A four-step framework for CFOs to control AI token costs.

Read Now

CFO's Budget Season Playbook: Christian Wattig's Top Frameworks

Last updated: November 202659 min watch

Speakers

  • Christian Wattig, Founder, Inside FP&A

  • Kirk Kappelhoff, Senior Director of Strategic Finance, Drivetrain

It’s the first week of budget season, the kickoff email has gone out, the templates are sitting in everyone's inbox, and the deadline is a comfortable four weeks away. Plenty of time, right?

Well, you would think so. Then the submissions arrive (on the day of the deadline, naturally), and the bottom-up total lands nowhere near the number leadership has in mind. As a result, you are the one who has to go back to every department head and ask them to do more with less, and you have about a week to do it.

If this sounds familiar, you're in good company. Christian Wattig has run eleven annual planning cycles at companies like P&G, Unilever, and Squarespace, and the top-down and the bottom-up never matched. Not once. But Christian doesn't see that as a problem. In his view, the mismatch is what starts the conversation about what's realistic. Finance's job is to reconcile the two numbers, ideally without going back to department heads every time they disagree.

In the webinar below, Christian shares his budget season playbook with Kirk Kappelhoff, Drivetrain’s Director of Strategic Finance.

No time to read? This article covers it all, with the live Q&A and full webinar transcript at the end.

Here, we explain what top-down and bottom-up budgeting are and why the two numbers almost always diverge. We'll also show you how to build each one so they can be compared, and how to challenge a department's numbers without creating friction.

59:07

What is top-down vs. bottom-up budgeting?

A top-down budget starts with the end result in mind. It is expectation-setting, and it usually comes from senior leaders or from finance. However, Christian's rule is that it still has to tie back to real business drivers, such as headcount, advertising spend, product development, or market expansion. "We should grow eight percent" is not a top-down budget. It is a wish.

In contrast, a bottom-up budget starts from scratch. Each department builds it up from the initiatives and campaigns it actually plans to run next year, along with what those will cost and what they are expected to return.

Each side also comes with a predictable bias:

Bottom-ups tend to be conservative (people want to under-promise so they can over-deliver), while top-downs tend to be aggressive since they come from leadership and are a bit removed from the day-to-day.
Christian WattigFounder, Inside FP&A

Both sides have to build their own version for the comparison to mean anything. If finance builds the whole plan alone, the assumptions are too thin to be useful when you are trying to explain variances later in the year. On the other hand, if departments simply send in numbers and finance adds them up, finance is not adding the value it should. Neither one is a budget on its own, but together, they are the start of one.

What does the gap between top-down and bottom-up tell you?

For example, imagine leadership's top-down calls for $12 million in new ARR next year, and the sales team's bottom-up comes in at $9 million. Neither number is wrong. Sales built theirs from the reps they have and the ramp times they have seen, while leadership built theirs from the growth the board expects and the market they believe is there.

Now, you could split the difference and call it $10.5 million. But nobody built that number. So there’s no plan behind it: no hiring target, no ramp assumption, and nothing to measure against as the year goes on. If you fall short, you can't tell whether it was due to a problem with execution or an unrealistic target, because nobody said what the extra $1.5 million depended on.

The better move is to put leadership's number in the same terms sales used. What does $12 million assume about headcount, hiring dates, and ramp time? Set that beside sales' plan, and the $3 million gap becomes a question of sales capacity: something like eight more reps, starting by March. That's a question sales can answer.

Christian's method applies the same logic to every department. He builds the top-down from a handful of tactic-level drivers, chosen with the business teams, so that each gap turns into a question about specific assumptions rather than a total.

How do you build a driver-based, top-down budget?

Christian is a big fan of driver-based top-down planning. Instead of starting from a growth rate, you model what the business is going to do differently next year and what results those changes should yield, so that the conversation with a department is about the drivers rather than the total.

For example, at a subscription company, he modeled unaided awareness, direct-response spend, inflation, and promo strategy across three scenarios: balanced, aggressive, and conservative. That let him show the board exactly where the plan was taking risk, and it made the same model useful all year long, because you can pull it out monthly or quarterly and adjust it for reforecasts.

It's worth noting that a driver-based top-down has its drawbacks, too. It requires a real understanding of the business, it can be time-intensive, and more drivers mean more compounding uncertainty. This is why Christian follows these best practices when developing the driver list:

  • Drivers live at the tactics level: "Marketing" and "sales" are line items. Unaided awareness, direct-response spend, and promo strategy are drivers.
  • Five to ten of them, not a hundred: Apply the 80/20 rule. A handful of drivers explain most of the result, and every extra one adds uncertainty.
  • Finance can't pick them alone: It takes real collaboration with the business teams, because they're the ones who know which tactics actually move the number.
  • Every assumption has to be measurable, yours or a department's: If you can't tell afterward whether it held, you can't explain the variance.

How do staggered budget deadlines work?

