Online food retailing

Online Grocery Retailing

This article will: prevent you from going bankrupt, demonstrate Internet opportunities for traditional retailers and show cost/profit modeling for effective e-business planning

Motivation for this article is the analysis of commercially successful and commercially failing online grocery retailers. And the experience as wholesale supplier of two well managed but failing online grocery businesses.

Management Summary

Success of online grocery shopping depends on market potential and distribution costs.

These key issues are interdependent. If market potential is high, distribution costs are moderately high. If market potential is low, distribution costs are extremely high. Because market potential of online grocery shopping appears to be low, online grocery retailers try to gain market share by lowering consumer prices (and margins). Result is a low margin operation with extremely high operating costs: the ultimate nightmare of every entrepreneur.

It is the ultimate challenge to develop an exception to this rule and the solution to this dilemma. There are three possibilities:

  1. Niche marketing: focus on a small, affluent and service oriented consumer group
  2. Focus on margin rich products, that sell well on the internet
  3. Focus on information distribution

These three solutions should be combined into a powerful Internet strategy, that should be executed in an excellent way. But there are many pitfalls and critical success factors. The pitfalls and success factors cover the entire marketing mix, the logistics system, information technology and internal organization.


High expectations but low online grocery penetration

In 2000 online grocery shopping was a hype. Some major e-commerce consultants were very optimistic about market potentials. Andersen Consulting used to predict a market share of online grocery shopping of 20 % in the year 2003. These optimistic predictions were used by to explain its high potential, and in 2001 defunct as support of a nationwide investment program of one billion US$.
Cap Gemini was even prepared to predict a percentage between 30 and 40 %, but the influential Verdict Research tuned the market potential of online retailing down to 3 % in 2004 for the total retail sector and 2.33 % for the grocery sector.
Fact is that in the year 2012 the online food market in the USA and in Europe is less than 0.3 % of total food sales. Potential for the next five years is not more than 0.3 – 0.5 % in the USA and not more than 0.2 – 0.4 % in most countries of Europe.

This low percentage of online grocery shopping is remarkable, because Internet penetration has grown sky high in USA and Europe and is growing with a high rate in other regions of the world:

Online grocery shopping

Why online grocery shopping is not popular

There are four basic reasons why online grocery shopping will never become popular by consumers. Home-shopping of groceries is no fun, it adds complexity to your lifestyle and it is more expensive than buying at the supermarket. On top of those disadvantages: traditional supermarkets are fighting for consumer loyalty by improving their marketing mix and increasing their efficiency.

1. Online grocery shopping is boring. In my view online shopping for dry groceries and perishables is boring. It does not even come close to the fun of buying books at the joy of buying clothes and shoes at, or the excitement of ordering exotic products in China by There is absolutely no advantage here over the weekly trip to the supermarket. I dare you to try online grocery shopping a couple of times and see for yourself. Only when an online business focuses on special products and on rich information content, online shoppers will become interested and stay interested.

2. Online grocery shopping is complex. Online shopping is less time consuming than traditional shopping, but it adds complexity to your lifestyle. Let us assume in an optimistic mood that every “household manager” will master the skill of shopping online. After ordering online you first have to be at home and make sure that the goods are properly received. Second you often have to go to the store anyway for miscellaneous articles. Third you have to check proper billing and payment. Fourth you have to follow up on order-picking mistakes and delivery errors.

3. Online grocery shopping is expensive. The distribution costs of home-shopping are twice as high as the costs of traditional food retailing, and most consumers are not willing to pay the extra 15 %. Internet startups will first try to gain market share with low prices and low service fees, but when the shareholders cash is consumed they will have to ask higher than “normal” prices to cover the costs and survive. Of course there is a small niche market for expensive home-shopping services: affluent PC-minded and service oriented consumers and consumers with no easy access to a nearby store.

4. Online grocery shopping is not competitive. In recent years Efficient Consumer Response and Category Management had a significant and positive impact on quality and efficiency of traditional supermarkets. Food retailing has always been a very competitive business, and in recent years super-marketing has become a professional science that is constantly improving the value to the consumer. In the USA, Europe and now also in India, China and South America very competitive stores with Every Day Low Prices and high service levels are gaining market share and are making the food business a war zone for new entrants.

Conclusion: the average consumer – with access to a good supermarket – does not have a good reason to go grocery shopping online.


High Distribution Costs of Online Grocery Retailing

Distribution costs of online grocery retailing are twice the distribution costs of traditional supermarkets. Traditional channels are extremely efficient, because consumers perform most of labor intensive work for the retailer: order-picking in the store and transportation of goods to the home. Costs of these activities are approximately 10 % of consumer price. Online retailers have to pay well trained and professional employees to do this work without errors. An additional 5 % of costs are spend on information technology, order management and after sales service. The additional costs are even higher in case of low market penetration or sub standard execution.

There are roughly three scenarios for online food distribution:

  1. Distribution from an existing store
  2. Direct home delivery from a dedicated central warehouse
  3. Home delivery from a central warehouse via satellite stations

Each scenario has its own specific benefits and value chain characteristics. Distribution from an existing store is appropriate for a small online business. Direct home delivery from a dedicated warehouse is appropriate for an online business with a high market share in a medium size densely populated area. Home delivery from a central warehouse via satellite stations is appropriate when the area is larger and the market share is lower.
The next table is an overview of supply chain activities and cost structure of each scenario, including traditional retailing. The figures are costs per activity as a percentage of consumer value.

Online grocery shopping
Cost/profit analysis by RetailEconomics.
[This cost structure is an indication only. In every country and each situation the costs per activity will be different. Outsourcing the delivery process to logistics service providers may add efficiency but will also increase complexity costs].

