From Reactive to Predictive: The AI-Driven Demand Forecasting Evolution

Demand forecasting is the engine of modern retail strategy, powering decisions from product planning to final markdowns. Retailers and brands can no longer afford to guess what consumers want.

Outdated demand forecasting methods built purely on historical data and instinct, crumble under today’s pace of change. AI-powered forecasting flips the script, helping brands get ahead of demand with accuracy that drives revenue, optimizes inventory and reduces risk. That’s because an inaccurate forecast can lead to expensive consequences.

Every missed forecast means lost revenue, excess inventory or margin erosion, costs that add up quickly in today’s volatile environment.

What changed? Shopping habits are now digital-first. Consumer preferences shift overnight. And there’s a flood of data from every channel. No human alone can process this complexity at scale. That’s where AI-powered demand forecasting comes in. It analyzes billions of data points, from social sentiment to sell-through trends, to predict what will sell, when and where. Retailers gain next-level clarity on:

• What products will perform by style, size and color.
• Which channels and stores need what assortment/inventory.
• How to time pricing and allocation decisions.

And the impact goes beyond the numbers. AI is transforming the roles of merchandise planners, buyers, allocators and pricing teams—freeing them from manual work and enabling strategic, real-time decisions.

This eBook dives deeper into:

• 5 key demand forecasting roles and their challenges.
• The impact of AI on demand forecasting.
• Role-specific situations and use cases for success.

Demand forecasting is the backbone of any modern retail planning ecosystem, where retailers and brands anticipate consumer needs to stay in line with future buying behavior. Inaccurate forecasting can lead to costly outcomes, from excess inventory to missed sales and an overreliance on markdowns.

Not long ago, forecasting demand meant relying on spreadsheet formulas, gut instinct and the mental math of a single seasoned planner—often working in isolation and with limited data. Today, retailers and brands are overwhelmed with data streams from multiple channels and rapid market shifts, which are faster than any individual—or spreadsheet—can keep up with.

Forecasting fatigue is real.

One person, one spreadsheet and thousands of SKUs? That’s not scalable. AI removes the guesswork, so teams can focus on strategic decisions—not data wrangling.

Consumer behavior has also changed dramatically. Shoppers no longer rely solely on physical stores to discover, evaluate and buy products. Today, discovery happens through clicks, scrolls and social feeds—shifting demand signals to digital.

That’s where artificial intelligence (AI) steps in. With AI-driven demand forecasting, retailers and brands can anticipate what will sell next month, which sizes and colors will be most popular and even what key messaging will resonate with specific audiences.

As a result, the roles of merchandise planners, allocators, buyers, sales and pricing teams are evolving. These teams are moving beyond spreadsheets and manual updates to make faster, more strategic, data-driven decisions.

Forecasting is the foundation. AI is the accelerator.

Accurate, AI-powered forecasting connects every planning function—from pre-season buy planning to in-season replenishment and end-of-life pricing. It enables SKU-level precision, faster response times and confident decisions across every channel.

With AI, the roles involved in demand forecasting are transforming. In the past, demand planners would analyze summary-level data, total sales by category, channel or region, occasionally drilling down to SKU-level data for replenishment or exceptions.

This limited, structured view of data only scratches the surface. Today, AI and machine learning can analyze massive data sets across multiple dimensions including by channel, region, product attributes and even consumer behavior patterns, surfacing insights no human alone could uncover.

What was once a manual, reactive process is now automated, proactive and continuous. AI surfaces high-, mid- and granular-level recommendations across short, medium and long-term horizons, aligned to margin goals, revenue targets and operational KPIs.

Demand forecasting roles are entering a new era, defined by smarter, faster decision making across every team.

This shift empowers teams with greater control and productivity. Even high-performing businesses are investing in AI forecasting to supercharge decision making and boost team output across locations, departments and roles.

Watch-out.

Despite the upside, some brands or retail organizations still fall into common traps—like relying too heavily on outdated spreadsheets, or underestimating the need for clean, aligned data inputs.

 1. Merchandise Planner/Financial Planner
Owns the seasonal and long-term planning process, setting sales, margin and inventory targets. Oversees product category performance and supports assortment strategy decisions.

Key Challenges:

• Must forecast demand without complete historical data or finalized assortments.
• Lacks time to assess potential at the style-color-size (SKU) level.
• Needs to align demand plans with financial goals and operational constraints.

 2. Buyer/Merchant/Assortment Planner
Selects and purchases the seasonal assortment across categories, channels and geographies. Collaborates with planners and allocators to ensure product relevance and commercial success.

Key Challenges:

• Faces pressure to localize assortments by store cluster or region.
• Must balance depth and breadth of buy to avoid over- or under-stocking.
• Limited ability to forecast demand for new or seasonal styles with no sales history.

 3. Allocator/Inventory Manager/Replenishment Planner
Manages inventory flow to stores and channels, oversees replenishment cycles and ensures the optimal stock levels with the right stock in the right place at the right time.

Key Challenges:

• Balancing stock across diverse store formats, regions and product types.
• Reacting to sell-through without a real-time view of evolving demand.
• Managing transfers and reorders within tight time, space and budget limits.

 4. Sales Representative/Wholesale Manager (brand side)
Drives revenue through wholesale accounts. Oversees seasonal bookings and co-plans assortments with retail partners.

Key Challenges:

• Low visibility into in-season demands and sell through at the partner level.
• Needs to forecast orders before bookings are confirmed (“blind buys”).
• Competes in a promotion-heavy market while protecting full-price margins.

