Demand Planning Under Uncertainty: How Mid-Market Companies Are Breaking Free From Gut-Feel Forecasting

The operations director for a mid-market food producer once told me his demand plan for 2022 was finalised in a single three-hour meeting. The senior sales team reviewed last year’s numbers, added a bit for growth, tweaked for a big new retail contract, and sent the spreadsheet to production. It was simple. It was familiar. And six weeks later, it was worthless. An energy price shock doubled his packaging costs, a key ingredient supplier was hit by a cyberattack, and the new retail contract was delayed by logistics snarls. His carefully crafted annual plan, based on decades of collective gut-feel, was shredded by a reality it was never designed to handle.

The Great Fracture

The period between 2022 and 2025 did not just disrupt supply chains; it fractured the logic that underpinned decades of operational planning. Volatility is now a structural condition, not a temporary exception. Geopolitical conflicts, from Eastern Europe to the Red Sea, have rerouted global trade and injected persistent uncertainty. The old rhythm of annual budgets and quarterly forecast reviews is too slow for a world where freight costs can spike overnight and supplier lead times change weekly.

Gut-feel forecasting, the art of seasoned managers using their experience to predict the future, was the default for a reason: it worked when the past was a reasonable guide to the future. But when the underlying patterns break, experience becomes a liability. It anchors you to assumptions that are no longer true. Across industries, average demand forecast accuracy hovers between a dismal 50-70%, according to the Institute of Business Forecasting. That’s not a stable base for committing millions in working capital.

The companies that navigated this period successfully didn't have a better crystal ball. They had a better process. They abandoned the illusion of the perfect annual forecast and embraced a faster, more frequent planning cadence.

From Monthly Ritual to Weekly Cadence

The brief for this article was simple: the companies that survived had a data model that updated weekly. This isn't just about running the numbers more often. It’s a fundamental shift in mindset. A monthly Sales & Operations Planning (S&OP) cycle, long considered best practice, is now dangerously slow. It encourages teams to save up problems for a big monthly meeting. A weekly cadence forces a continuous process of sensing and responding.

What does this look like in practice? It’s not a four-hour weekly executive review. It’s a series of short, sharp, data-driven check-ins. A demand review on Monday, a supply and constraints check on Tuesday, and a reconciliation meeting on Wednesday to frame decisions for leadership. The goal is not to re-plan the entire business every seven days. The goal is to spot deviations from the plan early and make small, corrective adjustments before they become major crises.

The Signals Are Already Inside Your Business

This faster cadence is impossible if your data is scattered across a dozen spreadsheets and disconnected systems. The signals you need to make better short-term forecasts are almost always already inside the business, trapped in departmental silos. A truly responsive forecast doesn't just use historical sales data. It fuses it with live operational signals:

• Deal Pipeline: Your CRM isn't just a sales tool; it's a demand sensor. The weighted probability of deals in your pipeline is a powerful leading indicator of future orders. When deals accelerate or stall, your forecast should adjust automatically.
• Marketing Engagement: Are prospects suddenly downloading a whitepaper on a specific product? Are webinar sign-ups for a certain feature spiking? These are weak signals that, in aggregate, can predict a shift in demand weeks before orders are placed.
• Service Tickets: A sudden rise in customer service tickets about a product nearing its end of life can be an early warning of a drop-off in demand. Conversely, pre-sales technical questions can signal buying intent.
• Supplier Lead Times: The lead time in your master data is often a fantasy. The real lead time is the one quoted on your latest purchase order. Tracking this variance in real-time provides a much more accurate picture of supply constraints.

These signals exist in every company. The problem is that they live in different systems that don't talk to each other. The sales forecast is in the CRM. The marketing data is in a separate platform. The purchasing data is in the ERP. By the time someone manually extracts, cleans, and reconciles this data, the opportunity to react has passed.

The Unavoidable Technology Question

You cannot run a weekly planning cycle on a monthly data architecture. It's that simple. The pre-requisite for this kind of operational agility is a single, unified data model. When your CRM, ERP, and supply chain modules share one database, the data reconciliation problem disappears. A deal marked 'Closed-Won' in the CRM can immediately inform the demand plan, which in turn checks inventory levels and flags potential constraints — all in the same system, all in near real-time.

This isn't about buying more technology; it's about committing to less, but more integrated, technology. The challenge for many mid-market firms is that they have been sold a patchwork of 'best-of-breed' applications that require brittle, expensive integrations to function. The latency built into these architectures makes a high-frequency planning process impossible. A platform like Response365, where every business function from sales and purchasing to manufacturing and finance operates on a single data source, isn't a nice-to-have. It's the enabling infrastructure for the speed that modern markets demand.

The biggest challenge in forecasting during supply chain volatility is not just unpredictability – it's reacting too slowly or overcorrecting. Volatile conditions amplify small errors, making agility and continuous updates more important than perfect accuracy.

From Prediction to Reaction

The pursuit of the perfectly accurate forecast is a fool's errand. In a volatile world, no algorithm can predict the next geopolitical shock or container ship blockage. The objective has changed. The new competitive advantage is not predictive accuracy, but reaction speed.

A forecast is no longer a static number to be hit. It is a live hypothesis to be tested against reality every single week. The weekly S&OP cycle is the mechanism for that test. It asks four simple questions: What did we expect to happen? What actually happened? Why was there a difference? And what are we going to do about it?

Companies that build the process and systems to answer those questions every week are the ones who will break free from the cycle of reactive firefighting. They are the ones who will turn uncertainty from a threat into an opportunity.

---