Example 1: is the oldest and best known online food retailer in the USA. Peapod once delivered from local stores, but in 1999 the company switched to direct delivery from central warehouses. In this case the delivery fee changed from US$ 16. (very high) to a minimum of $ 6.95 and a maximum of $ 9,95. Nowadays is a full subsidiary of Dutch Ahold, and turnover and profit figures are no longer published.
In the first 9 months of 1999 turnover was US$ 57.305, excluding additional income of
US$ 14.453. Gross margin was extremely low: 5.9 % of turnover. Distribution costs were exactly 30 % of turnover, in line with the cost/profit analysis table.
Other costs were: marketing and sales (13.2 %), information technology (5.9 %) and other overhead (19.7 %). The result was a loss of 37.6 % of turnover. Note that this is not a recent startup, but that this company has been in business since 1989.
The “about us” section of the website of made in 2000 a remarkable referral to “industry experts” who predicted a 20 % share of online groceries in the year 2003.

This example shows two things very clearly. It proves the fact that distribution costs of online grocery retailing are very high. But it also shows that in order to gain market share over traditional supermarkets the company had to invest heavily in price reduction and sales promotion. On top of that: overhead costs of exceeded the total distribution- and store costs of any well managed traditional supermarket chain.

Example 2: was in 1999 a new player in the American market. The company planned to develop 26 automated warehouses in populated areas all over the USA. These warehouses would deliver groceries to consumers via satellite stations. The Webvan case is interesting because it was a combination of high investments in buildings and equipment, and high customer service like a 30 minute delivery window.
The very professional “help section” of the Webvan website explained the pricing policy (“prices up to 5 % less than in local grocery stores”), and the product policy of a large and high quality assortment of groceries, perishables and drugstore items. Webvan offered free delivery for orders US$ 50 and over and a delivery fee of US$ 4.95 for orders under US% 50. The handling costs were “50 % less than the grocery industry average”.
The problem was that had based its business plan on the same optimistic online grocery market estimations as Peapod. The company filed bankruptcy in 2001. was caught in the same trap of every dedicated online grocery retailer: reducing the extremely high order fulfillment costs means high turnover; high turnover means low prices and high service. And when turnover stays low, the combination of low prices and high service means a very negative bottom line.

First Solution: Niche Marketing

“Niche marketing” is focusing the marketing mix on a special target group. This strategy is particularly effective for traditional retailers with a high market share and an affluent customer base. These retailers can develop an online channel as service offering, as market development strategy and as barrier against new entrants. In the USA this would be the recommended approach for successful local chains, in the UK for Tesco and in The Netherlands for Albert Heijn.
Niche online retailers use their quality image to reach for example 10 to 15 % of their own customers and 5 to 10 % of the rest of the market. Customers will spend on average 20 % of their grocery shopping online. In this way a company like Albert Heijn has the potential to reach about 10 % of total customers and win 2 % of total market share online.
The niche target group will consist mainly of affluent service oriented double income families, who are willing to pay the extra service fee. The other niche group are consumers who do not have easy access to a supermarket of choice. Every target group must have affinity to shopping on the Internet.

Second Solution: Focus on Margin Rich Products

A focus on margin rich products is the most debated solution for online food retailers.
Many e-commerce analysts assume, that consumers will prefer to order their heavy bulk products online, and visit the store for the rest of their grocery needs.
This assumption about consumer behavior is not supported by practical findings. In practice consumers will try to order enough products to avoid the delivery fee, in most cases set at an order amount of above 100 US$ or the equivalent in Europe. For online grocery retailers the sale of mainly bulk products is the quickest road to bankruptcy, since these products are the loss leaders of the grocery business.
There are already many examples of a clear product focus in online grocery retailing. The best example is wine, because successful wine selling requires professional information about the product and its use. Another important online food group is vitamins and supplements. In the USA there were some very attractive examples:, and I personally order my favorite brand of vitamins at
Online food retailers are also differentiating their business to financial services like banking, insurance and mortgages. Other service areas are video/DVD rental and dry cleaning, but the most attractive product development is non-food. This category is related to food and is therefore an expected extension to the online food business.

Walmart USA concentrates its Internet venture on its large non-food assortment, and use online food retailing as traffic builder. The Belgian food retailer Colruyt used online non-food retailing not only as an important margin rich extension of the food assortment, but also as a traffic builder for the supermarkets. Consumers could order the non-food articles online, but had to collect the merchandise at the store.

Third Solution: Focus on Information and Communication

“Information distribution and communication” is the name of the Internet game. Internet is the perfect channel for targeting different consumers and other stakeholders of the company.
In traditional advertising every commercial message is targeting a specific consumer profile with a specific message. Alternatively an Internet site has a menu structure, which makes it possible to communicate a whole range of messages to a wide range of interested target groups.
The other major advantage of Internet is the ability to really communicate in two directions. This important feature of Internet already has a great impact on the internal and external organization of any company. David Siegel, the author of “Creating Killer Web Sites” explained the impact of web communication on the organization in “Futurize your Interprise”.

Communication with different customers may focus on special products, like low fat or sugar free assortment. Allergy information gives the opportunity to add rich content to specific consumers. Other products that benefit from information are wines, private label products, vitamins and food supplements.

Retail organizations also use their website to inform their customers about their internal processes. (Safeway) is an excellent example of a description of the distribution process, and it enhances the professional image of the company.
In the USA the Internet is frequently used to distribute coupons. A good example is However the disadvantage of Internet distribution of coupons is fraud: the possibility to change rebate figures by the PC.