 5. Pricing/Lifecycle Management team
Manages and owns the pricing strategy from launch through to markdown and clearance. Manages margin optimization, lifecycle tracking and inventory sell through.

Key Challenges:

• Must balance price, margin and inventory risk in volatile conditions.
• Struggles to time and scale markdowns optimally across channels.
• Lacks real-time demand signals to adjust pricing dynamically.

Instead of looking back at last year’s historical data or last season’s trends to determine what “I think I should sell”, AI ingests a myriad of data—including consumer reviews, browsing histories, purchase patterns, social media sentiment and more—to anticipate consumer needs and retail preferences and forecast sales “with astonishing accuracy.”1

In short, AI-driven demand forecasting is evolving the role of merchandise planners, allocators, buyers, etc., enabling those teams to look at the business more holistically.

Additionally, these teams can look at specific situations or use cases, run different scenarios and remain more data-informed—to keep their businesses more aligned to current trends, what’s being bought and consumed by their consumers and the market.

Here’s a breakdown of how AI demand forecasting supports the:

Merchandise Planner/Financial Planner:

• Delivers accurate pre-season department/channel forecasts right down to SKU-level forecasts, even for new products.
• Provides financial guardrails for planning investments.
• Enables scenario modeling to guide strategic and operational buy decisions.

Role evolution: From spreadsheet-heavy number cruncher to strategic decision maker. AI frees this role to focus on high-value planning and cross-functional alignment.

Buyer/Merchant/Assortment Planner:

• Models the demand potential of new items using advanced visual or attribute similarity reference modeling.
• Identifies optimal buy quantities by channel or store cluster.
• Guides initial assortment decisions with demand-backed insights.

Role Evolution: From intuition-led buying to data-validated assortment curation. AI strengthens creative decisions with predictive confidence.

Allocator/Inventory Manager/Replenishment Planner:

• Forecasts demand at the store x SKU level to guide precise allocations.
• Enables dynamic in-season adjustments and inventory transfers.
• Integrates space, sales velocity and location-specific trends.

Role Evolution: From reactive firefighting to proactive inventory controller. AI empowers smarter decisions across the supply chain network.

Sales Representative/Wholesale Manager (brand side):

• Projects full-season sales based on early bookings (“blind forecast”).
• Reduces stockouts with improved demand visibility by customer/channel.
• Enables smarter stock commitments that protect full-price sales.

Role Evolution: From reactive seller to proactive partner. AI helps sales teams provide value-added insight to retail partners and improve seasonal performance.

Pricing/Lifecycle Management team:

• Uses demand signals to set optimized launch prices.
• Adjusts prices dynamically based on real-time sell through and inventory.
• Simulates pricing impact on KPIs across locations and channels.

Role Evolution: From static pricing operator to lifecycle strategist. AI enables more responsive, profit-protecting price management across the season.

From the first buy to final markdown, AI-driven demand forecasting boosts accuracy, speed and alignment across every team and decision.

1 Gary Drenik. “AI Shifts Decisions From Reactive To Predictive Power,” Forbes

Centric Software® (centricsoftware.com)

From its headquarters in Silicon Valley, Centric Software provides an innovative and AI-enabled product concept-to-commercialization platform for retailers, brands and manufacturers of all sizes. As experts in fashion, luxury, footwear, outdoor, home, food & beverage, cosmetics & personal care as well as multi-category retail, Centric Software delivers best-of-breed solutions to plan, design, develop, source, comply, buy, make, price, allocate, sell and replenish products.

• Centric PLM™, the leading PLM solution for fashion, outdoor, footwear and private label, optimizes product execution from ideation to development, sourcing and manufacture, realizing up to 50% improvement in productivity and a 60% decrease in time to market.
• Centric Planning™ is an innovative, cloud-native, AI solution delivering end-to-end planning capabilities to maximize retail and wholesale business performance, including SKU optimization, resulting in an up to 110% increase in margins.
• Centric Pricing & Inventory™ leverages AI to drive margins and boost revenues by up to 18% via price and inventory optimization from pre-season to in-season to season completion.
• Centric Market Intelligence™ is an AI-driven platform delivering insights into consumer trends, competitor offers and pricing to boost competitivity and get closer to the consumer, with an up to 12% increase in average initial price point.
• Centric Visual Boards™ pivot actionable data in a visual-first orientation to ensure robust, consumer-right assortments and product offers, dramatically decreasing assortment development cycle time.
• Centric PXM™, AI-powered product experience management (PXM) encompasses PIM, DAM, content syndication and digital shelf analytics (DSA) to optimize the product commercialization lifecycle resulting in a transformed brand experience. Increase sales channels, boost sell through and drive margins.

Centric Software’s market-driven solutions have the highest user adoption rate, customer satisfaction rate and fastest time to value in the industry. Centric Software has received multiple industry awards and recognition, appearing regularly in world-leading analyst reports and research.

Centric Software is a subsidiary of Dassault Systèmes (Euronext Paris: #13065, DSY.PA), the world leader in 3D design software, 3D digital mock-up and PLM solutions.

Centric Software is a registered trademark of Centric Software, Inc. in the US and other countries. Centric PLM, Centric Planning, Centric Pricing & Inventory, Centric Market Intelligence, Centric Visual Boards and Centric PXM are trademarks of Centric Software, Inc. All third-party trademarks are trademarks of their respective owners.