Is Your Revenue Projections Accurate, or Just Wishful Thinking?

Let's not mince words. The majority of revenue projections are flawed the minute they're made. Not because they contain bad math but because the thought process behind them is unclear. If your projection sounds more like a wish than a plan, it likely is.

Forecasting is Not Hoping

A forecast does not represent what you wish to happen. It's what is most likely to occur based on tough data. That involves knowing where your numbers derive from, how reliable your sales machine is, and where risk is lurking. If your forecast begins with, "If everything goes right," then you're in trouble already. That's not a plan. That's a fantasy.

The Lazy +10% Trick

Plugging in last year's figures and adding a growth percentage—often 10% or 20%—may be aesthetically pleasing in an Excel sheet. Still, without the context that the environment hasn't changed, it signifies nothing. In actual business, something always varies.

If you added new customers in the last year due to a fluke PR campaign or a one-off discount, and that's not in your plan anymore, you're not going to repeat last year. If your product got better and brought in more revenue, but your roadmap has broken down this year, then you're driving on vapors.

A prediction that guesses the future will be like the past, with no good reasons, is a guess in a suit.

When Sales Optimism Distorts Reality

Your sales team might report that there is $3 million in the pipeline. Sounds wonderful. But what's really in contract negotiation? How much of it is still in the early talk stage? What's the past close rate on comparable deals?

Forecasts inflated by sales optimism can sink a company. There's a difference between what the team wants to close and what's likely to close. Suppose you're not measuring forecast accuracy against actual close rates and time-to-close history. In that case, you're using a hope, not a forecast.

Forgetting Churn? That's Dangerous

If you're just adding numbers and not subtracting any, your forecast is a lie. Churn is real. Customers do leave. Some downgrade. Some cease paying. If your forecast doesn't adjust for that, it's broken.

Imagine that you are producing $100,000 in new business each month. But if you're losing $30k on churn and cancellations, your net revenue is $70k—instead of $100k. If you're developing your plan around the incorrect number, all your subsequent decisions will be incorrect.

Worse, if your churn rate is increasing and you're not factoring that into the forecast, your numbers will be okay—until they aren't, and your bank account isn't.

Your "Best Case" Should Not Be Your Forecast

It's okay to model different scenarios: best case, worst case, and a middle line. But suppose your actual forecast sits closer to the dream scenario because it looks good to investors or your board. In that case, you're setting yourself up for a crash.

Projections need to be in the middle of what happens—not at the height of your fantasy. Your budgets for hiring, marketing, and cash flow all hang on that number being good. If your plan is only valid when everything works out, it's not a plan. It's a bet.

What a Real Forecast Looks Like

An actual forecast doesn't simply consider revenue. It considers the complete picture—pipeline health, close rates, churn trends, market timing, product readiness, and even the bandwidth of the team. It's a live number, not a rigid goal put down three months prior.

You require pristine assumptions. If you claim to close $500k next quarter, where is that from? What phase are those deals in? Are they repeatable? Is the sales cycle trending on time?

You require sound data. It's not advisable to predict the future based on gut feel. You need to be drawing from genuine close rates, average time-to-close and deal progress data.

You require genuine churn arithmetic. How many clients are falling off? Why are they? Can it be prevented? If you don't know, you're predicting income you'll never be able to retain.

You must stress-test the weak spots. If one large customer is late, or if your next new product launch is pushed back by 30 days, how will that affect revenue? If your entire plan collapses from a tiny variation, the forecast is too sensitive. According to Gartner’s research on sales forecasting, over 50% of sales leaders say their forecasts are unreliable—highlighting the urgent need for more accurate, data-driven projections.

Ask These Five Tough Questions

This is your gut check. If you can't respond to these with figures, not guesses, your forecast isn't tough enough.

1. How much of our projection is dependent upon deals that haven't even entered the pipe?

2. Are we factoring seasonality, customer behavior changes, and pricing trends—or ignoring them?

3. Are we subtracting churn and downgrades from the forecast or simply adding new revenue?

4. What happens if our largest lead doesn't go through? Do we bounce back or implode?

5. Are we relying on historical close data to inform this prediction, or are we just doing what seems pretty?

   
   

Good Forecast Examples

1. Microsoft – Cloud Revenue Growth

• Approach: Tracks actual Azure usage, recurring revenue, seasonality, and churn.

• Result: In Q2 FY 2023, Microsoft projected ~30% growth in cloud revenue, reporting 31% actual growth—demonstrating <1% variance and data-driven precision .

2. Netflix – Subscriber Modeling

• Approach: Combines machine learning with historical subscriber data, regional behavior patterns, and dynamic pricing.

• Result: Following its password‑sharing crackdown in 2023, Netflix correctly forecasted a short-term subscriber dip followed by recovery, validating its scenario-based forecasting.

3. Slack (Before Acquisition) – ARR Forecasting

• Approach: Forecasted ARR based on Monthly Active Users, account expansion, and renewals.

• Result: Maintained ARR forecasts within 3–5% of actual results each quarter—demonstrating disciplined, data-focused forecasting accuracy (Note: internal Slack data).

Bad Forecast Examples

1. WeWork – Overinflated IPO Projection

• Issue: 2019 S‑1 predicted sky-high growth assuming full occupancy and negligible churn.

• Result: IPO pulled, valuation collapsed from ~$47B to ~$10B, and later bankruptcy, revealing catastrophic forecast failure

2. JCPenney – Pre‑Pandemic Sales Forecast

• Issue: Assumed stable foot traffic and no disruption from e-commerce.

• Result: Pandemic shattered assumptions; bankruptcy followed in 2020 due to lack of scenario modeling and digital responsiveness.

3. Tesla – Model 3 Production Target

• Issue: Musk forecasted 5,000 Model 3 units/week by end‑2017 without accounting for production or supply chain constraints.

• Result: Actual output reached only ~2,400 units/week; production was delayed into 2018

The Honest Truth

A prediction isn't a narrative you use to make people enthusiastic. It's a cutting-edge instrument you use to guide your business. If you get it wrong, your hiring will be misplaced. Your expenses will be misguided. Your investors will doubt. And your runway may not make it.

Wishful thinking is a nice slide deck. Nonetheless, it is solely through honesty that the doors will continue to remain open.

Make your forecast real. Then, improve it.