The most promising communication feature is the possibility for customers to respond to the retailers information and marketing mix elements. A retailer may solicit assortment suggestions and complaints. Internet makes it possible to use customer complaints as powerful drivers for quality improvement in the company.

Communication with other stakeholders of the company requires much attention but opens many opportunities. Most professional websites already have a special section for shareholders (annual and quarterly reports), potential employees (vacancies and working climate), suppliers (category management approach) and the media (latest news). Each target group gets maximum attention without annoying the others.
In case of communication with other stakeholders each target group must have a professional contact in the company, who is responsible for content and direct response. If this is organized in a professional way, the Internet communication will strengthen external relationships and internal responsiveness to the market.

Running and winning the “Longest Mile”
Physical distribution of goods ordered online has been called the “Achilles heel” of e-commerce. It is the week spot of a powerful development.
In the USA physical distribution is now referred to as “the longest mile”. It is an old trade, that was mastered by an ex Wallstreet investor like Jeff Bezos of Jeff Bezos solved the sore knees of employees (and himself!), who were packaging books sitting on the floor, by issuing knee protectors. His partner suggested the idea of purchasing some packing tables, so he told with laughter on TV.

When the home shopping market increases, professional service providers will develop distribution networks for general use by different online retailers. This will reduce the transportation costs by approximately 30 % and the total additional online grocery shopping costs by approximately 1.5 % points. is using this concept by sending books by mail, but a 1.5 % point reduction of food distribution costs is too small a gain to change the less then positive outlook of online grocery shopping for the best.


Cost/Profit Modeling

Costs of order management and distribution alternatives can be evaluated from a distance using “cost/profit modeling”. It is even better to use this tool in the business plan preparation situation. Cost/profit modeling was used to develop the value chain presented in the table at the beginning of this article.
A good example is the home delivery process. This is a major element of order fulfillment. Online retailers have to optimize the following cost elements of distribution:

  • The average distance from the warehouse to the customers
  • The average distance between customers
  • The stop time at customers
  • The loading- and unloading time
  • Handling efficiency
  • Cost per hour
  • Vehicle fill rate and utilization
  • Capital investment

The optimization process of distribution elements will be covered in a different article. The point is that every distribution decision has an impact on all eight elements. It is interesting to show the most important effects of some home delivery alternatives on these distribution cost elements.
The decision of Peopod to change delivery from stores to delivery from central warehouses improved handling efficiency but also increases capital investment and the average distance to the customer.

The semi automated warehouses of required huge capital investments, but were intended to increase handling efficiency. The use of local satellite stations increased capital investments but also had a positive effect on vehicle fill-rate and vehicle utilization.
The very narrow 30 minutes delivery time window of greatly increased average distance between customers. This service element was impossible to maintain, because it tripled the already extremely high transportation costs.


Critical Success Factors of Online Grocery Retailing

Any successful online retailer has to work hard and intelligently on a large number of critical success factors. These factors are critical because failure to excel in one of them may damage the company considerably.
We already covered the critical success factors customer targeting, focus on margin rich products and focus on “information distribution” and communication.

Other critical success factors are:

Trust. Online customers must have absolute trust in the retailer, because the retailer takes control of most of the shopping process. It is a good start if the retailer has an excellent reputation to start with, but when new channels are used, the retailers must be certain that this trust is not violated

Tailoring. The site should be tailored to the individual customer. An excellent example is, that greets you personally when you open the site, and gives you a choice of products that match your personal preferences

Choice. Each target customer group must have excellent choice within every product category. The online retailer must exceed the product offering of traditional competitors in order to create loyalty to the business. The best strategy is to add high end products to the assortment, and promote these products heavily on the site

Quality. Each service element of the online retailer must have excellent and predictable quality. Mistakes in online retailing have a much bigger impact than in traditional retailing

Price. Internet makes pricing very transparent and this is a great concern for online retailers. Most American online retailers charge prices 3 to 5 % lower than traditional retailers, with the effect that gross margins do not cover the costs. European online retailers tend to charge 5 to 8 % higher prices than normal, but they will be exposed by price comparing websites. The solution is to charge market conform prices and an additional service fee

Promotion. Each marketing mix element must be promoted on the site, and the site must be promoted in traditional promotion channels. This emphasis on site promotion is a substitute for the physical store and the physical signals in the store

Information Technology. Web technology features many commercially attractive possibilities like feedback on input, preference lists, cross-selling, inventory checks, differentiating service and cost structures, and professional order management

Payment. This is a major subject in itself. Basic is error free and easy payment. Optional is the service of payment alternatives

Economies of scale. High investments in information technology and marketing must at least in the long term be covered by high turnover. Even more important is the necessity of a high market penetration and a dense customer network, in order to optimize delivery costs.

Take the Challenge Seriously

Earlier publications on which this article was based have already evoked much discussion. The author invites readers to supply feedback, constructive criticism and new scenario’s to make online food (and non-food) retailing profitable.


drs Joost van der Laan
J.W. van der Laan Marketing & Logistics BV
Prins Clauslaan 3
3852 DA Ermelo, Netherlands
Tel.: 0031-6-53846927
Chamber of Commerce 320 56641, VAT number NL802737213B01



DPP for profitability of food retailers

Direct Product Profitability

The goal of a Direct Product Profitability (DPP) project is to:

Improve sales and gross margin by changing: product assortment, article presentation in the store and consumer prices

Reduce costs by changing: process (logistics, store handling) and product characteristics (package size, item size)

Results of a DPP analysis of 5 different products.

