Preseason planning is a complex process for any fashion business. Your financial, merchandise, and assortment plans must agree with each other, but changes you make at one level have side effects at other levels. Your overall plan and specific plans for each store and channel also impact each other. That’s what makes manual planning so tricky.
What makes manual planning so messy?
You might begin your plan at the financial level by setting budgets and revenue targets. Those decisions constrain your merchandise plan, including OTB budgets and channel splits. Similarly, your merchandise-level decisions place restrictions on your assortment plan, including how many units of each category and SKU you can order.
On the other hand, you might begin your plan at the assortment level and work your way up. For example, you could estimate how many units of each product you want to order for stores in each store cluster. Those estimates will help you set the range of each category, your monthly OTB budget, and so on. Your merchandise-level decisions will in turn help you set realistic revenue targets.
Here’s the challenge: when you walk through all three levels of planning in one direction, you might begin to see figures that are much larger or smaller than they should be. That forces you to adjust them manually and then walk through the levels in the opposite direction. The changes you made will cascade to the other levels and the numbers will usually start to look more reasonable. You might have to go back and forth with these adjustments several times before you’re satisfied with the compromises you make.
And that’s just one season! In Fashion, seasons often overlap, which makes your annual plan even more complex.
The “magic” of machine learning
Wouldn’t it be wonderful if you could take a form, fill in a few numbers that you’re sure are right, and magically receive a complete plan that works? That’s kind of what happens with machine learning… except it’s not magic.
A machine learning solution that can access your financials, supply chain details, and historical sales data can make planning much simpler. It knows how much you can invest in merchandise, how much each product costs, and how popular each of your products are in each of your stores. If you provide it a map of your stores and distribution centers, it can also calculate transportation costs and optimize delivery routes. Throw in a complete list of product attributes, and it can forecast sales for newly introduced products by drawing similarities with existing SKUs.
That’s not all that machine learning can do. It can segment your customers more precisely. It can suggest the right mix of expensive, mid-range, and budget items for each product category. It can tell you which product to offer as an alternative to something that’s out of stock. It can produce an optimized “shopping list” when it’s time to order more stock. It can keep an eye on social media to figure out what’s hot and what’s not. Census data, weather forecasts, stock market indices, and commodity prices are other types of information that machine learning can use to forecast demand and recommend ways to maximize your bottom line.
And it does all this on an ongoing basis. The moment something changes, machine learning will raise a red flag and let you know. If you’ve given it permission to place orders and make other changes, machine learning can fully automate many of your supply chain processes.
Each level of your fashion brand’s preseason plan is important, but you can finalize any of them independently of the others. Manual planning can take days or weeks, especially for large businesses. Using machine learning is much easier, quicker, and more accurate.
To learn more about how machine learning can make major improvements to your preseason and in-season demand planning, read our eBook, “Predictive demand planning across the fashion supply chain”.
About the author
Combining human centered AI with deep domain expertise, Algo’s analytics enriched supply chain intelligence platform helps suppliers and retailers plan, collaborate, simulate and execute a more efficient supply chain.