The planner – the person who matches demand to supply – has traditionally been seen either as an absolute whizz with spreadsheets or a super-boffin capable of fathoming the mysteries of computer packages which baffle even the most techno- savvy.
Planning software has tended to sit, brooding, on standalone PCs and has been approached in silence and in awe. But not any more. Just as web-based system have brought slick analytics to management desktops, so it is with planning. Systems are becoming more adaptable and suitable for use by a far wider range of purchasing and supply chain staff. They’re also appealing to smaller and mid-tier businesses, not just the traditional giants capable of supporting both the investment and the long-term staffing needs.
The days of what AMR Research terms BASS planning applications (Big Awkward Spread Sheets) belong firmly in the past as smaller companies embrace new planning technology.
‘One of our most recent customers is Poundland,’ says Andrew Fowkes, director of operations at Marketmax, which was acquired in October by SAS. ‘The company realised that it simply couldn’t do what it needed to plan effectively by using spreadsheets anymore. Like other new customers they’re starting with a simple application and will gradually develop this over time.’
Marketmax offers a range of planning tools including financial and assortment planning but in Europe most of its retail customers start with space optimisation; matching the available shop floor and stockroom capacity with forecast demand and range options. ‘Space is the big constraint,’ says Fowkes, ‘there are limitations on opening new stores so we’re seeing businesses trying to make the most of their available selling areas.’
SAS is currently evaluating how it will integrate the Marketmax planning tools into its wider business intelligence offer. ‘Our demand forecasting systems are largely driven by events,’ says Richard Hey, retail strategist at SAS, ‘They model how promotions, seasonal activities or the weather will impact sales. We’re working on predicting the impact of these events on the supply chain and the Marketmax planning systems will fit in well.’
Other companies, such as Compass Software – acquired this summer by Alphameric – are developing easy-to-use tools which are transforming the way merchandise teams assemble ranges, plan assortments and allocation, and strive to adapt supplies to demand in real time. Its Smartdecision system – launched formally at Retail Solutions 2003 last June – is a .Net planning tool that has been developed in association with Marks & Spencer (which also pioneered Marketmax’s Space Optimiser package).
Microsoft’s .Net environment not only enables mix-and-match software development but also ensures flexibility in the way systems can be deployed so that staff working from home or on buying trips using laptops can still access the planning capability in a functionally-rich environment.
‘Merchandising planning systems traditionally run as a thick client,’ says Simon Bullock, head of IT solutions (general merchandise) at M&S, ‘but they need to evolve and give greater flexibility. Our move to three head office locations meant that we needed a technically flexible planning system that would not be restricted to particular desks or PCs. It was a real challenge for Compass to create a product that could do this.’
The system uses web services technology so that users can download additional tools and components as needed. And it does not need a powerful PC to run functions such as rate of sale calculations or to model likely demand against various weather scenarios. Buyers in the Far East are able to directly input make-up costs and production details to ensure collaboration with head office teams for range planning. Smartdecision incorporates a workflow element guiding planners through the steps needed to bring a range to market with alerts for any deviations from plan. It also enables collaboration with suppliers.
‘The .Net environment means that we can deploy components of the system once and they can then be used many times in different formats by users as needed,’ says Ian Bowater, chief executive of Compass. ‘It also means that the system is agile so as to keep pace with changing requirements and can adapt to future planning needs.’
Similarly focusing on the wider use of easy planning systems, although with a rather different application, is Per Una – the highly successful own-label operation developed by George Davies for M&S. Designers and merchandisers here are using two Buyerplan modules – Range Planner and Merchandise Planner – from Decision Systems. These applications have also been integrated with Marks & Spencer’s internal applications for contract management and also with a Mercatus Open system from NSB running on Compaq servers and used by Per Una to run merchandising, stock management, distribution and logistics systems.
Range Planner is used to prepare product budgets and process the first styles from a line plan while Merchandise Planner is used to assess the initial sales and calculate additional stock requirements should demand prove higher than forecast. The systems were implemented within a couple of months of starting Per Una operations two years ago when it became clear that the traditional mix of spread sheets and word processed documents, each containing product photographs, would be inadequate for such a fast moving operation.
‘We work on very tight lead times,’ says Per Una buying director, Emma Trayner. ‘It is usually only eight weeks between design and shipping stock into stores and in that time we have to sample the style, go through the range selection process, make and ship.
At any one time Trayner and her team can look at the new lines in the development pipeline, complete with photographs of samples and styles, to gain a quick overview of progress. This integration of pictures, text and numbers has saved significant time when it comes to discussing new lines with M&S. Costs have been cut, too, since there is no longer a need for style sheets to be prepared by graphic designers for M&S meetings. Importantly, the Per Una team can access the planning information via the internet from wherever they happen to be – in the office, at home, or from any of the company’s hubs in Italy, Hong Kong or Turkey.
Italy-based TXT e-Solutions is another specialist in the high fashion sector, providing forecasting tools to the likes of Gucci, Prada, Benetton, Lacoste, Jean Paul Gaultier and Moschino Cheap and Chic. TXT – which set up operations in the UK two years ago – is partnering there with Prima Solutions, a supply chain specialist whose customers include Jockey, Panache Lingerie and Jaeger. TXT has developed a forecasting tool which enables clothing retailers to predict likely demand even when there is little or no sales history for comparable lines.
