October 24, 2007
CPGmatters: New Solutions Proposed to Reduce Out of Stocks
By Jack Grant
Through a special arrangement, what follows is an excerpt of a current article from CPGmatters, a monthly e-zine, presented here for discussion.
A set of recommendations for reducing retail out of stocks (OOS) is outlined in a new comprehensive study. The research grouped the root causes of OOS into three areas: data accuracy, measurement and shelf merchandising practices.
“Data accuracy must be addressed first, as it is the foundation for ordering and forecasting,” said Thomas Gruen, Ph.D. of the University of Colorado. “We found major problems with retailers’ product item data accuracy.”
Mr. Gruen conducted the research along with Dr. Daniel Corstein, IE Business School Madrid. It was funded by a grant from Procter & Gamble.
The study identified a variety of store issues that create perpetual inventory system inaccuracy.
“The level of PI (Perpetual Inventory) inaccuracy was stunning, as PI accuracy (where the PI exactly matched the on-hands) ranged from 32 to 45 percent in the four studies we conducted or examined,” said Mr. Gruen. “Improvements in PI accuracy cut OOS in half.”
The study found that manual audits and PI measurement of OOS levels are inaccurate, do not focus on the lost sales associated with an OOS item, and do not adequately point to solutions.
In shelf merchandising, the study found that most OOS sales losses are due to a relatively small number of items, and few of these items have adequate shelf space relative to their demand.
“Reworking planograms to account for the demand of the faster moving (and high OOS) items can reduce the level of OOS and the store labor necessary to continually restock these items,” said Mr. Gruen. “Second, basic retail practices that encourage three well-known links to OOS need to be enforced: one, don’t cover holes; two, don’t hide product, and three, shelf tag accuracy. We found that simple adherence to these practices had a huge effect on out of stocks, reducing OOS levels by about 40 percent.”
The study outlined a three-step approach to address OOS:
- Create Ranking: Create both a product ranking and a store ranking when assessing OOS; in other words, which products have the highest level of lost sales and which stores have the highest level of lost sales due to OOS.
- Target Reduction Desired: Determine the amount of OOS reduction desired and then allocate the amount of gain to be achieved from “product-based” OOS reductions (store or ordering types of solutions) and from “store-based” OOS reductions (shelf or operations types of solutions).
“For example,” the study states, “if a goal of $500,000 lost sales reduction has been established, determine how much of the goal will be obtained from ordering-type solutions such as PI data accuracy and how much should be gained from store-type solutions such as demand-based planograms. In this example, it may mean focusing on the top 300 products and the worst 24 stores to get them to their target rate.”
- Apply Solutions: Apply the solutions in the assessment to those products (across all stores) and those stores (across all products). This will provide an estimate of the resources needed to achieve the goal.
Discussion Questions: What do you think are the root causes of out of stocks (OOS)? What do you make of some of the recommendations in the study to reduce OOS? What solutions did they miss?
BrainTrust
Recent Discussions
Discussion Questions
Poll
While this topic and poll include some great ideas, I believe the “low hanging fruit” falls into the “Other” category.
The most common missing link between planograms and the rest of the retail organization is the communication of MIN and MAX data to replenishment systems.
Retailers have done a great job of building enough planograms that are localized to the right markets. These planograms include MIN data, the lowest number of units by SKU that is acceptable so you still look like you’re in business, and MAX data, which is typically the capacity of the SKU in each planogram.
Retailers win when they connect the MIN/MAX data from their planograms to their replenishment systems. The really good retailers do it many weeks in advance so they know when to fill up or sell down their warehouse inventories in advance of space allocation changes.
When the retailer gets this link connected the consumer and the manufacturer win along with the retailer.
To Kai’s point, many of the out-of-stocks visible to the customer may not be visible to the retailer. If the merchandise in question is in the stockroom but not on the shelf (or showing in the store’s inventory even though it may be en route from a distribution center), the company’s central inventory management will not have visibility to the problem. A lot of effective management of out-of-stocks comes down to good store managers “walking the floor,” plus appropriate use of payroll for restocking.
