Last quarter, an FP&A executive I've known for years caught me up over coffee. Her team had just spent nine months rolling out the third FP&A platform they'd tried to make fit in five years. Different vendor, same arc: a confident demo, a long implementation, a launch that landed at "good enough," and a quiet drift back to Excel for the parts that actually mattered.
What stuck with me wasn't the cost. It was how unsurprised everyone in the room was. The misfit isn't a bug in any one product. It's the default mode of an entire category.
The misfit always shows up in the same three places
I've now sat with enough finance teams to map this. The friction is never random. It clusters in three predictable spots:
The chart of accounts
Every business books revenue, costs, and headcount in a structure that mirrors how it actually operates. Vendor templates are generic. Your team translates every cycle.
The planning cycle
Some teams reforecast monthly, some quarterly, some run a rolling 18-month view. The platform assumes one rhythm. You reshape your process or build Excel workarounds.
The reports the board reviews
The variance deck, the scenario page, the headcount view your CFO has reviewed every month for eight years. It is specific to how your company thinks. Templates rarely get there.
None of these are exotic. They're the normal contours of a real finance team. And yet they land, every time, in the "professional services" column of the SOW.
Why this happens (it's the economics, not the engineering)
The misfit isn't because vendors are lazy or their engineers aren't capable. It's because the business model demands it.
The legacy FP&A category was built on a stack of incentives that all point the same way: sell the platform, hand off to implementation, refer to a partner network for anything custom, then meter every change after go-live as a billable hour. Custom configuration isn't a missing feature. It's a profit center.
Most FP&A platforms don't sell software. They sell software plus a permanent dependency on the people who can change it.
So the platform stays generic on purpose. Your hierarchy lives behind a consultant. Your custom report lives behind a quote. The "tool" that was supposed to give your team leverage becomes another vendor in the queue.
Newer cloud-native players moved faster on UX. The economics under the hood didn't change.
What "fits" actually means
A platform fits when three things are true on day one:
Your structure is the structure. Your GL, your departments, your hierarchy, your roll-ups - configured as the native model, not bent into a template.
Your cycle is the cycle. The way your team plans today is what the software supports out of the box, not a "phase 2" item.
The reports your CFO reviews are the reports the platform produces. Not approximations of them. The actual deck.
None of this is technically hard. It's organizationally hard. To fit a real team, the people configuring the software have to understand finance and ship code in the same week. Most vendors have split those two skills across three companies (vendor, implementation partner, your in-house ops team) and called the seams "best practice."
The trust question
I'll be direct about the obvious objection: "every vendor says they're different."
The reason I'm comfortable making this argument is that I spent the decade before Novaplan doing the work that this problem actually requires. I ran BI teams and watched what happens when finance data is finally modeled the way the business runs, not the way an ERP template assumed it would. I shipped AI in production at Fortune-scale and learned the part of that work no one puts in the keynote: AI is only as useful as the data layer underneath it. And last year I co-founded an agentic-AI platform from zero, which is the kind of work that teaches you fast where "AI" stops being a slide and starts being something a team relies on Monday morning.
I'm not pitching that experience. I'm naming it because the thing FP&A actually needs is a stack that's data-modeled like a BI system, configured at the speed of modern AI tooling, and operated by the people who built it. That's a specific combination. It's the combination Novaplan is built around.
What changes when it fits
The numbers we're seeing on the design-partner side are not subtle. A board-ready variance deck in 10 minutes instead of a week. From signed contract to live-and-configured in 14 to 28 days instead of two quarters. Zero handoffs between the team using the platform and the team that can change it.
A board-ready variance deck.
Ask once. Get a slide deck before your coffee cools. Used to take an FP&A team a full week.
What I find more interesting is the second-order effect. When the platform fits, finance stops being the team that produces numbers and starts being the team that interprets them. The reforecast doesn't eat the month it was meant to plan. The strategy conversation, the what-ifs, the moment a number actually changes a decision - those are the parts of the job that originally pulled most of us into FP&A. They're also the parts generic platforms quietly take away.
We're at the moment when the technology stack underneath finance can finally be rebuilt around how teams actually work, instead of the other way around. That's the bet. If anything in here resonates with where your team is, I'd love to hear about it.