‘This is a major benefit for fashion retailers,’ says Stefano Lena, TXT’s international sales manager. ‘We are already working with a UK retailer and the product is proven. If you just have one or two days of EPoS data then we know this can give a 98 per cent improvement in forecast accuracy with potential stock reductions of 25-30 per cent.’
TXT’s customers tend to be vertical businesses involved in both manufacturing and retailing although Lena believes that the new forecasting system is also relevant for chains with from 30-40 branches. Over the past couple of years TXT has been ‘modularising’ its product to create specific packages for vertical markets. Many customers are active in collaborative supply chain initiatives and TXT believes that its planning applications are especially relevant here. The company signed a marketing deal with Microsoft in the summer and the latter is now promoting the Demand Planning system as part of the Microsoft Business Solutions suite with a simplified version integrated into the Microsoft ERP offering.
While the planning systems themselves are becoming more user friendly, they are also – as with Compass’s Smartdecisions and the Marketmax/SAS plans – integrating more widely with execution functions across the supply chain. By tracking the implementation of the plan with alerts for any deviation, systems can help improve operational efficiency.
Demand chain optimisation systems are also increasingly being linked with more conventional supply chain systems. Optimisation is fast becoming one of the most fashionable buzzwords with talk of price optimisation, promotions optimisation, markdown, replenishment and space allocation optimisation functions. Many of the IT vendors offering these products are highly specialist – such as Profitlogic with its markdown optimisation systems, or KhiMetrics and Demandtec focusing more on the price/promotions space. Others are emerging from mainstream players: Retek, for example, has launched a markdown optimisation tool while i2 Technologies – now back on a more secure footing with reaudited accounts – has moved into price optimisation.
i2 acquired the Makoro planning suite from IBM three years ago. This is now rebranded as Merchandise Planner and integrated with Demand Planner which acts as the forecasting engine and Replenishment Planner which manages inventory levels from suppliers through warehouses and distribution centres to match actual stock levels with forecast demand in a fully integrated system. On top of this i2 has added its new optimisation tools; the set includes a Category Price Optimisation function, developed initially with Spanish supermarket chain Caprabo, which adds price optimisation attributes such as margin pricing and competitor’s prices into the equation.
US customers include Payless Shoes which has added price optimisation as part of its demand planning toolset: ‘Payless is an established i2 customer,’ says Lee Gill, retail industry director for EMEA, ‘They see this sort of optimisation as an adjunct of forecasting. Using optimisation techniques the company is making better chain-wide decisions and believes it can save $20million a year by improving over-all gross margin.’
But while markdown optimisation is currently a hot topic on both sides of the Atlantic, Gill believes that it is a solution that should not be needed: If you get the forecast tight then there should be no need for markdowns to clear surplus stock,’ he says. ‘We are developing improved pre-season planning tools that will give greater forecast accuracy’ – an area of great interest.
NCR Teradata, too, is looking to tweak forecast capability with its Promotions Manager system which combines with its established Demand Manager. This application takes basic EPoS data, uses algorithms to smooth anomalies, and then produces a demand calculation matched to seasonal profile. Promotions Manager sits on top of this capability monitoring every product in every store to see the effect on overall category performance. It can calculate losses due to cannibalisation of sales and recalculate to produce a recommended promotional strategy which optimises performance at the margin.
‘In the US J C Penney is using both the Demand and Promotions Manager tools,’ says Sean Cassidy, programme manager for demand chain solutions with NCR. ‘The system uses interactive search techniques to look at product performance and history and will produce an average for comparable activities which buyers can then manipulate further.’ The system also uses historical information about the effectiveness of different types of media campaigns on to the demand forecast.
A better forecast
A retailer, for example, might be planning a cut price teabag promotion. It can use the system to check how sales were affected by previous comparable schemes – such as how a local radio campaign for cut price coffee compared with one using newspaper ads – and this information can then be used by the demand application to give a better forecast of likely sales.
Early customers for the product include Austrian DIY chain Baumax which has already seen a 30 per cent reduction in store inventory by using the Demand Manager product.
‘There has been steep learning curve for buying departments,’ says Walter Hitziger, a member of Baumax’s executive board, ‘Before, they estimated order size but now they must accept what the system tells them and that is a big change.’
JDA, too, is looking at how promotional activity can be integrated into demand planning systems and is working on a new application called ‘event lift forecasting’ which takes information about stores, customers and previous results of promotions and builds this into a more precise and relevant forecast model.
‘Demand forecasts and assortment plans are often built on averages,’ says Peter Charness, senior vice president for global marketing and chief product officer at JDA, ‘but average is irrelevant. Only one store in a chain is ever likely to be average so it is much better to work with small clusters of four or six and model the likely impact of events on a clusterby-cluster basis.’
JDA acquired Engage – a marketing and digital management software specialist – in the summer. As this product is integrated into the wider suite it will help match local variables and collaborative inputs with demand and supply chain information to give better forecast accuracy. ‘We can integrate workflow functionality into the model as well as space optimisation,’ says Charness. The result, he argues, will be store-level plans and forecasts that are individually relevant and tailored rather than simply being a corporate average.
Integrated and flexible systems such as these – melding space and merchandise planning with financial performance and promotional activity – are already transforming supply chain activity. Unwieldy plans and BASS-dependence are rapidly being consigned to history.