The Stocking Problem That Costs HVAC Distributors on Both Ends
HVAC distribution has a unique and painful inventory problem: it loses money on both sides of the demand curve. During peak cooling and heating seasons, stockouts cost sales - a contractor who can't get a specific condensing unit from you calls the next distributor. Between seasons, overstock accumulates - equipment that didn't move during the peak sits in the warehouse depreciating, eventually sold at a discount or returned to the manufacturer at penalty.
The industry data confirms the cost. 60% of HVAC distributors report that inadequate seasonal stock compromises their sales. These are not small operational annoyances - they are direct revenue losses to competitors who happened to have the right inventory at the right time. At the same time, end-of-season inventory markdowns across the industry represent hundreds of millions of dollars in margin erosion annually.
The paradox is that HVAC demand is seasonal but not unpredictable. It follows regular patterns anchored to weather, construction activity, and equipment lifecycle. The distributors getting both ends right - stocked for the peak, clean inventory going into the off-season - are not using fundamentally different data than their competitors. They're using the same data differently.
The 2.1% Margin Case: What Actually Changed
A $75M regional HVAC distributor achieved a 2.1 percentage point margin improvement within 60 days of implementing a data-driven demand planning process. That's not a forecast improvement - it's a realized margin improvement that showed up in their P&L in two months. Here's how it worked.
Before the change, this distributor used a standard planning approach: take last year's seasonal sales by SKU, apply a growth factor, and place stocking orders accordingly. This produced two systematic errors. First, the growth factor was uniform across all SKUs, ignoring that high-efficiency models were growing share while older models were declining. Second, the approach anchored on the previous year's demand without adjusting for forward-looking signals - a cooler-than-average summer forecast, or a slowdown in new construction activity, would not reduce their stocking orders until after the sales miss occurred.
The change was three-pronged: they integrated 90-day weather forecasting for their service territory, adjusted stocking models by equipment tier (premium vs. standard efficiency) based on sales mix trends, and built in a rolling 30-day recalibration process during the selling season. The result: they entered the season with 18% less overstock on slow-moving models and 12% more availability on high-velocity premium equipment. End-of-season markdowns dropped by 40%, which translated directly to the 2.1% margin improvement.
The Leading Indicators Most Distributors Ignore
Traditional HVAC inventory planning relies primarily on lagging indicators: last year's sales, last quarter's turns, trailing 12-month demand by SKU. Lagging indicators are necessary but insufficient. They tell you what happened; they don't tell you what's about to happen.
Leading indicators that have high correlation with HVAC seasonal demand and are largely free to access:
Weather forecasting data: NOAA extended outlooks provide 90-day temperature and precipitation probability forecasts by region. A high-probability above-normal summer temperature forecast for your geography is the single most reliable leading indicator of cooling season demand. Services like Weather Trends International specialize in HVAC-specific demand impact forecasting and are widely used by large distributors.
Building permit data: Local and county building permit databases, available free through government portals, show new construction starts 3-6 months before equipment is needed. A surge in residential building permits in Q1 predicts increased new equipment demand in Q2-Q3. This data is particularly valuable for forecasting commercial equipment demand, which has longer lead times from permit to installation.
Contractor backlog surveys: Many HVAC trade associations publish quarterly contractor business surveys that include backlog data. A contractor community reporting growing backlog and increased hiring is signaling strong upcoming demand. A community reporting declining backlog is signaling a slower season ahead.
Utility cooling and heating degree day forecasts: Degree day projections correlate closely with residential HVAC service call demand, which drives parts and refrigerant consumption. Your utility company or a service like Degree Days.net provides this data free.
Equipment Mix Forecasting: The Segment Most Planners Get Wrong
Total demand forecasting - predicting how many condensing units you'll sell overall - is less valuable than getting the mix right. HVAC equipment mix is shifting structurally in ways that historical averages will systematically misforecast.
Three mix dynamics dominating the current market:
Efficiency standard transitions: The 2023 DOE transition from SEER to SEER2 standards phased out a significant portion of the legacy product catalog. Distributors who continued stocking legacy efficiency equipment based on historical demand patterns ended up with obsolescence write-offs. Upcoming heat pump incentive expansion under various state programs is creating a shift from gas heating to heat pump that will take 3-5 years to fully show in sales data.
Premium tier growth: Despite economic pressures, premium efficiency equipment (20+ SEER2) is growing share faster than standard efficiency equipment as contractor customer education about operating cost savings improves. Distributors weighting their buys toward historical mix (predominantly standard efficiency) are chronically understocked on premium equipment and overstocked on standard.
Regional climate adaptation: Extreme heat events in regions historically dominated by milder summers (Pacific Northwest, Mountain West) are driving demand shifts toward higher-capacity systems in markets that hadn't previously needed them. Distributors serving these markets need to adjust their capacity tier mix accordingly.
In-Season Inventory Recalibration: The Process Most Distributors Skip
The most underutilized demand planning practice in HVAC distribution is in-season recalibration: adjusting stocking orders based on actual sales velocity in the first 30-45 days of each selling season.
The concept is simple. Your pre-season forecast is your best guess based on available leading indicators. Once the season starts and real sales data flows in, that data either confirms your forecast or reveals deviations. A product selling 40% faster than forecast in weeks 1-4 will stockout unless you expedite additional inventory. A product selling 30% slower than forecast will create end-of-season overstock unless you reduce incoming orders.
In practice, most HVAC distributors don't have a formal recalibration process. Pre-season orders go in, and adjustments happen reactively - when a stockout is imminent or when a warehouse manager notices unusually full shelves. By the time reactive signals are strong enough to notice, the correction window is often too late to prevent either the stockout or the overstock.
A formal recalibration process sets explicit thresholds: any SKU with 4-week velocity 20%+ above or below forecast triggers an order adjustment review. This review happens at week 4, week 8, and week 12 of the selling season. The process takes 2-3 hours per review cycle and prevents the majority of end-of-season inventory problems.
Building the Demand Planning Infrastructure
Advanced demand planning doesn't require a dedicated data science team. It requires three infrastructure components that most regional distributors can build in 60-90 days:
First, a SKU-level demand history clean enough to build category-specific models. This typically means cleaning 24 months of transaction data to remove anomalies (COVID disruption years, one-time bulk orders, etc.) and ensure consistent SKU coding. This data cleaning step is often underestimated and is the primary cause of delayed implementations.
Second, a structured process for ingesting leading indicator data and translating it into forecast adjustments. This doesn't require sophisticated software - a spreadsheet model that takes weather forecast deviation from normal and calculates demand impact by category is a reasonable starting point. Purpose-built demand planning software (Slimstock, Blue Ridge, Logility) provides more sophisticated capabilities when you're ready for them.
Third, a recalibration cadence with explicit ownership. Someone needs to be responsible for comparing actual-to-plan during the selling season and triggering adjustments. In most mid-market distributors, this is a buyer or operations manager role with 4-6 hours per month of dedicated demand planning time. The discipline of the process matters more than the sophistication of the tools, at least initially.
All Tools