Item A is a slow-moving product with high gross margin. Item E is a fast-moving product with low gross margin. Items B, C and D are medium-moving products with average gross margins and different prices.

Using a traditional profitability ratio (gross margin per week) and even gross margin per week per m3 shelf space, the fast moving product E appears to be the most profitable.

Item C is most profitable. Using the DPP model and after assigning direct product costs to these products, item E shows to be a mega looser, the high gross margin product A also shows a net loss, and only the other 3 products appear to deliver a net profit.

Direct Product Profitability tableDPP analysis had a considerable impact on assortment tactics and pricing strategy.


The Seven Step DPP process

The DPP model is capable of calculating net profitability of individual items of fast moving consumer goods. Working with the DPP model is a seven-step process:

  1. DPP model fine tuning: the classical DPP model is adapted to specific product characteristics of your industry
  2. Input of process characteristics: process characteristics of the logistics chain are entered as activity drivers in the DPP model (examples: delivery frequency, productivity ratios)
  3. Input of general ledger resource costs: resource costs of the central depot, transportation and the store (examples: transportation cost per km, costs per working hour)
  4. Calculation of activity costs: activity costs are calculated in the DPP model
  5. Input of product characteristics: all characteristics of individual products are entered as cost drivers
  6. Calculation of direct product costs: activity costs are allocated to products
  7. Calculation and presentation of direct product profitability ratios


Suggested project approach

A. Agree on project plan during initial meeting:

  • Project scope: which assortment, which pilot stores
  • Project organisation, involvement of consultant
  • Project timing
  • Project budget

B. Execute the seven step DDP approach

C. Suggest improvement plans and calculate cost- and profitability effects in 5 areas:

  1. Process changes
  2. Product characteristics
  3. Assortment
  4. Pricing structure
  5. Article presentation in the store

D. Suggest measures to implement improvement plans

E. Final report


Example: resource costs versus activity costs

2% cost reduction. Resource costs are normally well documented and controlled. Because resource costs are continually reported and controlled, there is often no opportunity (<2%) for substantial cost reduction. See the example for transportation resource costs in the next diagram.

Transport resource costs

20% cost reduction. Activity costs are normally not reported by classical administration systems. To show these costs, a special activity cost driver model is employed. The advantage of analyzing activity costs is the possibility to substantially (>20%) reduce costs by improving the process.

In the transportation example the activity cost reduction possibilities are: improve load factor, reduce delivery frequency, reduce waiting times, etc.

Transport activity costs


Please contact RetailEconomics


drs Joost van der Laan
J.W. van der Laan Marketing & Logistics BV
Prins Clauslaan 3
3852 DA Ermelo, Netherlands
Tel.: 0031-6-53846927
Chamber of Commerce 320 56641, VAT number NL802737213B01


Category Management for Supermarkets

Category Management

Category Management is the management of product categories as strategic business units, were assortment, pricing, inventory levels, shelf-space allocation, promotions and buying are all managed as a whole.

This intensive training course is all about implementing Category Management in your organization.

Workshop program

1. Development of Category Management

  • The pioneers: Coca-Cola, Procter & Gamble and Wal*Mart
  • The impact of Efficient Consumer Response
  • The impact of retailer concentration and globalization
  • Future development trends

2. Category Management from Retailer Perspective

  • Store format optimization
  • Assortment, pricing, presentation, logistics
  • Sales, gross margins, costs
  • Out-of-stocks, availability
  • Consumer happiness

3. Category Management from Manufacturer Perspective

  • Trade marketing versus retail marketing
  • Aligning category management and sales targets
  • Activation of different distribution channels
  • Eight trade marketing strategies to cope with category management

4. Win-Win Strategies for trading partners

  • Retailers improve retail marketing
  • Manufacturers improve trade marketing and brand value
  • Trading partners improve cooperation
  • International opportunities for manufacturers
  • How to handle confidential information

5. Successful Category Management Process

  • Definition: consumer segmentation and category structure
  • Role: cross category analysis and profitability evaluation
  • Assessment: identifying sales and profit opportunities
  • Balanced scorecard: performance targets
  • Strategies: assigning strategies to each category segment
  • Tactics: planning activities for results
  • Plan implementation: effective action
  • Review: measuring against benchmarks

6. Manufacturer Response to Specific Category Strategies

  • Traffic building
  • Point-of-Sale output
  • Category profitability
  • Cash generation
  • Creating excitement and customer triggers
  • Image enhancement

7. Importance of Effective Use of Data

  • Consumer behaviour (GfK household panel data, specific consumer research)
  • Market share and causal data (Nielsen data etc.)
  • Geographics
  • Retailer sales, margins, cost indicators, store locations, space management, promotions, introductions
  • Manufacturer turnover, margin indicators, promotional budgets, promotion results, introduction budgets and results

8. Effective Analytical and Planning Models (“enablers”)

  • Space management
  • Direct product profitability
  • Account profitability
  • Activity Based Costing
  • Pricing strategies, brandpricing, price elasticity

9. Management of Specific products

  • Impulse presentation of profit makers
  • Pricing strategy and presentation of private labels
  • Availability and shrinkage of theft sensitive products
  • The critical success factor of Wines
  • The cat-food mystery
  • Timing of fresh foods
  • Lean logistics of mega-losers
  • Packaging of slow-movers

10. Cutting-Edge Strategy for Category Management Projects

  • An effective and detailed approach
  • The role and commitments of each participating company
  • Team members and their work
  • Project planning
  • Avoiding and solving pitfalls
  • Internal and external communication

