it is out of stock in a particular store, call the broker and have more product delivered. If a real hot promotion is not proving as effective as it should be, the brand manager can look at competitors' pricing and react immediately. Access to real-time information from more stores is really significant," she adds.
Here is what some brand marketers and brokers told Brand Marketing about their use of real-time information:
Phil Benvenuto
retail information manager
Coors South Central Field Business Area
Menard, Texas
We are working with EMS. One of the real advantages of real-time information is the fact that you can monitor your progress a lot quicker. You don't have to wait four weeks. You can see how it did yesterday in terms of distribution and sales.
We also use daily data for analyzing promotions and displays and we track pricing with it. The real advantage is that it is actionable information. We are not doing promotion analysis a week after the fact. With daily data you can pull up information about a promotion in progress and still have three to four days to make changes if necessary.
We did a promotion analysis with one of our major supermarket customers. We found that we did not have displays of our featured package in all of the stores. We were able to effect a change and transformed a mediocre execution into a real success.
We have had success marrying EMS data with our own demographic information to target the stores most likely to sell a new product. The information is quick and accurate.
John Bird
category manager
Frito-Lay northern California area
Pleasanton, Calif.
We use real-time information for category management and growing the entire category at retail, and also internally to develop selling stories as a way to help grow sales for ourselves. Down the line we may be able to tie it into our delivery system, but that may be pretty far along.
The neat thing is it is really actionable. We also have InfoScan but we don't get it for four weeks. It is hard to evaluate a promotion four weeks later. With daily data you can see right away that "Wow! That was a great promotion," or "That was not a good promotion." For every dollar you spend, it is a lot more actionable, plus you get every single store. The data is very clean, which is also nice.
Michael Munch
category development manager
Nestle Frozen, Refrigerated and Ice Cream Cos.Chicago
We get information the next day for every store of a major Chicago chain. We use it to analyze promotions. We can see if the sale is truly effective while it is still in progress. That is especially important now, with ECR. Retailers want to focus on getting exactly the right product to the right store at the right time. Now we can put out what each store sells to maximize promotions and increase sales by the consumer.
We have seen increased sales. We can satisfy more consumers with fewer out-of-stocks. When a retailer runs a promotion, the biggest complaint is usually out-of-stocks. The daily data works almost like demographic information. It gives us an understanding of how we sell vs. our competitor and how much space our product deserves.
We want to know what is the optimum price where we can get the most customers into the stores and still maintain the retailer's margins. Daily data has helped us understand the promotion. If we can sell the item for 2 cents more per package and get the same amount of customers, why not? We have to find the break-even point.
Stephen Yoder
chain account manager, Tennessee
Miller Brewing Co.
Atlanta market area
If we run a feature with an account for one or two weeks, the daily data from EMS allows me to track the number of displays we had in the stores. In the past we had to rely on the retailer for that information, and it was not always totally accurate, due to various factors. With daily data I can see how many units were sold, what the lift was, and if we outsold our competitors. The daily data we get from EMS allowed us to get in the door with the Southeast division of a major national chain and become the category captain for 1995, in effect controlling the beer sets in those stores. Our access to real-time information will also help us assist the retailer in determining the benefits of a more impactful beer display with fewer cases.
Daily data has also been helpful during a regional rollout, as we recently did with Red Dog beer. We can pull up a store listing, see what package a store has and spot any distribution voids. We can track displays, where they are in the store, if they are primary or secondary.
Bob Gaedke
director of marketing
Carey, Ahrens & Raynsford (a broker)
San Ramone, Calif.
We were the first broker with CIR. [In tests] we have been picking mainly seasonal businesses that have a short window in which to make mistakes.
Our objectives are to really do some effective store tailoring, and find out which stores are responsible for our business. That type of information cannot be found in the syndicated data.
Normally the trade is ahead a minimum of six weeks, often eight to 12 weeks setting up promotions. If you are dealing in a seasonal business with a 20-week window, you cannot afford to wait five weeks because then there is nothing you can do about it. We needed something that would allow us to say what was happening in a promotion and what was happening on store level, such as which stores were most responsible for results.
Daily data also helps us determine price points. For example, a can of tuna was priced 3 cents higher at one chain vs. another. We were able to show the chain with the higher price what that 3-cent differential meant to its sales vs. the competition. After two months of consideration, the chain lowered its price and saw the sales of that tuna increase.
We also do post-promotion analysis with daily data.
Doug Morrison
marketing manager
PMI (a broker)
Lombard, Ill.
I see the progression of daily data and census information becoming very valuable to the broker and manufacturer. I think within the next five years all the information will become census rather than sample stores. We will be working with information about all stores. I don't know about the daily data, but because of the progress in technology, it seems that daily data would be the next step by default.
Real-time information is terrific for store door delivered products. If you are an ice cream vendor and drop product at the stores every day, you can use daily data to find out how much you need on your load on a daily basis.