Good grief! Here we are, still talking about this problem and coming up with the same old solutions which have yet to solve anything. We all know there are very few SKUs which sell in large quantities, more than we would like not selling at all, and the vast majority somewhere in the middle.
I would suggest three things could be done with today’s systems and technology. First, identify the best sellers by store and make it easier for the order writer to know at a glance what they are. If there is any stock in the back room, it should only be these items (not counting seasonal and promotional goods) so restocking them must have priority since that is the only stuff there. Second, Get rid of the dogs. This sounds easily done, but in reality it is extremely difficult to accomplish. Operations people do not like the write-offs and staff people do not like to admit mistakes. Take your medicine and get rid of them. Third, the new found space can now be used for expansion of the best sellers or for the introduction of new high-potential items.
At the risk of over-simplification,those steps would go a long way to reduce store level/controllable out of stocks. As for store replenishment, that is another story.
These are great practices that all retailers should use, however, the article forgets one of the most common problems and that is simple data usage. Although many retailers have sophisticated inventory management and planograms in place, they still don’t properly mine and use the data that is available to them. This includes basic POS data cross compared to shelf stocking quantities, velocity tracking and comparisons across stores, regions and categories (many items have definite correlations like milk, eggs and butter). Furthermore, stores have access to backroom inventory systems that should be enforced every bit as much as the inventory in the front of the store. This is a huge deficit for most stores and accounts for one of the key reasons for poor shelf space and OOS in the front of the store despite inventory in the back.
There are basically two objectives in inventory management for the self distributing retailer. For the warehouse, the objective is to maintain a balanced time supply of products in order to minimize handling and transportation charges. In the store, the goal is to “keep the shelves full.”
The warehouse requires accurate forecasts for both demand and delivery schedules along with rational expectations of what can be received. Too often, I have seen self inflicted inventory problems at retailer warehouses by over optimistic receiving schedules. They are almost like the airlines, expecting to “land” all their deliveries at the same popular times.
Store inventory management begins with allocation of presentation stock or shelf inventory. To some extent manufacturers have contributed to challenges in this area by offering so many variations of their products that allocating sufficient space becomes impossible. But retailers located in highly seasonal areas must also be flexible in the shelf allocations and be willing to invest in the resets necessary to accommodate demand shifts.
I think when addressing store out of stocks, you must divide your items into three categories: those items whose normal sales between deliveries is less than the shelf inventory (they require no backroom inventory and represent the overwhelming majority of the “center store” items), those items whose normal sales between delivery exceed the shelf space but the shelf space quantity is enough to cover the time between stocking periods (they require rework from backroom inventory between deliveries), and items whose normal sales exceed the time between stocking periods (i.e. between night crew shifts) that require special handling to make sure the shelf area is in-stock.
It is this last category that provides the most headaches, because items are constantly shifting in and out of this category based on promotion schedules. Store personnel must be given a “heads up” on the items whose expected movement between stocking periods exceeds the shelf allocation. It would be great if these items could be highlighted for the point of sale systems and a notice issued each time the shelf area capacity is approached. That way store personnel could be proactive in making sure these items are kept in stock.
The out of stock prevention steps seem reasonable. But the financial impact may be overstated. Very often, shoppers simply buy a substitute when the desired item is out of stock. That substitute is often bought at the same store. So has the store lost money on every out of stock item? Yes, the shopper’s annoyance may undermine her loyalty, but measuring that financial impact is very different from measuring the financial impact of allegedly lost sales. Furthermore, there’s a point where being 100% in stock at all times can be financially unreasonable versus 99% or 98% or 95%. Getting that last few percentage points can be disproportionately expensive.
The first step in controlling out of stocks is to measure them. Your data warehouse and Business Intelligence tools should be constantly evaluating inventory to detect out of stocks, impending out of stocks, number of days out of stock, impact on selling rates for companion items when items are out of stock, etc.