11. Case Study 1

  • A case of category management failure
  • What went wrong and the solution out

12. Case Study 2

  • Analysis of a successful category management case
  • Critical success factors

13. Retailer Buying Models and Manufacturer Response

  • Pure buying and negotiation
  • Merchandising and below the line focus
  • Format management and manufacturer investment in trade marketing
  • Decision making unit (DMU) and input of production and logistics
  • Integrated category management and retailer-manufacturer teams

14. Manufacturer Organization of Trade Marketing and Category Management

  • Four basic organization models of trade marketing
  • Tasks of category management, account management and brand management

15. Conclusion: Bring Everything to Action

  • Category management as part of the commercial organization
  • Day to day working between retailers and suppliers
  • How to share costs and benefits

The Category Management workshop is multi-company, single company or for two retailer/supplier partners. A full Category Management Workshop takes two days. It is possible to make a selection (“menu”) and organize a workshop of one day.


Please contact RetailEconomics for your Category Management workshop


drs Joost van der Laan
J.W. van der Laan Marketing & Logistics BV
Prins Clauslaan 3
3852 DA Ermelo, Netherlands
Tel.: 0031-6-53846927
Chamber of Commerce 320 56641, VAT number NL802737213B01


Supermarkets high productivity

Retail Store Productivity

Since net margins of retailers are in the range of 2-5%, 10% reduction of labor costs will double net profit!

Productivity improvement by “Internal Benchmarking” has been developed on the shopfloor, to improve productivity with the consent of employees!

Labor cost management has three levels of productivity drivers:

  • Strategic level: the marketing mix of the store format drives the activities in the store
  • Tactical level: productivity planning and human resources drive the standard time per activity and labor costs per hour
  • Operational level: store management executes the plans in a dynamic environment of fluctuating sales, customer and employee demands, logistics problems and other deviations from plan.

Retailers must use the best productivity planning method. See: “Internal Benchmarking”.


Objectives of a successful productivity project are:

  • Measuring and analyzing labor productivity and labor cost ratios of a retail chain, in particular the relative performance of individual stores in a multi outlet retail chain
  • Optimal productivity standards for each store and department (sales per work hour)
  • Optimal labor cost standards (costs per work hour)
  • A functional labor cost planning system
  • Minimal 3% decrease of total labor costs

For a retailer with 500 million turnover per year and 12% labor costs, the potential savings are 1.8 million per year. A project costing 200.000 will have a pay-back period of 2 months.


A productivity project will focus on control of labor “cost drivers”:

  • The right activities for optimal customer satisfaction (no weakening of the store format)
  • High productivity standards per department and per activity
  • Tactical planning of working hours in line with turnover fluctuations
  • Prevention of “idle time” (service departments)
  • The right activities executed by the right employees
  • Control of costs per hour per function

The project will take into account all external factors that have an effect on labor costs: service or self-service, turnover shares of product groups and departments, average price per article etc.


Suggested project approach:

  1. Discussion of suggested project plan with company leadership
  2. Informing the organization
  3. Detailing the plan with operational managers
  4. Defining departments and activities
  5. Measuring and analyzing present situation
  6. Setting new standards for productivity and costs per work hour, using our efficient “Internal Benchmarking” method
  7. Assessment of deviations of standard per department per store
  8. Implementation of productivity improvements and cost reduction in the stores
  9. Implementation of planning system
  10. Organizing continuous improvement

Each retail company should do this work every three years, to keep the organization lean and sharp. We are glad to help you with this profitable action.


Please contact RetailEconomics for your Productivity Improvement.


drs Joost van der Laan
J.W. van der Laan Marketing & Logistics BV
Prins Clauslaan 3
3852 DA Ermelo, Netherlands
Tel.: 0031-6-53846927
Chamber of Commerce 320 56641, VAT number NL802737213B01


Automatic replenishment

Automatic Replenishment

Automatic Store Replenishment is a powerful strategic weapon for retailers. Seven benefits of Automatic Store Replenishment (or Computer Assisted Ordering) and cycle-counting.

New features are: Nice Shelf Stock to improve commercially attractive shelves, Counting by Exception to improve data integrity and store productivity and Store Order Leveling™ to improve logistics productivity.

Automatic Store Replenishment is: “the preparation of orders by a computer integrating information about product movement (as recorded by point of sale equipment), outside factors that affect demand (such as seasonal changes), actual inventory levels, product receipts, and acceptable safety stock levels. Inventory data integrity is maintained by cycle-counting. (Source : ECR US – Automatic Store Replenishment, Computer Assisted Ordering, Sales Based Ordering, Automatic Ordering)

The seven benefits of Automatic Store Replenishment and cycle-counting are:

  1. Lower out-of-stocks and higher sales
  2. Lower inventory costs and higher margins
  3. Lower labor costs by cycle-counting
  4. Shrinkage reduction
  5. Store Order Leveling™ and truck-load optimization
  6. Reduction of price mark-downs
  7. Reduction impulse-buying of loss-leaders

How to Realize Maximum Profits from Automatic Store Replenishment

This article shows how to benefit from each aspect of Automatic Store Replenishment. How to build high data integrity. What are pitfalls to avoid. How to manipulate ordering parameters. What inventory choices to make: do you prefer minimum inventories or full shelves. And what are benefits of warehouse production leveling and truck-load optimization.


Benefit 1. Lower Out-of-Stocks and Higher Sales

A rule-of-thumb is that 1 % out-of-stock results in 0.5 % lower turnover. When products are out-of-stock, retailers suffer from customers buying elsewhere or buying not at all. Manufacturers suffer from brand substitution and also from non-sales. Strong brands show more store-substitution and weak brands show more brand substitution. Effects are different for stores with large assortments than for stores with small assortments. And sales effects of out-of-stocks are limited for both retailers and manufacturers by package-size substitution and delay of purchase.

