Five FP&A Stories Finance Leaders Should Be Paying Attention to Right Now
Finance teams are heading into 2027 planning with an unusual combination of pressures.
CFOs are becoming more concerned about inflation and interest rates. At the same time, many are remarkably optimistic about their own companies’ profitability.
AI is moving from experimentation into actual finance workflows.
FP&A teams are still spending too much time on low-value work.
Finance leaders are worried about talent.
And somewhere in the middle of all of this, FP&A is being asked to build next year’s plan.
Taken individually, these look like separate finance stories.
They aren’t.
They point to a larger shift happening inside the finance function.
The economics of how Finance operates are beginning to change.
Here are five developments we’re watching at The Schlott Company, and more importantly, what finance leaders can actually do about them.
1. Stop trying to predict the economy. Understand your exposure to it.
New Federal Reserve survey data reported by Reuters shows corporate finance leaders raising their inflation expectations, with interest rates becoming an increasingly important concern.
That matters for FP&A because macro assumptions eventually become operating assumptions.
Interest rates become interest expense.
Inflation becomes compensation pressure, vendor increases and pricing decisions.
Changes in financing conditions can affect customers, capital projects, acquisitions and hiring.
The traditional planning response is to debate the “right” assumption.
Will inflation be 3% or 4%?
Where will rates be six months from now?
What will the Fed do?
Those conversations have value, but they can also consume enormous amounts of time without answering the question management actually needs Finance to answer:
What happens to our business if we’re wrong?
What Finance can do now
Take the three or four external assumptions capable of materially changing your 2027 plan and identify the exposure attached to each one.
For every major assumption, ask:
What are we assuming?
Write it down. Don’t leave it buried in a formula.
What happens if we’re materially wrong?
Don’t model twenty scenarios. Model the ones that could actually change a decision.
Where would we see the impact first?
Cash? Margin? Revenue? Hiring? Covenants?
What action could management take?
A scenario without an available management response is mostly an interesting spreadsheet.
The goal isn’t to correctly predict every macroeconomic variable.
It’s to make sure management isn’t surprised by what those variables do to the business.
2. CFOs expect profit growth. FP&A should understand exactly where it comes from.
Grant Thornton’s Q3 2026 CFO survey produced an interesting contrast.
Only 46% of respondents said they were optimistic about the U.S. economy, yet 80% expect their organization’s net profit to grow over the next 12 months. Thirty-five percent expect profit growth above 10%.
That’s a significant amount of optimism inside a fairly uncertain operating environment.
For FP&A, “profit will grow” isn’t really an assumption.
It’s an outcome.
The useful work begins when Finance works backward from it.
Where does that growth actually come from?
Pricing?
Volume?
Headcount productivity?
Gross-margin improvement?
Lower operating expenses?
AI?
And what has to be true for each of those things to happen?
Turn the profit target into an assumption map
For every material piece of expected profit improvement, identify:
Driver → Assumption → Owner → Evidence → Risk
Suppose the plan contains $5 million of margin improvement from productivity.
Don’t stop at $5 million.
Ask what operational change creates it.
If AI is supposed to create productivity, where should the benefit appear?
Fewer hours?
Slower hiring?
Reduced contractor expense?
Greater transaction volume with the same team?
Shorter cycle times?
This is particularly important with AI.
The same Grant Thornton survey found that 65% of respondents rated the performance and quality of their AI technology as good or excellent.
We’re moving beyond the stage where “we’re investing in AI” is enough.
Eventually the investment has to meet the P&L.
FP&A should help make that connection visible.
3. Don’t use AI to make unnecessary finance work faster.
FP&A Trends’ 2026 research found that 47% of FP&A teams are still involved in low-value-added activities, while only 19% can run scenarios in real time or in less than a day.
That’s worth considering as companies race to introduce AI into Finance.
Because there are two very different AI strategies.
The first asks:
How can AI perform our existing work faster?
The second asks:
Why are we doing this work in the first place?
We prefer starting with the second question.
Otherwise Finance risks automating processes that probably shouldn’t exist.
A report nobody uses doesn’t become valuable because AI produces it in thirty seconds.
A reconciliation caused by inconsistent upstream data doesn’t become a good process because an agent performs it.
A fifteen-step workflow doesn’t necessarily need fifteen automated steps.
