Free Trial Now Available | See Fintastic on Your Own Data
Get Started Free →Yes. Tell me more.

How to Reduce Financial Planning Cycle Time, and Why Speed Is the Wrong Target

https://www.linkedin.com/in/anthony-losurdo-b430b93/

Anthony Losurdo

|
6
min read
|
November 17, 2026

Ask a finance leader how long their forecast takes and you get a number. Five days. Ten days. Three weeks in a bad quarter. Ask what would happen if that number got cut in half and the answer is usually some version of "that would be great," followed by a pause, because nobody is quite sure what the team would do with the time.

On a recent panel on what changes in FP&A next year, two finance leaders who have spent the past year rebuilding how their teams plan were asked the obvious question. How much shorter is your cycle now?

Neither of them had a shorter cycle.

The short answer

You reduce planning cycle time by removing the work that happens before analysis starts: finding data, reconciling it, moving it between systems, and rebuilding models every time an assumption changes. That is where most of a forecast cycle actually goes.

But the teams that do this well usually do not end up with a shorter calendar. They end up with the same calendar and a different forecast. That distinction matters more than the number of days, and it is the one most planning projects fail to make.

Where the days actually go

Break a typical five day forecast cycle into what people are doing hour by hour and it looks something like this.

  • Pulling and reconciling actuals. A large share of the cycle, and it produces only a version of the numbers people trust enough to use.
  • Chasing inputs from business owners. Another large share, producing a complete set of assumptions.
  • Rebuilding or adjusting the model. A moderate share, producing a model that reflects this month.
  • Analysis and scenarios. A small share. This is the thing the forecast is for.
  • Review and conversation. A small share, and where decisions actually get made.

The first three are preparation. They do not produce a single insight. They exist because the data lives in several places, the model cannot absorb a change without being reworked, and nobody trusts a number until they have traced it themselves.

When a finance team says the cycle is too long, this is almost always what they mean. Not that analysis takes too long, but that they barely get to it.

Why compressing the calendar is the wrong target

Here is the trap. If you set out to cut a five day cycle to three, the fastest path is to cut scope. Fewer scenarios. Fewer business partners consulted. A lighter review. You hit the number, and the forecast gets worse.

The SVP of FP&A at Nextech described the alternative on the panel. Her team still takes about five business days from the end of close to issuing a forecast. What changed is that the searching, compiling, and manual input that used to fill those days is now close to instant. The days did not disappear. They got refilled with analysis, risk adjusted scenarios, and longer conversations with the business owners who actually drive the numbers.

Her framing was that the timeline did not change. The quality and value of what happens inside the timeline did.

The o9 Solutions panelist, who leads long range planning there, made the same point from a different angle. Just because you remove work does not mean total work goes down. His team's revenue forecast used to take one to two days of checking and assembling. It now runs in about two hours. That time did not get handed back to the calendar. It went into questions the team could never afford to ask before: did renewals land on time, did they land at the rate that was modeled, and if not, what is driving the gap.

The test that tells you which kind of improvement you got

After a planning project, teams usually report a time saving. The useful question is what the saved time bought. Three signals separate real improvement from a faster version of the same forecast.

  1. Scenario count goes up, not down. If building a second scenario still means duplicating a model and spending two days validating it, your team will keep producing one. When a scenario is a change to a few inputs in a model everyone shares, the number of scenarios per cycle rises on its own.
  2. Assumptions live in one place and cascade. If changing a headcount ramp or a renewal rate means touching six files, the model is not responding to the business. It is being rebuilt to match it.
  3. The conversation changes. The clearest sign is what business partners ask for. When they stop asking finance to pull a report and start asking what the options are for hitting a target, the forecast has become a decision tool rather than a record.

None of these show up in a cycle time metric. All of them show up in the quality of the decisions made at the end of the cycle.

What has to be true underneath

The preparation work is not a discipline problem. It is a consequence of how most planning environments are put together. Data sits in one place, the model in another, and the connection between them is rebuilt every cycle by people. Every structural change to the business, a new product line, a new region, a new pricing model, means reworking the model before it can produce a number.

The teams that cleared the preparation work did it by moving to a planning environment where the data and the model are the same thing: transaction level detail available in the model itself, assumptions held centrally and cascading to everything downstream, and scenarios created by changing inputs rather than copying models. That is also the precondition for AI being useful in planning. An agent asked to explain a variance can only trace the answer if the detail and the relationships are already there.

Fintastic is built that way, which is why both of these teams spend their fifth day arguing about the business instead of assembling the file.

The better question for 2027

Stop asking how to shorten the cycle. Ask what the last two days of your cycle are currently spent on, and what you would put there instead if the first three days ran themselves.

If the answer is scenarios you have never had time to build, you can test that directly. Start a free trial of Fintastic and build one of them against your own numbers.

Looking for a smarter way to analyze your data?

Book a Demo