Your alert systems should count the number of stockouts this week, measured against last week, and last year. Vendors, stores, planograms, and categories should be evaluated against out of stock performance objectives.
Data accuracy problems should be rooted out as out of stock issues are researched. It should be effortless for a central inventory manager to call for a stock count for a particular item and store. Conversely, if an item is suddenly not selling at a particular store, it should be a clue that there is a stockout at the store despite what the computer says.
Stockouts are an area where every retailer can make continuous improvements every day, every week, every season. But it takes the right informational resources in data warehousing and Business Intelligence.
This ranks up there with the most important discussions to take place on RetailWire. OOS are the single largest cause of lost sales in supermarkets. Unfortunately, this excerpt from CPGmatters is just another identification of the problem, which they conceded.
I totally disagree with all of the esoteric, process-driven “systems” advocated here. If they worked, they’d have been successfully employed decades ago. They make sense, but they’re not used. Consultants study the situation and present verbose work-arounds as part of their business. But there’s got to be a reason that their work-arounds aren’t working. Turns out, it’s implementation.
A supermarket veteran I met during my early days in the bidness spelled it out for me: “The number one reason for empty shelves,” he said, “is store managers running registers.” This hasn’t changed. Every time you see a store manager running a checkout in order to reduce labor costs, holes appear on the shelves. This is not a function of smaller back rooms, inadequate “densing up,” or whiz-bang computer solutions gone awry. It’s because the boss is not walking the aisles and holding his employees accountable for their sections.
This is not a technology issue. It’s a management issue.
There is not one issue but a myriad of issues which combine and result in an out-of-stock problem. While tools and studies might lend insights, a real solution provides store-level, easy-to-use tools powered by accurate data.
Approximately 32% of all store employees are between the ages of 16 – 24. To ensure a successful reduction in out-of-stocks, fail-safe solutions which clearly support category management and merchandising plans need to be implemented.
Even in this technological day and age, stock is often missing from grocers’ shelves to an unacceptable degree when the shopper is standing right there wanting it. When this happens to me, I try to find someone to help. If I do locate a store person to ask, and if they do agree to “check in the back,” in a high percentage of cases they do actually find what I wanted and bring it out. Total elapsed time for the whole process can be as long as ten minutes, however. I personally am willing to wait, albeit grudgingly, because in my worldview it is better than driving to another store for the item and I feel I am also helping the next customer who may want the same item. Many shoppers have neither the time, patience, or inclination to make this effort.
Most consumers, I think, can forgive an occasional out of stock situation, especially if it is related to weather, a manufacturing production slowdown or a warehouse catastrophe. What people cannot accept is a disconnect between the back storeroom and the public shelves of the same store.
I realize much of the discussion here relates to the grocery business, which is not my area of expertise, so let me confine my comments to general retail.
OOS issues ultimately begin with bad inventory data. You cannot replenish effectively with inaccurate data. Garbage in, garbage out. Maintaining accurate inventory data remains one of the greatest operational challenges in retail, despite all of the technological advances.
When OOS can be traced back to bad data (and that’s pretty easy to do), it is absolutely necessary to audit the transaction stream to identify the operational breakdown. If this is done as soon as the OOS is identified, finding the operational breakdowns is frequently pretty quick work, and staying on top of OOS and their causes will reveal any patterns in the breakdowns.
On the other hand, show me a retailer that has a well developed cycle count program in place, and I’ll show you a retailer who’s very good at identifying operational issues BEFORE out of stocks appear.
Every point applies from the study! However, isn’t a key need to find labor to fill the shelves? And this is an ongoing negotiation between retailers and major CPG Companies. Meaning, having the CPG company pay for shelf filling…or having its own merchandising teams support this retail function!
Anyway, as one views our Industry and its issues, the usual need is man and women power, e.g. labor cost, to operate a meaningful retail business that shoppers are loyal to!