The first of the three budget-season challenges Christian described is late inputs. You set one deadline four weeks before the board meeting, thinking that is plenty of time. Then, departments wait until the day itself and submit something that is not where it needs to be, in depth or in the numbers, and now you are scrambling with no time left.

The fix is a staggered timeline with multiple deadlines. For example, kicking off in early November, a first-pass deadline [at the] end of November, a second pass mid-December, and a final lock in mid-January.
Christian WattigFounder, Inside FP&A

The early deadlines assume the first pass won't be what you need. People go back and improve it for the second pass, and finance gets more time to analyze the submissions and run scenarios. But there is a catch. This only works if senior leadership (ideally the CEO or your SVPs) has bought in, because the first-pass deadline only gets respected when there is an actual business review afterward that sends things back for changes.

Through the whole stretch, finance acts as the project manager, following up, reiterating deadlines, and explaining the reasoning behind them. Simply put, if nobody is chasing, nobody is submitting early.

Why should you ask for budget inputs as a range?

Here is the part of the webinar that got the most nods. The bottom-up almost never matches leadership's expectations on the first pass, which means finance ends up sending people back to do more with less, sometimes more than once. Christian's fix is in how the inputs are collected in the first place: ask each department for a range rather than a single number, starting with the very first submission.

That way you don't have to keep going back to department owners every time there's a gap. You can just use the more aggressive end of their own range, and the width of that range also tells you something about their level of uncertainty.
Christian WattigFounder, Inside FP&A

In other words, you get a second number from every department without a second meeting, and the width of each range tells you how confident each team is in its own plan. A department that submits $800K to $850K knows its business. Likewise, a department that submits $600K to $1.2 million is telling you something, too.

How do you challenge a department head's number without friction?

An anonymous audience question got the most practical answer of the session: how do you challenge a department head's assumptions without causing friction? After all, you are usually more junior than the person across the table, and they know their business better than you do. Christian's answer is to do your homework before you walk in.

The best way is to do your homework first — build your own top-down estimate, even a rough back-of-envelope one, before the conversation. Then go in and say, 'here's my own projection and my assumptions, walk me through where yours differ' — completely different tone than opening with 'your number looks low, we need it higher.'
Christian WattigFounder, Inside FP&A

Coming in with your own assumptions turns the meeting from a negotiation over a total into a comparison of two views, driver by driver. It is also usually how you find a way to genuinely help the department rather than simply push on its number. For a department that does not want to engage at all, Christian's line is that a high-ROI idea gets funded through this process, and that what a team is allowed to spend can flex with business realities even though the budget commitment itself does not change.

How does the full reconciliation run, start to finish?

Let's walk through the full reconciliation, start to finish:

  • Build the top-down first. Five to ten tactic-level drivers, with balanced, aggressive, and conservative cases.
  • Collect the bottom-up as ranges by the first-pass deadline.
  • Compare driver by driver. Start from your assumptions and ask where theirs differ.
  • Consolidate with version control. Know who submitted which version, when, and whether the department head actually signed it off.
  • Use the aggressive end of their range where the total falls short, before you ask for anything new. Only go back for a real change.
  • Review, send back, and repeat through the second pass. Lock in mid-January.

Once the plan locks, the number becomes a commitment to your stakeholders about what the company will achieve and how it plans to get there, which is why Christian says it shouldn't change six months in. That's the core difference between a budget and a forecast: the budget holds, and the forecast updates the picture as the year unfolds.

Get the bottom-up from the team, challenge it with your top-down, and go through the iteration. It can be painful having multiple rounds where the bottom-up is far from your top-down, but that pain is exactly where the value and the real understanding comes from.
Christian WattigFounder, Inside FP&A

How to use ChatGPT to prepare for a budget conversation

Christian also showed how he uses ChatGPT to prepare for budget conversations. He follows a prompting structure borrowed from Greg Brockman, OpenAI's president, which has four parts: the goal, the return format, warnings about what the model should check, and a context dump. He demonstrated it live, playing a finance manager preparing to discuss a CMO's bottom-up submission. His prompt covered each part:

  • Goal: Talk to the CMO about their bottom-up submission and get their thoughts on improving marketing ROI.
  • Return format: Define a specific structure for the output.
  • Warnings: Make sure each question can be used to run financial analysis and lead to a concrete recommendation.
  • Context: "I'm more junior than the CMO, I want this to feel collaborative rather than imposing, and I don't want to seem like I'm posturing for our CFO."

The output was five questions. The first asked which marketing programs had punched above their weight over the last twelve months, which Christian liked because it lets the CMO start with wins. The next asked about the CMO's biggest bets and the assumptions behind them, and along the way, the model suggested asking for customer lifetime value data.

Christian calls brainstorming one of his favorite uses for AI in budgeting, because it doesn't matter if the model makes something up when you're not relying on it for facts. For facts, a search engine is usually the better tool.

Leverage technology to take the pain out of consolidation

Of all the steps in the reconciliation, consolidation is the one a tool improves most directly. In the webinar poll, 90% of attendees were still running the process in Excel or Google Sheets. If files from ten departments take more than a day to roll up into company revenue and profit, there are already new updates by the time you finish. And as Christian pointed out, it isn't only about time. You're also never quite sure the formulas are right or that you used the right version.