It might need six steps.
Before automating a finance process, ask five questions
1. Who uses the output?
Not who receives it. Who actually uses it?
2. What decision does it support?
If nobody can identify one, that’s worth investigating.
3. Why is the process manual today?
The answer often reveals a system, data or ownership problem.
4. Which steps actually require judgment?
Protect those. They’re usually where Finance creates value.
5. What would happen if we stopped doing it for 30 days?
Sometimes the answer is surprisingly little.
Automation should come after simplification.
Otherwise we’re just giving unnecessary work a very fast employee.
4. Look for the “human middleware” inside Finance.
The 2026 Finance Architecture Survey highlighted another problem: half of respondents identified talent and skills gaps as the greatest risk to the stability of their finance function over the next year.
The research also raises an idea that will be familiar to almost anyone who has worked inside Finance:
Human middleware.
System A produces data.
Someone downloads it.
They clean it.
Someone maps it.
Another person reconciles it against System B.
The file gets emailed.
Someone pastes the result into Excel.
Eventually it reaches a management report.
All the technology technically works.
It just requires several human beings standing between the systems to make it work.
We’ve seen versions of this throughout Finance.
And because people become extremely good at keeping these processes running, the underlying problem can remain hidden for years.
Try a human-middleware audit
Pick one recurring finance process.
It could be:
- monthly close,
- forecast refresh,
- management reporting,
- revenue reconciliation,
- board reporting,
- accounts payable.
Follow one piece of information from source to final output.
Every time a person touches it, write down what they did.
Downloaded.
Reformatted.
Mapped.
Reconciled.
Copied.
Approved.
Uploaded.
Corrected.
Explained.
Then ask:
Why did a human need to perform that step?
You’ll usually find three categories.
Some work requires genuine judgment.
Some work exists because systems don’t communicate properly.
And some work exists because that’s simply how the process evolved.
Those categories shouldn’t receive the same solution.
5. AI needs to become a financial-management question.
As companies prepare 2027 budgets, AI spending is creating a new FP&A challenge.
CloudZero recently highlighted how difficult AI costs can be to forecast because consumption can spread across departments and the connection between usage and business output isn’t always obvious.
This problem is likely to get more important.
AI spending can appear as enterprise software, individual subscriptions, API usage, embedded functionality inside existing platforms or consumption-based charges.
Without some discipline, Finance could quickly find itself asking the same question it has asked about cloud spending for years:
Who owns this bill?
We’d go further.
Finance should ask three questions about every meaningful AI investment.
Who owns the consumption?
Every material cost needs an accountable business owner.
What economic outcome are we buying?
Faster processing isn’t necessarily an economic outcome.
Are we avoiding hires?
Reducing outside spend?
Increasing capacity?
Accelerating close?
Improving collections?
Increasing revenue?
Reducing errors?
Define the expected benefit before the spending disappears into the software budget.
How will we know if another dollar is worth spending?
This may eventually become the most important question.
Some AI usage could have extraordinary incremental returns.
Other usage may amount to expensive experimentation.
FP&A should help management distinguish between them.
The bigger opportunity isn’t AI. It’s capacity.
There’s a thread running through all five of these developments.
Finance is dealing with greater uncertainty.
Management still expects growth.
AI can perform more finance work.
Finance teams remain buried in low-value activities.
Talent remains difficult to find.
Those conditions create an interesting question for CFOs:
If technology materially reduces the cost of producing financial information, what should Finance do with the capacity it gets back?
Hopefully, the answer isn’t more reporting.
The opportunity is to move Finance closer to decisions.
More time understanding what’s changing in the business.
More time challenging assumptions.
More time with Sales, Operations and leadership.
More time evaluating capital allocation.
More time identifying risks before they appear in the financial statements.
And less time moving information between systems simply because that’s how we’ve always done it.
At The Schlott Company, we think that’s the more interesting version of finance transformation.
Not simply replacing spreadsheets.
Not adding another dashboard.
And not putting AI on top of every existing workflow.
Start with the work.
Understand why it exists.
Decide what requires human judgment.
Remove what doesn’t need to happen.
Then use systems, automation and AI to redesign what’s left.
Because the goal isn’t to build a finance function that produces more.
It’s to build one that helps the business decide better.