Ordering mistakes by store employees are a main cause of out-of-stocks. Other main causes are delivery delays of manufacturers, out-of-stocks and order-picking mistakes of the central warehouse and the quality of product introduction and deletion. But in-store manual ordering mistakes amount to at least 50 % of out-of-stocks. No manual ordering mistakes is one of many benefits of Automatic Store Replenishment.

Why do store-orders improve by Automatic Store Replenishment? Inventory levels at time of delivery are estimated by a forecasting system. The system takes into account any shipments underway and turnover during lead-time from order to delivery. This forecasting is almost impossible by classical manual ordering. But there are some conditions for Automatic Store Replenishment to work correctly.

Conditions for correct store-orders are:

  • Low non-scanning rates to generate correct sales data
  • Discrete (separate) scanning of varieties of products (colors, flavors)
  • Regular match of physical and database inventory levels through cycle-counting
  • Correct input of minimum and maximum shelf-stock levels

If these conditions are met, lower out-of-stocks will have a positive impact on net profitability. See the next calculation.


Realistic net profit estimate for out-of-stock reduction by Automatic Store Replenishment, based on following assumptions:

  • Out-of-stock level by store ordering mistakes = 2 %
  • Real out-of-stock reduction by ASR = 1.5 % (75 % of out-of-stocks caused by in-store ordering)
  • 1 % out-of-stock reduction results in 0.5 % sales increase
  • ASR is used for 60 % of sales
  • Average direct product profitability of ASR assortment = 8 %

Potential net profit of out-of-stock reduction by Automatic Store Replenishment = 180.000 for every 500 million of total sales.


In my practice some retailers already have low out-of-stocks by solving every order-picking / ordering mistake by extra deliveries, shipments and returns. This solution proves extremely expensive. In these cases net profits from Automatic Store Replenishment are up to 350.000 for every 500 million of sales.


Benefit 2. Lower Inventory Costs and Higher Margins

Automatic Store Replenishment makes it possible to choose between many different ordering techniques. Most important are “fill-to-minimum” and “fill-to-maximum”. Fill-to-minimum is ordering just enough to stay above minimum stock levels. Fill-to-maximum is ordering for full shelves. Advantage of fill-to-minimum is low average stocks, but disadvantage is a good chance of unattractive looking shelves. Advantage of fill-to-maximum is attractive full shelves, but disadvantages are higher average stocks and increased possibilities of product shrinkage.

Highest profit from Automatic Store Replenishment – and every solution to any problem! – is obtained by maximizing advantages and minimizing disadvantages. For store ordering this means: choosing a middle road between fill-to-minimum” and “fill-to-maximum”. You can work with a percentage of each ordering technique, or you can define so-called “nice stocks”.

“Nice Shelf Stocks” are minimum shelf stocks to give shelves an attractive look. “Nice Shelf Stock”-levels are often larger than safety-stock levels, which are maintained to avoid out-of-stocks when turnover is extra high. A simple way to define “nice stocks” is: one or two items per facing.

Other inventory benefits of Automatic Store Replenishment are “sales anticipation” and “on-order information”.

“Sales anticipation” means that the ordering system makes a forecast of turnover during lead-time between order and delivery. This forecasted turnover is added to the order, so shelf-stock is just right when delivery is put on the shelf.

“On-order information” takes into account that during ordering a former delivery has not arrived yet. In classical manual ordering this problem is often tackled by lowering order frequencies. Automatic Store Replenishment has no problem with deliveries on route. These are just added to in-store stocks when calculating an order.


Realistic net profit estimate for stock reduction by Automatic Store Replenishment, based on assumptions:

  • Ordering from fill-to-maximum to ASR (fill-to-minimum and nice stock)
  • Average shelf-stock level fill-to-maximum = 2 weeks
  • Average shelf-stock level ASR = 1.4 weeks
  • ASR is used for 60 % of sales
  • Average net buying costs ASR assortment = 62 %
  • Capital costs = 6 %

Potential net profit of stock reduction by Automatic Store Replenishment = 130.000 for every 500 million of total sales. This is excluding opportunities for gross-margin increase.


Space for new assortment is created by Automatic Store Replenishment, because fast-moving items often need less facings when stock levels go down. New assortment creates more choice for customers and higher sales. Facings of slow-moving items are not effected by lower stock levels, because facings of slow-movers totally depend on case-size. See article on optimal case-size.

Automatic Store Replenishment makes it possible to substitute low-margin sales by high-margin sales. Fast-moving items – especially A-brands – often have very low gross-margins and negative net product profitability. Lower stock levels reduce impulse buying of these loss-leaders. New assortment can be chosen from high margin niche groups. Or you give high margin private brands extra facings to induce impulse-buying. Resulting average gross-margins will rise. If free space is used optimally, resulting net profitability will be higher than net profit from stock reduction.


Benefit 3. Lower Labor Costs by Cycle-Counting

Automatic Store Replenishment reduces labor costs, because manual ordering is substituted by automatic ordering. It is true that new tasks are introduced to maintain correct stock data, but these tasks require fewer and cheaper labor.

