In this article, learn about:
The basics of forecasting
The difficulties of forecasting with a new retailer
How to ensure that your forecasting is as accurate as possible while you collect the necessary data
Getting that first "yes" from a major retailer is a milestone every brand dreams of. But once the initial excitement of the partnership fades, a daunting reality sets in: you have to actually fulfill those orders. The most critical, and often most terrifying, part of this process is the demand forecast.
Forecasting in general is hard enough. Forecasting for one you’ve never sold to is a different problem entirely. Most suppliers take their existing forecasting methodology, which is usually a blend of historical sell-through, seasonality, and a planner’s "gut feeling,” and apply it to the new partnership. The result is almost always a guess dressed up in spreadsheet formatting.
This way of forecasting (called “cold start” forecasting) often leads to one of two outcomes: the supplier overcommits inventory and absorbs massive carrying costs, or they underdeliver against the initial orders and burn trust with the retailer in the first 90 days. To succeed, you need to move away from trend extrapolation and toward a scenario-based discipline.
Related Reading: Forecast Accuracy and Forecast Bias
Why the Standard Forecast Fails the Cold Start
When you enter a new retail relationship, your usual baseline is gone. Sell-through trends don’t exist, seasonality is purely hypothetical for that specific channel, and your planner’s gut is calibrated against benchmarks that may or may not be accurate.
The penalty for being wrong has also never been higher. With inflation raising carrying costs and retailers enforcing stricter compliance (like OTIF) and chargeback programs, the penalty brands pay for a bad forecast are wider than they were just five years ago. If you treat this like any other forecast, you are setting yourself up to either miss shipments or drown in excess inventory.
Anchoring Your Forecast Without Historical Data
Since you don’t have any history at this retailer, you must look for external data sources to reduce uncertainty. Here is how to anchor a forecast when historical sell-through doesn't exist:
Analog brands: Look at how comparable brands in your category performed at this specific retailer during their first 12 months.
Category and channel data: Leverage syndicated panel data, retail media data, or public POS data to understand the broader movement in the category.
Retailer-shared planning: While the buyer's initial order quantity is a data point, it is not the whole story. These numbers are usually calibrated to the retailer's risk tolerance or a launch-quantity convention rather than an actual demand prediction. Treat the initial PO as one data point, not the forecast itself.
Existing account data: Use your own DTC or existing wholesale data as a directional input but be careful; the conversion rate to a new, major retailer is rarely 1:1.
Related Reading: Why More Supply Chains Are Pairing Forecasting and Demand Sensing
The Scenario-Based Framework
Rather than producing a single point estimate number, forecasting requires building a range of scenarios. This allows you to calibrate your spread to the actual level of genuine uncertainty you face.
The low case: This is your protection scenario. It answers the question: What happens if the launch is slower than expected?
The base case: This is your primary production and inventory plan, anchored in the best available analog data.
The high case: This is your success scenario, that helps you answer the question of what if your product goes viral or the retailer's footprint expands faster than planned?
You should calibrate these scenarios based on the retailer. For example, a brand entering its tenth Whole Foods store has much narrower uncertainty than a brand entering its first Walmart. Each scenario should have explicit assumptions attached to it so you can test them as real data begins to roll in.
The First Six Months: A Deliberate Learning Period
It’s impossible to be perfectly accurate with your initial forecast. Rather, the goal of your initial forecast should be to help you be wrong in the deliberately right direction. This means structuring the first three to six months as a learning period that updates the forecast as you go.
What does deliberate learning look like in practice? It requires:
Real-time data feeds: You need immediate visibility into POS sell-through, store-level distribution, and velocity by week.
Strict review cadences: Set weekly reviews for the first eight weeks, then shift to biweekly.
Specific triggers: Define ahead of time what velocity thresholds will trigger a forecast update or an inventory reallocation.
Building Operational Buffers
Your forecast is not living in a vacuum; it affects your production and supply chain strategy. To absorb the inevitable miss in a new launch, you must design your operations to be flexible.
This involves creating inventory buffers calibrated to your scenario spread and maintaining production flexibility. This might mean smaller batch commitments, staging raw materials rather than finished goods, or having contingency capacity ready to go. While these buffers and flexibilities cost money, that cost is simply the price of forecasting under genuine uncertainty. It is an investment in your long-term relationship with the retailer.
Managing the Success of the Launch
Getting the forecast right is the foundation upon which everything else stands or falls, from EDI compliance to item setup. By moving to a scenario-based model and treating the first 180 days as a data-gathering exercise, you protect your margins and, more importantly, your reputation with your new retail partner.
The cold start is one of the highest-stakes moments for any brand. Don't leave it to a guess. Use the data available, build for multiple outcomes, and be ready to pivot the moment the real numbers hit your dashboard.
Unlock the Power of Network Intelligence with MAX
As you navigate the critical first months of a new retailer launch, visibility is your most valuable asset. MAX by SPS is an AI-driven supply chain agent designed to help you turn that learning period into a competitive advantage.
By drawing on proprietary insights from over 300,000 trading relationships and billions of transactions, MAX Chat and MAX Monitor can spot patterns that signal potential issues before they become chargebacks or rejected shipments. Whether you need to automate routine tasks or gain direct access to network-wide intelligence, MAX helps you resolve problems in real-time, ensuring your first 90 days lead to a lifetime of retail success.
