Date posted: 09/04/2020 5 min read

How agility produces better financial models

An off-the-shelf product is impossible with financial modelling. That’s why an agile approach is useful to ensure it is fit for purpose.

In Brief

  • Financial modelling teams spend a significant amount of time exploring and fleshing out model requirements.
  • A flexible approach is effective because it ensures stakeholders can buy into a model’s specifications.
  • Working through multiple iterations of a financial model means its final form will better reflect the business’s nuances and quirks.

If you ask anybody from my team, “What is the last thing you do before delivering a financial model to our clients?” you’ll hear something like “Open Excel and build the model.” They may deliver this with a wry smile, but it’s a core value that has been drilled into them since the day they joined the team.

The team spends a significant amount of time exploring and fleshing out model requirements. We prepare an unambiguous description of every aspect of the model, and we call this a specification document. And all that’s before we even touch Excel.

By the time that specification document has been iterated upon, updated for comments and further requirements from various project stakeholders, and is finally finalised and fully formed, it can account for 30-60% of the total time and cost on a project.

This may sound like a lot of time is spent without actually opening a modelling tool (and it’s not always Excel) but it’s an investment that pays real dividends. The model blueprints have been finessed and agreed on, and a skilled modelling team can rapidly build the model, perfectly aligned to spec, to deliver the agreed outcome.

The role of our senior team members is like an architect’s; we create the plans that underpin your house so it provides you with the features you need.

We are so confident in this approach, it’s taught in PwC’s Professional Model Build Course to more than 100 new and aspiring financial modellers every year. It’s a process that delivers the required model functionality while managing the significant associated risks of any project.

This approach has saved my skin more times than I can count in my 19 years of building models. It’s been battle hardened – by me and others in my profession – and will continue to be used into the future.

But that said, we’re seeing signs the world is changing. And modellers are changing with it.

Horror! Clients don’t always read the spec doc

Often a specification document is not fully understood or read by the ultimate model owner. But faced with the daunting task of trawling through a long spec doc, they will give a green light to progress to the build phase of work.

A modeller may then do a flawless job creating a masterpiece model, exactly to spec. But through no fault of their own, they fail to hit the bullseye because the target has shifted. All that hard work is for nothing, as the model owner’s first real exposure to the tool’s functionality doesn’t align with their expectations.

That usually results in significant changes being required at a very late stage of the project when its project budget is severely depleted and the risk of errors being introduced is high.

We have learned that a more flexible approach is more effective. It ensures stakeholders are brought along on the journey and fully buy in to the model’s specification.

There is a fantastic opportunity for our profession to move towards an agile world. Instead of a specification being a single document, it needs to be given the room to evolve in small sprints over time. Its specs can be drawn from a collection of modules, prototypes, workshops, emails and other documents that are collaboratively developed throughout the project.

This sounds like a small change but to some – including me not that long ago – this is tantamount to blasphemy.

Let’s get loose

Where there is this mix of client engagement, commitment and underlying team technical and project management skills, we are able to provide increased value to clients. But such an approach requires deep experience in modelling projects and exceptional project management skills.

It’s a fundamentally new approach. We still rely heavily on the tried-and-tested framework, however the client experience is vastly improved, and the frequency of interactions and updates is much higher than our traditional approach.

Revisiting the analogy, rather than just playing the architect in this scenario, our senior team also take on the role of site manager and oversee an ever-changing set of plans. This control and regular communication is critical in order to continually triage the development priorities as the model architecture evolves.

Working ‘with’ instead of ‘for’ our clients

As budgets continue to be stretched (and get tighter) and model outcomes are needed sooner, stakeholders are demanding value for money and greater clarity on progress. In the past, there was an appetite to wait while the work happened in a dark room. Now there are evolving prototypes and a need to show continual progress.

“In the past, there was an appetite to wait while the work happened in a dark room. Now there are evolving prototypes and a need to show continual progress.”

The market is demanding our profession innovates to stay ahead of changing requirements. We must find new ways to provide value and insight to businesses.

We have an obligation to improve how we work together. My team and I are excited to embrace this change and evolve our profession by developing new approaches, methodologies and leveraging new technologies to use on our engagements.

If we all embrace this change, we will all play a role in driving our industry forward.

Points to remember

When it comes to financial models, the underlying business or problem being modelled – and the purpose of the financial model – is unique. That means an off-the-shelf approach is impossible.

When building any forecast model, you have to fully understand the nuances and quirks of your business and the context in which you operate to ensure the model is fit for purpose.

At PwC, we are finding that with new approaches and technology, we discover the underlying business drivers through multiple iterations, allowing us to continuously improve as we edge closer to the final model.

This idea of multiple iterations, using feedback on the previous version to drive continuous improvement, is central to Lean management ideas, first developed in Japan in the mid-20th century. A lot of these were later absorbed into the “agile principles” for business, which put “satisfying customers” at the top of the list.

While an agile approach can benefit finance modelling, don’t forget these three important points:

  • As with any critical part of your organisation’s ecosystem, continual review and inspection is required to ensure forecast tools reflect the underlying operation.
  • Models need to be flexible to incorporate the changes to your operations as your business evolves, however…
  • Care must be taken to manage the risk that comes with making these changes.

Ian Bennett FCA leads PwC Australia’s deals modelling team and has more than 19 years’ experience as a professional financial modeller. He was assisted on this article by Tim Linke, associate director, PwC Deals Modelling.


Have a financial modelling or spreadsheet question you’d like answered in Acuity? Send it to [email protected] with ‘Ask Ian’ in the subject line.


Storytelling with Data: Let’s practice!

Storytelling with Data: Let’s practice! offers a clear guide to creating graphs and visualisations that make sense of data.

Download the ebook from CA Library