Cycle-counting is the technique of choice to maintain correct stock data. Instead of checking shelf stocks at each order cycle, stock levels are counted every 4 to 8 weeks and compared to database information. Labor cost savings are substantial:

  • Cycle-counting frequencies for ASR are much lower than manual ordering frequencies
  • Cycle-counting is a simpler task than manual ordering and therefore cheaper
  • Timing of cycle-counting is not time-critical and can be done at quiet moments
  • ASR techniques as “Counting by Exception” improve efficiency and data-integrity

Counting by Exception is an important feature of Automatic Store Replenishment, which makes cycle-counting even more efficient. Exceptional stock levels trigger counting: zero physical stocks, zero database stocks, zero turnover and vulnerable products. Effect: obvious problems are tackled with a minimum of work.

Zero physical stocks are checked to detect order picking mistakes and theft. In takes a few minutes to walk the isles after every shelf filling cycle, and to check empty – or almost empty – shelf spaces and corresponding database stock-levels. Zero physical checks are important: if database stocks are above minimum stock levels and physical stock levels are zero or very low, the ordering system will not generate necessary replenishment orders. Sales will drop to zero.

Checking zero (or negative) database stocks is method of choice to manage shelf space for unexpectedly successful products, deleted products, new product introductions and products temporarily out-of-stock. A list of zero stocks is compared with physical stocks every day. This also takes little time. Deviations require some analysis and adjusting of shelf space.

Zero turnover checks make it possible to detect problems with product availability and visibility of fast-moving products. If a product is normally purchased 20 times a day, it only takes a few hours to produce an automatic trigger for store management to check if anything is wrong with products on shelf. Zero turnover detection requires sophisticated statistical software, but is a powerful tool to improve attractiveness of the store.

Vulnerable products are articles which have a high probability of shrinkage and other causes of stock level deviations. Examples are perishable goods, or small expensive items like cosmetics, cigarettes, batteries and razor blades. These products (or product groups) are pre-defined or are detected by frequent stock level mutations. They are cycle-counted more frequently than normal. The effect is that counting frequencies for “normal” products are limited to a minimum level, and average counting frequency remains low.


Realistic net profit estimate of labor cost reduction by Automatic Store Replenishment, based on assumptions about manual ordering and cycle-counting:

  • Traditional ordering requires 5.0 % of total man-hours of a store
  • Traditional ordering costs € 12 per man-hour
  • cycle-counting and order management of ASR requires 2.8 % of total man-hours of a store, and costs € 9 per man-hour
  • Total man-hours of a store = 1 hour per € 135 of sales
  • ASR is used for 60 % of sales

Potential net profit of labor cost reduction by ASR = € 775.000 for every € 500 million of total sales.


Benefit 4: Shrinkage Reduction

High frequency cycle-counting of vulnerable products will detect shrinkage problems in an early stage. When store management addresses these problems, cycle-counting makes it possible to measure effects and evaluate solutions.

Shrinkage of vulnerable products can be resolved in many ways, for example:

  • Increasing visibility of small and expensive items by placing them next to point-of-sale, but maintaining self-service presentation
  • Reducing shelf stock to a minimum of a few hours of sales
  • Reducing “sweeping” of multiple products by placing high priced items in racks
  • RFID tagging
  • Placing products behind a service counter to reduce theft by customers and employees.

Each solution has advantages and disadvantages. For example: placing products behind a service counter will result in lower sales and higher costs. This is a high price to pay for shrinkage reduction. It is better to develop solutions that increase sales, at the cost of some closely monitored shrinkage.

Placing small expensive items next to point-of-sale has two distinctive advantages. We measured the effects in two interesting shrinkage studies. First benefit: increased visibility for POS workers and lower shrinkage. Second benefit: increased visibility for customers and higher sales. Increased visibility is very interesting for high margin batteries and razor blades. These products generate high gross margins (up to 60 % of an average price of € 4,50), and they have a high visibility elasticity or high impulse factor. Customers buy these products when they see them. Batteries and razor blades are the best products to place on a counter or next to point-of-sale. Shelf profitability of this high-opportunity store real estate is much higher then for the usual low-price/ low-margin candy bars.

We were able to obtain these results only through the availability of item-level inventory and cycle-counting.


Benefit 5. Store Order Leveling and Truck-Load Optimization

Turnover is not equally distributed over days of the week. More customers shop on Fridays and Saturdays than on other days of the week. These two days together may even constitute half of total weekly sales, as in the next diagram.

ASR turnover per day

When stores order every day, they will generate large orders around the weekend and small orders in the middle of the week. This disrupts the entire supply chain:

  • Unbalanced workload at distribution centers and suppliers
  • Unbalanced truckloads and average lower utilization of trucks and truck space
  • Unbalanced workload at store

This problem is not new and retail organizations have developed different solutions:

  • Ordering and distributing slow-moving products only on quite days
  • Pre-producing fast-moving items on quite days
  • Maximizing daily orders

But every solution creates a new problem. Ordering slow-movers on quite days makes the ordering and distribution system more complex. Pre-producing fast-movers on quite days creates extra inventories in the pipe-line. Maximizing daily orders makes ordering more complex and is cause for increasing out-of-stocks on busy days.

The best solution is Store Order Leveling™ by simply manipulating input parameters of the automatic replenishment system. The system uses free shelf space of set-to-minimum order algorithms, or redundant stock levels of set-to-maximum algorithms. Orders can be leveled to an ideal distribution over days of the week. The next diagram shows flat orders fro Monday to Friday and reduced orders on Saturday.

Store Order Leveling

We have made many simulations for set-to-minimum and set-to-maximum order algorithms, and these simulations are tested in real-life situations. It is now possible to generate almost any order leveling distribution for any actual turnover distribution. It is also possible to shift automatic orders further to compensate for manual – non-leveled – orders.

