A more efficient ordering process through better exception management

Jan 5, 2012 7 min

Many companies switch to automated replenishment to improve shelf availability and inventory turnover, and to make re-ordering more efficient. Increasing automation generally produces positive results, especially for those products with a regular, predictable sales pattern. It’s easy to replenish them automatically.

It doesn’t, however, make your replenishment managers redundant. Far from it. Good systems both improve efficiency and highlight challenging areas. Your colleagues are now free to focus on those tasks that need their skills and experience and to extend improved results so they become great results.

Why Is Exception Management Necessary?

Many businesses seem to expect automated re-ordering to be the answer to all their replenishment problems and to work, unsupervised, in every situation. Not surprisingly many businesses find themselves disappointed. Moreover they stop trusting the system to handle the tasks to which it is best suited and spend even more time checking orders than they did before. The result: many of the gains from automation are lost.

So let’s stop to consider for a moment how automation works and, by extension, how computers work. In the old days (ie pre 1980 – the dawn of the PC era) many computer processes were referred to as routines. They would perform the same task over and over.

Computers are great at repetitive tasks. They don’t get bored. They’re also great at handling vast amounts of data. Complexity doesn’t faze them. To a computer your sales data is just a heap of 1s and 0s. It’s like ice cream to a five year old.

So if there are tasks that get done again and again and require reams and reams of figures to be analysed leave it to your systems.

So what are people good at? Other people, for a start. People manage computers way better than computers manage people, and people manage other people very well. Why? Because our survival has depended on it since the dawn of humanity. People also manage exceptions very well. Our brains have developed to spot differences and changes. It’s why you can spot the fact that someone has been into your room and moved one thing even though the other thousand things in there haven’t been touched. It’s a survival skill. Handling repetitive processes isn’t.

Deciding which routines to automate then becomes an assessment of the extent to which a particular process repeats itself and how easy it is to automate in an intelligent way. That leads to a cost benefit comparison between the effort and expense of automation against the effort and expense of manually repeating that task time and time again.

A good ordering system should take into account a variety of exceptional situations, such as campaigns and seasons. In practice, situations will still arise regularly where the experience and expertise of managers is needed. Some things are hard to predict: sales of new products, sales resulting from a campaign or a seasonal spike a few weeks earlier or later than in previous years. There are numerous external factors, such as weather, one’s competitors, the changing market or the economy. Managing large numbers of products also invites the possibility of data errors and these can lead to considerable problems if not dealt with quickly.

By putting in place an efficient exception management process, the attention of senior staff time can be refocused on the management of the most important products and situations: new products, campaigns, large orders (by volume or cost) according to the needs of the specific sector and the individual business. Most products don’t need constant attention, and they can and should be marked for automated replenishment. It is also good to bear in mind that not all products are equally important, and less important lines should command less attention.

Management of exceptional situations should also extend beyond the actual ordering process. With the right tools you can keep tabs on the reliability of each supplier’s deliveries. That helps you react to order delays quickly. Selection for each shop can be constantly refined by dropping lines with a poor turnover rate, and shelf availability can be improved by checking availability issues regularly.

Five Steps to Better Exception Management:

1. Prioritise the most important exceptions

It is easy to come up with a long list of exceptional problem situations. But monitoring too many can make it impossible to concentrate properly on any of them. That eats up time and produces bad results. Choose a few areas critical to your business that require action, and concentrate on them first.

If, for example, high inventory levels and low inventory turnover cause your company problems, prioritise reducing dead inventory and stopping over-large deliveries. You can leave the fine tuning of shelf availability for later.

2. Define corrective measures for each exceptional situation

Exception management is distinguished from normal reporting by the fact that exceptions should always lead to corrective measures. Continuously monitoring shelf availability, the accuracy of forecasts or your dead inventory does not of itself improve these key figures. However by determining the right response to each specific exception, work can be organised efficiently and measurable improvements can be achieved.

For instance, a shop can shift dead inventory by putting it out at sale prices or by transferring it to a location where it will still sell. At the same time you need to check that the store’s selection is up to date and that deliveries of problem products are stopped. There is no single ‘right solution’; your job is to determine the best response for your line of business, for the specific product and for the situation in hand.

3. Automate wherever possible

You may be dealing with a wide range of exceptions, but you may find that your responses fall into a pattern. When response measures are straightforward and the same every time, these can, at least in part, be automated. When exceptions occur the situation may prompt one of your establish corrective responses. These can be put to the manager responsible for fast-tracked approval. The efficiency gains free up yet more time to concentrate on more difficult cases that cannot be automated.

For instance, amongst the standard responses to dead inventory could be a search of alternative outlets to see where the non-moving items sell best, transfer to those outlets and an instruction to the system to change the order model to stop further deliveries of that item to the store.

4. Make sure that the number of exceptions is manageable

The point of implementing an efficient exception management process is to improve results by allowing key staff to focus where their attention is needed. Yet one of the most common problems is that the number of exceptions becomes unmanageable. For example, there can easily be hundreds of products with a zero balance at any given shop. If one employee is in charge of ordering for ten different stores then he or she suddenly has several thousand products to track. It effectively becomes impossible to take any concrete measures.

The number of products that staff need to oversee must be realistic and manageable. If you specify an exception criterion – such as ‘slow-moving stock’, or items that have not sold at all for more than two months but that are still in the inventory – that pulls up too many result for staff to manage properly, then you should refine the criteria to highlight the most important cases. For instance you could specify; ‘only premium products with a high collective stock value’, or ‘products with the slowest turnover rate’ to reduce the number of cases that need staff attention to a level at which colleagues can take effective action.

5. Continuously develop the exceptions being managed

By managing exceptions regularly the process develops and becomes more efficient. The number of exceptions decreases. This is not a cue to rest on one’s laurels. Rather it’s an opportunity to bring new exceptions under direct management or to broaden the management of old ones. In this way exception management becomes part of the everyday process of making your business more efficient.

In the beginning some businesses get excited and try to define all kinds of exceptions and corrective measures for them. This rarely works. It results in a workload that nobody can manage. One sitting is not enough time to come up with everything essential. Instead, exception management and its development should become a natural part of your ongoing business processes.

Dead inventory can be cleaned out by starting, for example, with the products which have sufficient stock to last for 180 days. When such products are continuously cleared out, and when selection management is used to ensure that they are no longer ordered, the number of exceptions is sure to decrease. In such circumstances you might then choose to decrease the stock limit to 120 days, or you might re-focus on availability issues.

Correct Problems Before They Happen

It’s all too easy to find yourself managing exceptions reactively – responding to problems that have already occurred; shelf availability issues go unnoticed until shelves are empty, dead inventory is only picked up after a product has gone unsold for a long time. These kinds of exceptions can be dealt before they become an issue, and in an ideal world, problems would thus be avoided almost entirely. This shouldn’t increase your workload – rather you become proactive, addressing issues when they are potential rather than actual problems. Of course, this requires support from good tools, so that the appropriate exceptions can be readily selected for management.

For example, in some cases stock-outs can be tackled before they happen. If a certain product is normally delivered once a week, the availability of the product can be monitored between delivery dates to see if there enough stock to last until the next delivery, if sales correspond to predictions. If there is an unexpected spike in the demand and a line threatens to run out, more stock can be ordered. That might mean scheduling additional deliveries or changing safety stock levels. You should not be aiming to do this for all products, but, for the most important ‘A-products’, you may decide it’s wise to ensure that the shelf is never empty.

Written by

Tommi Ylinen

Chief Product Officer