Home › Blog › Forecasting in Excel: Stop Reporting the Past
Forecasting in Excel: The Skill That Moves You From Reporting the Past to Owning the Number
Anyone can tell the room what happened last quarter. The person who can credibly say what happens next quarter is the one who gets the title.
Budget season. Your director asks the same question she asks every year: “So what are we going to do next quarter?” You pull up the actuals, scroll to the bottom row, and say what most people say — “Last year we did about 8% growth, so… probably something like that?” It’s a guess wearing a spreadsheet costume, and everyone in the room knows it.
Now picture the analyst across the table. She shares one tab. There’s a line that continues past today into a shaded band, a slider for three scenarios, and a single sentence underneath: “Base case is $4.1M, with 95% confidence between $3.8M and $4.4M — and here’s what breaks it.” She didn’t use Python. She didn’t use a $40,000 planning platform. She used Excel — and in about forty minutes.
What “forecasting in Excel” actually means
Forecasting has a reputation for being statistical and intimidating. It isn’t. In practice it’s three skills stacked on top of each other, and modern Excel hands you all three.
1. Project the trend — with a confidence range
Excel’s Forecast Sheet (Data tab → Forecast Sheet) takes a column of dates and a column of values and produces a chart, a projected series, and upper/lower confidence bounds in a couple of clicks. Under the hood it runs FORECAST.ETS — exponential smoothing that detects seasonality automatically. If your sales always spike in December, it knows.
2. Show the uncertainty instead of hiding it
The number that gets trusted isn’t the confident one — it’s the honest one. A single point estimate invites the question “how sure are you?” A range answers it before it’s asked. That confidence interval isn’t a hedge; it’s the thing that makes you credible.
3. Model the levers, not just the line
Trend projection assumes tomorrow looks like yesterday. Business rarely cooperates. This is where Scenario Manager, Data Table (one- and two-variable what-if grids), and Goal Seek earn their keep: they let you answer “what if churn runs 2 points higher?” or “what price gets us to $5M?” live, in the meeting, instead of promising to circle back.
The tell: If your planning deck has a number in it that came from someone eyeballing last year and adding a percentage — and nobody can explain the assumption behind it — you’re looking at the exact gap this skill fills.
Why employers pay a premium for it
Look at how the job titles are priced. General financial analyst roles in the U.S. sit around a $80,000 midpoint, and 2026 market data puts financial analysts with advanced Excel skills near $88,000 on average. Move one rung over to FP&A — financial planning and analysis, where forecasting is the job — and reported averages climb past $110,000, with the upper quartile well above $140,000.
That gap isn’t about knowing more functions. It’s about which direction you point the data. Reporting looks backward and is increasingly automated. Forecasting looks forward, requires judgment, and is what leadership actually buys. Hiring data across 2025 consistently flagged predictive analytics and data visualization as the fastest-rising skills attached to analyst postings — forecasting sits squarely at the intersection of both.
Reporting tells your boss what already happened. Forecasting tells your boss what to do. Only one of those gets invited back to the planning meeting.
There’s also a scarcity effect. Plenty of people can build a PivotTable. Far fewer can defend a projection — explain the model, name the assumptions, and say out loud where it would be wrong. That combination of technical work and honest framing is rare, and rare is leverage at review time.
How it translates to your career
- Raises: “My forecast came in within 3% for four straight quarters, and we stopped over-ordering” is a defensible, dollar-denominated result — the kind of sentence a manager can carry into a compensation conversation on your behalf.
- Promotions: Forecasting is the standard bridge from analyst to senior analyst, and from reporting roles into FP&A, demand planning, and revenue operations. It’s the skill that moves you from producing inputs to owning an output.
- Job security: Backward-looking reports are the first thing to get automated. The judgment call — which assumptions hold, which scenario to plan against — is the last. Be on the judgment side of that line.
The fastest way in is structured practice on real data. Our PivotTables class gets your history into a shape a forecast can actually consume, the Formulas & Functions class covers the FORECAST, TREND, and what-if machinery behind the scenes, and the Power Query & Power Pivot class keeps the underlying data refreshing itself so your model never goes stale.
Grab any two columns you already have — dates in column A, a number in column B, at least 24 rows if you can (two full years lets Excel find seasonality). Select both columns, then go to Data tab → Forecast Sheet. Before you click Create, open Options and set the Confidence Interval to 95%. Click Create. You now have a projected series with an upper and lower bound, plus live FORECAST.ETS formulas you can read and adapt. Then try one more thing: in an empty cell, enter =FORECAST.ETS.SEASONALITY(B2:B25,A2:A25) — that number is the cycle length Excel detected in your own data. Two minutes, and you’ve learned something about your business you didn’t know this morning.
Be the person who owns the number
Forecasting isn’t a statistics degree — it’s a handful of Excel tools plus the discipline to state your assumptions. Learn it properly, in a live class, on data that looks like yours, and walk into next budget season with a range instead of a shrug.
See the class calendar → Explore the Formulas & Functions class →FAQ
How much historical data do I need before a forecast is meaningful?
FORECAST.ETS will still return a number, but it can’t distinguish a seasonal pattern from noise, so treat the range as wider than it looks.Is Excel really good enough, or do I need Python or a planning tool?
What’s the difference between FORECAST.LINEAR, TREND, and FORECAST.ETS?
FORECAST.LINEAR fits a single straight line and projects it. TREND does the same thing but returns a whole range of fitted values at once, which is handy for charting. FORECAST.ETS is the modern one: it uses exponential smoothing, weights recent periods more heavily, and handles repeating seasonal patterns — which is why the Forecast Sheet button uses it. We cover all three, and when each is appropriate, in our Formulas & Functions class.