Hmmmmmmmmmmm
All of these are great ideas. Every option provided in this poll is important to a retailer’s overall out-of-stock and out-of-shelf situations. However, the one proven solution that consistently reduces OOS by a significant degree is missing as a choice. The implementation of a demand forecast based computer generated ordering (CGO) tool has, in my opinion, the greatest impact on OOS.
If the product was not ordered from the supplier (direct delivery or distribution center) a great shelf management solution and shelf stocking processes will not solve the problem. A sophisticated OOS solution uses sales as a basis for ordering, so the first choice in this poll is close.
Best-of-breed CGO solutions consistently deliver reduction in OOS of 50% to 80% and therefore should be the prerequisite to all of the other choices.
I have worked with Dr. Gruen and Dr. Corstein on an out-of-stock study and the problems usually occurred at store level. Adherence to planograms would be a major step in addressing the issues along with proper training and communication with store personnel. Unfortunately, we forget about the people who do the work, the ones who face the customer every day, the ones who understand the problems. It is a fact that the first area those hours are cut is at store level. Our biggest and best asset is our store personnel. We need to develop an open dialogue between Operations and Merchandising/Marketing. Once we understand the problems each of us face, we can move toward solutions.
Some of the root causes:
– Stock Transfer
– Merchandizing
– Price issues
– Principle OOS
– Principle supply
– Transportation
– Discontinued items
– Wrong stock on hand
– Disorganization at the back door….
More Discussions
Has Same-Day Delivery Become an Expectation Across Retail?
Are expectations around retail delivery speed in flux?OpenAI Names the Infrastructure Standard for Agentic Commerce
Will OpenAI or Google (or another entity) set the gold standard for agentic commerce?Is Five Below Poised For Even Greater Growth?
Five Below is hitting serious growth numbers. Can this trend continue?
While this topic and poll include some great ideas, I believe the “low hanging fruit” falls into the “Other” category.
The most common missing link between planograms and the rest of the retail organization is the communication of MIN and MAX data to replenishment systems.
Retailers have done a great job of building enough planograms that are localized to the right markets. These planograms include MIN data, the lowest number of units by SKU that is acceptable so you still look like you’re in business, and MAX data, which is typically the capacity of the SKU in each planogram.
Retailers win when they connect the MIN/MAX data from their planograms to their replenishment systems. The really good retailers do it many weeks in advance so they know when to fill up or sell down their warehouse inventories in advance of space allocation changes.
When the retailer gets this link connected the consumer and the manufacturer win along with the retailer.
To Kai’s point, many of the out-of-stocks visible to the customer may not be visible to the retailer. If the merchandise in question is in the stockroom but not on the shelf (or showing in the store’s inventory even though it may be en route from a distribution center), the company’s central inventory management will not have visibility to the problem. A lot of effective management of out-of-stocks comes down to good store managers “walking the floor,” plus appropriate use of payroll for restocking.
Good grief! Here we are, still talking about this problem and coming up with the same old solutions which have yet to solve anything. We all know there are very few SKUs which sell in large quantities, more than we would like not selling at all, and the vast majority somewhere in the middle.
I would suggest three things could be done with today’s systems and technology. First, identify the best sellers by store and make it easier for the order writer to know at a glance what they are. If there is any stock in the back room, it should only be these items (not counting seasonal and promotional goods) so restocking them must have priority since that is the only stuff there. Second, Get rid of the dogs. This sounds easily done, but in reality it is extremely difficult to accomplish. Operations people do not like the write-offs and staff people do not like to admit mistakes. Take your medicine and get rid of them. Third, the new found space can now be used for expansion of the best sellers or for the introduction of new high-potential items.
At the risk of over-simplification,those steps would go a long way to reduce store level/controllable out of stocks. As for store replenishment, that is another story.
These are great practices that all retailers should use, however, the article forgets one of the most common problems and that is simple data usage. Although many retailers have sophisticated inventory management and planograms in place, they still don’t properly mine and use the data that is available to them. This includes basic POS data cross compared to shelf stocking quantities, velocity tracking and comparisons across stores, regions and categories (many items have definite correlations like milk, eggs and butter). Furthermore, stores have access to backroom inventory systems that should be enforced every bit as much as the inventory in the front of the store. This is a huge deficit for most stores and accounts for one of the key reasons for poor shelf space and OOS in the front of the store despite inventory in the back.
