By Ben Heinl

In Hard Problem #4, I spent some time describing what I see as some of the Easy Problems of FP&A. Following on from that is a hard problem I don’t hear people talking about too often, which is how hard it is to make an informed decision when purchasing FP&A software.

Why is this a Hard Problem? Well, for starters most people that are part of making a decision to purchase FP&A software… it will be the only time they ever do it in their life. So they are inexperienced business software buyers. And they are up against one or more experienced software sales teams that are focused only on that every day. So for the most part, the salesperson has a massive advantage in terms of information.

Further, there is no solid information source out there to really help a buyer. G2 and TrustRadius give it a go, but how do you know which ones are legitimate and which ones are flowery reviews from the vendors and their partners? You really have to read between the lines and it is hard to fully trust the information there. Analysts like Gartner and Forrester are “pay to play” and their view of the market is much more limited than I think most people want to admit. Never mind trying to discern why “ability to execute” and “completeness of vision” as the axes in Gartner’s Magic Quadrants seem to be accepted in the industry as the best way to evaluate a market.

So buyers are on the back foot right away because they typically have limited experience in terms of what actually makes a difference to success in a real life production FP&A environment. Further, all the information they are getting is often focused on complementary features and hype and in the best case scenario, they may have some individual experiences themselves. (These experiences are almost assuredly the most valuable).

So given the difficult starting points, it is no wonder that people make decisions they regret after the fact. To help you avoid this, here the most common mistakes people make when faced with this buying decision that I have seen.

“I can do X in Excel and this software is ‘more modern than Excel’ so it probably can also do X.”

Probably it cannot do X. Excel is a super mature, highly flexible tool that has been put through its paces by literally billions of users in literally over a trillion spreadsheets. It has in the neighbourhood of a gazillion features. Whatever software you are looking at has way fewer features than Excel.

Common Mistake #1.1:

“I can do Y in IBM Planning Analytics (TM1), so surely I can do Y in this brand-new modern cloud-based AI-driven tool.”

Probably it cannot do Y. TM1 has been around for over 40 years – just as long as electronic spreadsheets like Excel and it is also battle-tested by innumerable companies in innumerable FP&A application scenarios across innumerable platforms with innumerable users of all experience levels. TM1 can solve a lot of problems and whatever new tool you are looking at was probably designed to focus on solving just one problem. That problem is probably either “Planning” or “Reporting / Dashboarding”, and don’t think that because the tool does one of those that it does the other.

I have seen time and again our IBM Planning Analytics (TM1) customers having buyer’s remorse shortly after buying a product claiming to replace “old, legacy TM1 with a modern solution”. “I assumed that because the software had nice planning capabilities that it would have a built-in ETL tool as powerful as Turbo Integrator” or “I assumed because it showed a graph on a screen that it had the flexibility of reporting and formatting that I get in my TM1 add-in” and so forth. You know what they say about what happens when you assume. : )

“My ERP vendor has a planning and/or reporting tool and therefore it must be better integrated with their own GL and easier to use than a third party tool.”

In pretty much every case I have seen, the ERP vendor’s Planning and Reporting tools are much worse than third party tools. This is because it is not their core competency to develop such tools and they also don’t have to compete with those products in the real-world market outside of their captured ERP customer market. Further, an ERP vendor is no better able to produce “one version of truth” with their own Planning and Reporting tools than a third party tool. It is copying the data from its source into some other database or tool just like everyone else.

“Best of Breed” is a good strategy – especially for something so important!  It may seem more expensive on paper up front, but in the long run, you will almost assuredly get a better result from a third party product.

Thinking you are doing a proof of concept (POC), when in fact it is just a custom demo. 

We almost always encourage customers to do a POC. We like to pick the hardest problem they believe they have and work with them to prove how it can be solved with TM1. But this process is often hijacked by our competitors into a situation where a skilled presales person from the vendor takes your data away, builds something with it and then presents it back in their tool to you. This is not a POC. This is a custom demo.  Such a custom demo does not prove any concept, unfortunately. (Believe me, I used to be a presales guy).

Why doesn’t prove anything? The first reason is that as a customer you didn’t watch them build it so you don’t know really how hard it is to build. Maybe they had a dozen people somewhere working on that thing. Second, what did you actually prove? Every software vendor will be able to demonstrate their product with your data in some way – but they haven’t had to prove how difficult it is to maintain or how it will integrate in your ecosystem or any number of other things that a crucial to success in the real world.

My recommendation is to sit down and build the POC together with your vendor – see every step of the way how it works. Also, use your full dataset – you need to find the edges of complexity where most products hit the wall and using a small sample dataset does not do that. Maybe add in some modelling complexity like foreign exchange calculations or allocations. Take note of performance and if it degrades significantly with added complexity and size. A couple of days doing this can save you months of heartache down the line.

In the end, try not to be dazzled by “complementary features” that don’t matter much or don’t differentiate the products. Instead, focus your POC on one of the Hard Problems of FP&A that are relevant to you and do a co-build POC focused on that.

In the end don’t take a software vendor’s word for it (including mine!) – make them prove it.

As a reminder, the Hard Problems of FP&A so far are:

• Hard Problem #1: Integrated Business Planning
• Hard Problem #2: What-If Analysis and Scenario Modelling
• Hard Problem #3: Excel Hell and Power BI Purgatory
• Hard Problem #4: Recognizing Easy Problems.