Kirk's demo showed how Drivetrain handles it. Each department head works from one template that shows only their own actuals, forecast, headcount, and attrition, so finance isn't "rebuilding the same report thirty times for thirty teams." Comments live in the budget itself and ping the owner in Slack, Teams, or email. And when it's time to investigate a variance between budget, actuals, and the reforecast, Drivetrain's AI drafts the commentary for you, with live references to the underlying metrics so the text updates as the numbers do.

Christian's reaction went straight to the send-back rounds and the flood of "this number needs to change" emails. You could comment in Google Sheets, too, he said, "but you'd still have the version-control problem, whereas everything stays organized in Drivetrain."

The budgeting use case shows the collaborative flow end to end, from collecting ranges to locking the plan. If you're a SaaS company building the top-down side, our top-down revenue modeling template is a good place to start. And if you'd like to see how your own budget season could run with a lot less scrambling, contact us today to book your demo.

About Christian Wattig

Christian Wattig has spent more than 14 years in FP&A, including eleven at Procter & Gamble and Unilever leading FP&A and accounting teams. He has an MBA from NYU Stern and was an FP&A leader at Squarespace through its IPO. He's the founder of Inside FP&A, a corporate training firm that has trained teams at Google, Merck, Lowe's, and Discover, and he directs the Wharton School's FP&A Certificate Program, which he created. He writes a weekly newsletter.

Summary

Top-down vs. bottom-up budgeting works like this: the top-down is the target that leadership or finance sets, tied to a handful of business drivers, while the bottom-up is the target from the combined budgets that each department builds from its own initiatives. You build both types of budgets separately, and you should expect them to disagree.

Christian's method for resolving the differences has three parts:

  • A top-down built on 5–10 drivers, not a hundred: Each driver is a specific tactic, such as promo strategy or direct-response spend, so the conversation with a department is about assumptions rather than a total.
  • A staggered calendar, not a single deadline: An early first pass, a business review that sends weak submissions back, and a second pass before the lock in mid-January, so you're not fixing everything the week of the board meeting.
  • Bottom-up inputs collected as ranges, not single numbers: When a department's number falls short of the target, you can often use the more aggressive end of its own range instead of sending it back to redo the plan. The width of the range also shows how sure each team is about its numbers.

Note that this page covers the reconciliation process, which is one stretch of a longer budget season timeline that starts well before kickoff, with a review of last year's process, a planning roadmap, and a RACI.

Summarize with AI:

Questions from the webinar

How do you build a living budget instead of a one-and-done exercise?

Reforecast at least monthly or quarterly, and tie each reforecast back to the original budget rather than running it separately. Once a year plus a mid-year update is not enough. Tying back is much easier when the budget was built at the driver level, and a driver-level budget is exactly the comparator you need for variance analysis.

How do you communicate the value of budgeting to leadership in a non-finance-driven organization?

Ask what they worry about, then run the process to answer it. For example, a fast-growing company cares whether the growth is sustainable, whereas a nonprofit may care most about what happens if funding falls through, so the budget becomes that scenario plan.

Which non-monetary goals belong in the annual plan?

Start from each team's strategy and tactics, then work back to the metric that tracks them. For example, if product quality is a goal, something like consumer hotline call volume works as a leading indicator. A spike there can point to a quality issue that will eventually hit sales, and in a subscription business, it shows up as churn.

Why does ChatGPT give a different answer each time?

For brainstorming, that is actually useful, so regenerate to get more options. For a fact, like the population of Paris, use a search engine.

How does the annual budget fit a five-plus-year strategic plan?

The strategic plan sets the guardrails and the starting expectations, and the annual budget is informed by it rather than reconciled to it line by line.

Speaker spotlight

Christian Wattig

Founder, Inside FP&A

Christian Wattig is an FP&A expert with 14+ years of experience in MNCs (P&G, Unilever) and high-growth tech startups (Squarespace). After earning his MBA from NYU Stern, he helped take Squarespace public. Now an entrepreneur, Christian directs the FP&A certificate program at Wharton and runs his firm, FP&A Prep, offering corporate training, online programs, and sharing insights through LinkedIn, newsletters, and his 'FP&A Tuesday' series.

Kirk Kappelhoff

Senior Director of Strategic Finance, Drivetrain

Kirk Kappelhoff is a financial modeling expert with a BBA in Finance & Accounting and nearly a decade of experience at Deloitte, EY, and KPMG, where he built models for pre-IPO companies, M&A transactions, and strategic planning initiatives. At KPMG, he led the Business Modeling Services team, specializing in equity stories and financial forecasts that help high-growth companies communicate their value to investors. At Drivetrain, Kirk writes about strategic planning, granular reporting, and modern FP&A best practices for rapidly scaling finance teams.

Ready to start your journey?

Fewer spreadsheets, faster planning, better decisions.

Book a Demo