There are basically two objectives in inventory management for the self distributing retailer. For the warehouse, the objective is to maintain a balanced time supply of products in order to minimize handling and transportation charges. In the store, the goal is to “keep the shelves full.”
The warehouse requires accurate forecasts for both demand and delivery schedules along with rational expectations of what can be received. Too often, I have seen self inflicted inventory problems at retailer warehouses by over optimistic receiving schedules. They are almost like the airlines, expecting to “land” all their deliveries at the same popular times.
Store inventory management begins with allocation of presentation stock or shelf inventory. To some extent manufacturers have contributed to challenges in this area by offering so many variations of their products that allocating sufficient space becomes impossible. But retailers located in highly seasonal areas must also be flexible in the shelf allocations and be willing to invest in the resets necessary to accommodate demand shifts.
I think when addressing store out of stocks, you must divide your items into three categories: those items whose normal sales between deliveries is less than the shelf inventory (they require no backroom inventory and represent the overwhelming majority of the “center store” items), those items whose normal sales between delivery exceed the shelf space but the shelf space quantity is enough to cover the time between stocking periods (they require rework from backroom inventory between deliveries), and items whose normal sales exceed the time between stocking periods (i.e. between night crew shifts) that require special handling to make sure the shelf area is in-stock.
It is this last category that provides the most headaches, because items are constantly shifting in and out of this category based on promotion schedules. Store personnel must be given a “heads up” on the items whose expected movement between stocking periods exceeds the shelf allocation. It would be great if these items could be highlighted for the point of sale systems and a notice issued each time the shelf area capacity is approached. That way store personnel could be proactive in making sure these items are kept in stock.
The out of stock prevention steps seem reasonable. But the financial impact may be overstated. Very often, shoppers simply buy a substitute when the desired item is out of stock. That substitute is often bought at the same store. So has the store lost money on every out of stock item? Yes, the shopper’s annoyance may undermine her loyalty, but measuring that financial impact is very different from measuring the financial impact of allegedly lost sales. Furthermore, there’s a point where being 100% in stock at all times can be financially unreasonable versus 99% or 98% or 95%. Getting that last few percentage points can be disproportionately expensive.
The first step in controlling out of stocks is to measure them. Your data warehouse and Business Intelligence tools should be constantly evaluating inventory to detect out of stocks, impending out of stocks, number of days out of stock, impact on selling rates for companion items when items are out of stock, etc.
Your alert systems should count the number of stockouts this week, measured against last week, and last year. Vendors, stores, planograms, and categories should be evaluated against out of stock performance objectives.
Data accuracy problems should be rooted out as out of stock issues are researched. It should be effortless for a central inventory manager to call for a stock count for a particular item and store. Conversely, if an item is suddenly not selling at a particular store, it should be a clue that there is a stockout at the store despite what the computer says.
Stockouts are an area where every retailer can make continuous improvements every day, every week, every season. But it takes the right informational resources in data warehousing and Business Intelligence.
This ranks up there with the most important discussions to take place on RetailWire. OOS are the single largest cause of lost sales in supermarkets. Unfortunately, this excerpt from CPGmatters is just another identification of the problem, which they conceded.
I totally disagree with all of the esoteric, process-driven “systems” advocated here. If they worked, they’d have been successfully employed decades ago. They make sense, but they’re not used. Consultants study the situation and present verbose work-arounds as part of their business. But there’s got to be a reason that their work-arounds aren’t working. Turns out, it’s implementation.
A supermarket veteran I met during my early days in the bidness spelled it out for me: “The number one reason for empty shelves,” he said, “is store managers running registers.” This hasn’t changed. Every time you see a store manager running a checkout in order to reduce labor costs, holes appear on the shelves. This is not a function of smaller back rooms, inadequate “densing up,” or whiz-bang computer solutions gone awry. It’s because the boss is not walking the aisles and holding his employees accountable for their sections.
