HOLY <BLEEP> : A Guide to Sales Forecasting
Sales forecasting gets a bad rap. People call economics “the dismal science,” but if you’ve ever sweated over a quarterly sales forecast spreadsheet, you know forecasting deserves that crown.
For many builders, it’s as nerve-wracking as public speaking. You dread it, you avoid it… until you’ve done it enough times to realize it won’t actually kill you. Like stepping on stage, the fear is mostly in the anticipation. The first time, your palms sweat. The second time, you’re still uneasy. By the tenth time, you start to notice patterns—and maybe even enjoy the craft.
But forecasting for a business, especially if you make tangible products, is more than a comfort-building exercise. It’s a critical function that ties directly to your cash flow, inventory, and operational efficiency. It’s the Goldilocks problem in business operations: not too much, not too little, but just right. And getting there is hard.
Even giant corporations with decades of experience, mountains of historical data, and advanced software tools routinely get it wrong. That’s why industries exist to deal with overstock—companies like Winners and Marshalls (TJX Group) or Liquidation World (LW Stores) profit from excess inventory created by forecasting misses.
So, if the pros still struggle, how do you, the builder of a small but growing business, approach forecasting in a way that’s practical, effective, and doesn’t drain your will to live?
Let’s start with the basics.
What Forecasting Actually Is
At its core, sales forecasting—also called demand forecasting—is making an educated bet on how many things your customers will buy in the future. You use that bet to place orders with suppliers, schedule production runs, or coordinate with a contract manufacturer (also called a co-packer) to make finished goods.
It sounds simple enough, but in practice, it’s anything but. Forecasting requires balancing multiple variables: market demand, production capacity, seasonality, promotions, competitor activity, and your own risk tolerance.
Forecasting accuracy is important because it directly impacts your most precious resources: time and money. A good forecast helps you optimize both.
Why Forecasting Is Hard—and Why It Matters
Forecast too low, and you risk running out of stock, losing sales, and frustrating customers. Forecast too high, and you tie up cash in unsold inventory that could have been invested elsewhere.
Some leaders swear it’s better to under-forecast, keeping inventory lean to protect the cash conversion cycle. Others insist over-forecasting is safer, ensuring you never miss a sale. I think both views miss the point. Obsessing over whether you’re slightly over or under is less useful than creating a rational, data-informed process—and then setting guardrails so your worst-case scenario doesn’t sink you.
Forecasting isn’t about perfection. It’s about creating a repeatable, adaptable system.
The Three Main Forecasting Approaches
Every forecast falls into one of three broad categories :
- Causal modelling — Looking at cause-and-effect factors that influence demand. This includes macroeconomic conditions (“Is the economy in recession?”) and micro factors (“We’re running a 20% off sale this month”). Price elasticity studies also fall here.
- Qualitative modelling — Using non-statistical information to build a picture of demand. This could be market research, surveys, or brand loyalty studies.
- Time-series modelling — Using your own historical demand data to predict future sales.
For early-stage builders of tangible goods, time-series modelling is usually the most accessible and immediately useful approach. You don’t need a PhD in econometrics or expensive software—just a spreadsheet, some data, and a clear process.
Four Time-Series Models You Can Start Using Today
1. Historical Comparison
The simplest method of all. Look at a comparable past period and assume you’ll sell the same amount in the future.
If you sold 228 widgets in January this year, you forecast 228 for January next year. It’s quick, but it’s also blunt—it ignores growth, market changes, and promotional plans.
Even with its limitations, historical comparison is foundational. It’s one of the few truly “known” data points you have.
2. Moving Average (MA)
The moving average method smooths out short-term fluctuations to give a clearer view of overall trends. It’s more nuanced than a straight historical comparison because it reduces the impact of outlier months.
Here’s how to build it:
- Gather a year of sales data, broken down by time period (monthly is common).
- Calculate the average for three consecutive periods (e.g., the last three months).
- Use that three-period average as your forecast for the next period.
This approach reduces volatility, but it still doesn’t account for predictable seasonal swings.
3. Seasonally-Adjusted Moving Average (SAMA)
SAMA adds a layer to your moving average to capture predictable demand spikes or dips—like holiday shopping surges or summer slowdowns.
To build it :
- Identify recurring seasonal patterns in your sales data.
- Calculate your moving average.
- Determine the percentage lift or drop in sales during seasonal periods compared to non-seasonal ones.
- Adjust your forecast accordingly.
If you’re using a year-over-year historical comparison, seasonality is already baked in (though not as precisely as when you calculate it directly).
4. Exponentially-Smoothed Bursting
This model is particularly useful for businesses that see intense demand spikes from promotions, PR pushes, or product launches—and for situations where you don’t have much historical data.
The idea is to create a “demand curve” based on past bursts and apply it to future events.
Steps :
- Identify past bursts of activity and gather sales data for the periods before, during, and after.
- Calculate your baseline sales.
- Determine the sales lift (delta) during and after each burst.
- Apply these lift percentages to forecast future bursts.
This approach blends the smoothing benefits of moving averages with targeted adjustments for short-term, high-intensity demand patterns.
What If You Don’t Have Historical Data?
Time-series models depend on past data. But what if you’re launching a new product or business and don’t have any?
If you have a similar product, you can use its historical data as a proxy, adjusting for differences in audience, price point, or use case.
If you have nothing, base your forecast on financial risk :
- How much capital can you commit to production without jeopardizing your ability to try again if sales flop?
- How confident are you in your product-market fit?
- How deep are you willing to discount to move product if needed?
Essentially, you work backwards from your risk tolerance. If you have $10,000 in operating capital and want to keep $5,000 in reserve, you produce as many units as the other $5,000 will buy. Launch, measure, adapt, repeat.
The Continuous Improvement Mindset
The most common forecasting mistake? Treating it as a static process. Businesses rarely track their forecast accuracy over time. They miss the opportunity to learn from misses and improve the model.
Your forecasting process should evolve with your business. That means :
- Reviewing actuals vs. forecasts regularly.
- Identifying which models work best for which products.
- Adjusting for new patterns as your business and market change.
Even small improvements compound over time, boosting both your bottom line and your operational confidence.
Forecasting as a Builder’s Superpower
Forecasting isn’t glamorous. It doesn’t give you the immediate dopamine hit of closing a sale or launching a product. But it’s one of the most powerful tools in your operational toolbox.
When done well, it frees up mental space. You’re not constantly scrambling to react to stockouts or overstock—you have a plan, and you trust the system you’ve built.
That’s the real goal here : not perfection, but predictability. Not clairvoyance, but clarity.
The best part? You can start today with nothing more than your sales data and a spreadsheet. Pick a model, apply it, measure your results, and keep refining. Over time, you’ll turn forecasting from a sweaty-palmed chore into a core strength of your business.
This article is a part of my series on topics for entrepreneurs, intrapreneurs, and people who just love building things. I podcast and post weekly with tools and guides on The Journey. Check out the companion piece here : https://6catalysts.substack.com/p/holy-bleep-a-guide-to-sales-forecasting