Demand-driven value networks sense and shape demand to deliver the profitable demand response. This is a stark contrast to traditional supply chains that match demand with supply, but are blind to profitability and true market demand. Companies that are demand driven, sense and align the value network for demand changes five times faster, and they are better at balancing growth and efficiency goals. But, how do they do it? What steps do they take?
1. Design for demand variability: Sounds simple, but it is not. Traditional supply-based network design techniques optimize costs, but do not account for demand variability in the design of networks. Companies that have done this well are Coors and Dow Chemical. If demand variability is not accounted for there are two primary issues:
1. Design for demand variability: Sounds simple, but it is not. Traditional supply-based network design techniques optimize costs, but do not account for demand variability in the design of networks. Companies that have done this well are Coors and Dow Chemical. If demand variability is not accounted for there are two primary issues:
a) Is the design feasible? The strategic optimization engines give an average response, that may or may not be a feasible response. Simulation tools help to see if the strategic network design technologies are giving a feasible response. For supply chains with demand swings, this is an important check.
b) Does it reflect the best design? For companies seeking to be demand-driven, what if analysis needs to include the probability of demand and the impact of demand ranges and patterns for important products. Demand variability is a strong determinant of push/pull boundaries. Currently, I only see 15% of companies including demand variability in network design calculations.
2. Focus on market-drivers: These leaders model the ship-to environment (not the traditional ship-from SKU-location) based on market drivers (weather, consumption drivers, consumer take-away, etc.) Great examples of companies making this transition are Cisco, Del Monte and DuPont. I loved Peter Murray, DuPont quote at the S&OP IBF conference in May in Chicago, “ The hardest change was changing to focus on market drivers. When we did this, our work on S&OP reversed a 12 year slide in margin. The downturn made us believers in strong horizontal processes at the executive level.”
3. Get clear on baseline forecasts. Increasingly, organizations are turning to supply chain to own the forecast. It has bounced around in organizations from function to function over the past ten years, but there is now a trend to return it to the supply chain function. Why? Marketing and sales groups tend to have a greater bias, and the supply chain group is seen as able to be a better litmus test of the true forecast. Best in class groups focus on understanding the baseline forecast–true market demand if there is no demand shaping (price, trade promotions, or marketing programs)– and become great modelers on true lift predictors. In calls, 95% of companies struggle with an accurate representation of baseline forecasting.
4. Take a holistic approach to demand shaping. For 40% of companies, price is managed separately than trade promotions, and new product launch has a significant percentage of trade promotion and marketing activities. Look at the impact of the multiple demand-shaping techniques through the use of strategic modeling tools like M-Factor, Thinkvine or the Henry model from IRI (now Symphony).
5. Move away from rules-based consumption: Traditional Advanced Planning Systems used rules to split a monthly forecast into smaller “buckets” for expected daily demand. However, supply chains are unruly. The spread of the forecast is hard to predict through fixed rules. As a result, 21 Fortune 500 companies have substituted statistical engines like those from JDA, Oracle, and Terra Technology to get a better predictor of daily demand. This has improved daily item forecast accuracy at a distribution center by at least 30%.
6. Give your Data a Scrubbing: Like a spring and fall cleaning, order and shipment data needs continual cleaning to align history, account for returns, and item changes. Clean data is an important foundation for demand-driven initiatives.
7. Where possible use Channel Data: Channel data helps to reduce demand latency (the time to sense true consumption in the channel). The more steps that a company is back in the value network, the more important to take this step.
No comments:
Post a Comment