This is not a technology issue. It’s a management issue.
There is not one issue but a myriad of issues which combine and result in an out-of-stock problem. While tools and studies might lend insights, a real solution provides store-level, easy-to-use tools powered by accurate data.
Approximately 32% of all store employees are between the ages of 16 – 24. To ensure a successful reduction in out-of-stocks, fail-safe solutions which clearly support category management and merchandising plans need to be implemented.
Even in this technological day and age, stock is often missing from grocers’ shelves to an unacceptable degree when the shopper is standing right there wanting it. When this happens to me, I try to find someone to help. If I do locate a store person to ask, and if they do agree to “check in the back,” in a high percentage of cases they do actually find what I wanted and bring it out. Total elapsed time for the whole process can be as long as ten minutes, however. I personally am willing to wait, albeit grudgingly, because in my worldview it is better than driving to another store for the item and I feel I am also helping the next customer who may want the same item. Many shoppers have neither the time, patience, or inclination to make this effort.
Most consumers, I think, can forgive an occasional out of stock situation, especially if it is related to weather, a manufacturing production slowdown or a warehouse catastrophe. What people cannot accept is a disconnect between the back storeroom and the public shelves of the same store.
I realize much of the discussion here relates to the grocery business, which is not my area of expertise, so let me confine my comments to general retail.
OOS issues ultimately begin with bad inventory data. You cannot replenish effectively with inaccurate data. Garbage in, garbage out. Maintaining accurate inventory data remains one of the greatest operational challenges in retail, despite all of the technological advances.
When OOS can be traced back to bad data (and that’s pretty easy to do), it is absolutely necessary to audit the transaction stream to identify the operational breakdown. If this is done as soon as the OOS is identified, finding the operational breakdowns is frequently pretty quick work, and staying on top of OOS and their causes will reveal any patterns in the breakdowns.
On the other hand, show me a retailer that has a well developed cycle count program in place, and I’ll show you a retailer who’s very good at identifying operational issues BEFORE out of stocks appear.
Every point applies from the study! However, isn’t a key need to find labor to fill the shelves? And this is an ongoing negotiation between retailers and major CPG Companies. Meaning, having the CPG company pay for shelf filling…or having its own merchandising teams support this retail function!
Anyway, as one views our Industry and its issues, the usual need is man and women power, e.g. labor cost, to operate a meaningful retail business that shoppers are loyal to!
Hmmmmmmmmmmm
All of these are great ideas. Every option provided in this poll is important to a retailer’s overall out-of-stock and out-of-shelf situations. However, the one proven solution that consistently reduces OOS by a significant degree is missing as a choice. The implementation of a demand forecast based computer generated ordering (CGO) tool has, in my opinion, the greatest impact on OOS.
If the product was not ordered from the supplier (direct delivery or distribution center) a great shelf management solution and shelf stocking processes will not solve the problem. A sophisticated OOS solution uses sales as a basis for ordering, so the first choice in this poll is close.
Best-of-breed CGO solutions consistently deliver reduction in OOS of 50% to 80% and therefore should be the prerequisite to all of the other choices.
I have worked with Dr. Gruen and Dr. Corstein on an out-of-stock study and the problems usually occurred at store level. Adherence to planograms would be a major step in addressing the issues along with proper training and communication with store personnel. Unfortunately, we forget about the people who do the work, the ones who face the customer every day, the ones who understand the problems. It is a fact that the first area those hours are cut is at store level. Our biggest and best asset is our store personnel. We need to develop an open dialogue between Operations and Merchandising/Marketing. Once we understand the problems each of us face, we can move toward solutions.
Some of the root causes:
– Stock Transfer
– Merchandizing
– Price issues
– Principle OOS
– Principle supply
– Transportation
– Discontinued items
– Wrong stock on hand
– Disorganization at the back door….