Estimated financial results of Store Order Leveling™ are very interesting:


Realistic net profit estimate of labor cost reduction and truckload optimization by Store Order Leveling™, based on assumptions about average turnover distribution over days of the week:

  • Saving of 2 full-time equivalents in distribution center = € 50.000
  • Saving of 2 trucks = € 250.000
  • Saving of 8 full-time equivalents in stores = € 160.000

Potential net profit of labor cost reduction and truckload optimization by Store Order Leveling™ = € 460.000 for every € 500 million of total sales.


Benefit 6. Reduction of Price Mark-Downs

Fashion retailers plan their budgets every season for item procurement and distribution to the stores. Logistics is mainly push oriented. Traditionally the leftovers are discounted at the end of the season with high mark-downs. These mark-downs are high because the season is over and leftovers are the least desirable items for regular customers.

A system with item-level turnover planning and inventory control makes it possible to monitor sales development from day one. After a few weeks it is clear which items are slower then expected and will not entirely be sold during the season. If price mark-down is started during the season, overall mark-downs will be lower. This is called “quick response”. The earlier a store starts its mark-down program, the lower the overall price reduction will be. See the next table.

ASR suggested markdowns

Profit effects depend on seasonality and fashion characteristics of the assortment. If 20 % of items is normally marked-down on average by 50% at the end of the season, and “quick response” makes it possible to lower average mark-down to 40%, the net profit effect is a staggering 10 million on every 500 million of turnover. There will be some substitution effects in the season, but early mark-down will also improve the price image and customer attractiveness of the store.

One additional aspect of “quick response” is the reverse of early mark-downs: additional production and distribution of fast-moving items. If sales of certain items are higher than expected, producers should be flexible and quick enough to produce and distribute these items to the stores in a very short time.


Benefit 7. Reduction Impulse-Buying of Loss-Leaders

Often there is no match between logistics requirements and commercial requirements. For example: logistics managers prefer low distribution costs and low delivery frequencies, but commercial managers prefer low store inventories, low out-of-stocks and high delivery frequencies. An easy solution is a compromise with a heavy weight on store requirements.

A better solution is to choose for high delivery frequencies to the stores, in combination with concentrated deliveries by one channel. In other words: minimal “direct store deliveries” and maximum deliveries via the central warehouse. This creates a win-win situation for the stores and for logistics.

But high delivery frequencies are not sufficient to prevent a costly problem in the stores. There are always 100 – 150 super fast moving items – with low gross-margins – that still require much shelf space in order to meet customer demand. Large shelf space means high visibility, and high visibility means high impulse sales. And here is the problem: super fast moving items are often loss-leaders: net product profitability is below zero. You do not want to sell these products, but you have to. Extra impulse buying works very negative on the stores profitability and must be avoided. But these items are traffic builders and they must never be out-of-stock, otherwise the store will lose customers to the competition.

Item-level inventory control – as part of Automatic Store Replenishment” – makes it possible to maintain low inventory levels and low visibility on the shelf, in combination with an extra backroom stock. The computer will take the two stock locations per item into account, for cycle-counting and ordering purposes. Unwanted impulse buying of loss-leaders is prevented and there is even shelf space available for additional high-margin assortment.

Net profitability effects are not easy to calculate, because ”space-elasticity” is different for every item. We also have to make estimates about substitution from low margin to high margin products. But this improvement concept is intuitively promising and can be tested in practice.


Conclusions and Recommendations

Potential results of Automatic Store Replenishment are high: 1.5 million hard benefits and another 1 million soft benefits for every 500.000 million of turnover. For fashion retailers potential benefits are much higher, because these retailers have the potential to cut back in expensive price mark-downs.

These benefits have a long life span. If you are making an investment decision for SCM software licenses and implementation, use at least a 5 years time span. At 10% DCF, the NPV of hard benefits is 5.5 million. This is about 10 times the cost of a SCM software licence and these benefits may cover the investment of an entire ERP system.

There are three important critical success factors for Automatic Store Replenishment:

  • At start: data integrity
  • During project: develop correct forecasting and order algorithms and avoid manual order corrections
  • After project: use Automatic Store Replenishment information for maximum benefits

These critical success factors must be managed to realize their specified results. This requires assistance from experienced implementers and users of Automatic Store Replenishment.


Strategy and Tactics for Optimizing Results:

RetailEconomics offers consultancy services to realize full potential benefits of Automatic Store Replenishment, by focusing on critical success factors.

Start an Automatic Store Replenishment project

  • Execute a quick scan and write a business case for Automatic Store Replenishment
  • Select and implement your Automatic Store Replenishment system
  • Develop right conditions for correct inventories and automatic orders (data integrity)

Execute an Automatic Store Replenishment project

  • Develop simple and effective order algorithms
  • Develop correct ordering parameters for every category
  • Develop effective system of Counting by Exception
  • Provide training
  • Monitor order corrections and develop structural solutions

Use Automatic Store Replenishment information for maximum benefits

  • Analyze shrinkage reduction opportunities
  • Propose Store Order Leveling™ parameters
  • Propose alternative assortments for higher net profitability
  • Propose alternative shelf presentation for higher net profitability
  • Develop procedures to “lock” quality and for continuous improvement

We are happy to deliver a detailed plan, based on your specific questions.


Please contact RetailEconomics for Automatic Store Replenishment


drs Joost van der Laan
J.W. van der Laan Marketing & Logistics BV
Prins Clauslaan 3
3852 DA Ermelo, Netherlands
Tel.: 0031-6-53846927
Chamber of Commerce 320 56641, VAT number NL802737213B01