2 questions you should always ask your financial model auditor
Ari Gold: Never scared to ask the hard questions.

2 questions you should always ask your financial model auditor

As someone who has spent almost 20 years building financial models, and software to improve the way financial models are built, it's probably most apt to describe my relationship with the model auditing sector as volatile.

Each time I've recommended a model auditor to a client, I've made a fast friend out of that model auditor, and for a brief moment in time we've bathed in the warm feeling of making money together.

But then there's the other side of the equation, the side in which I've become obsessed with reducing financial model build time, cost and risk. And this isn't something that excites my model auditing colleagues at all. In fact, it's safe to say that it scares the hell out of them.

In hindsight, I was very naive to think that I could just find cool ways of improving financial modeling and everyone would jump on board. I never really thought about the fact that there are many businesses out there that rely on models being confusing and risky to justify their fees. But I'm now very aware of it.

Put simply, there is a fundamental conflict of interest between financial model auditors and their clients, because the profitability of model audit businesses is directly related to how long it takes to audit financial models. And as a result, model auditors continue to fight evolution, even though I believe they are doing so at their own risk and at risk to all of us in the financial modeling sector.

In recent years model auditors have been forced to reduce costs, primarily because the industry has become a savage price war and some of the more aggressive parties started outsourcing the majority of the work to cheaper labor in developing countries. And once one party did this, many were forced to follow to protect their margins.

These market-driven improvements in efficiency have resulted in some cost savings to model audit clients, but there certainly hasn't been a complete pass through of these cost savings to clients. And, more importantly, there has been limited investment in obvious initiatives to improve model auditing processes, for the reasons I've mentioned above.

With this in mind, I wanted to focus on two areas which could so obviously be used to reduce financial model build and review time, risk and costs, but remain improperly addressed by most leading financial model auditors.

Are you open to creating a single global spreadsheet modeling standard?

I have quite a unique perspective, having been an investment banker, professional financial modeler, and senior software engineer. So when I see a spreadsheet I see a computer program built by entering formulas instead of writing code. Yet unlike code-driven software, most spreadsheets are random and chaotic, making it almost impossible to reuse, collaborate and share.

In 2002, I left investment banking to start a financial modeling consultancy called BPM, and the first thing we noticed was that the main cause of inefficiency and errors in financial models is inconsistency. Even in our small team of consultants, everyone had different views on everything, so it quickly became very hard for anyone to confidently collaborate. It was a mess.

The problem was further exacerbated by the fact that each of our clients had different views on everything, as did each advisor we worked alongside. So working on deals was like being part of a team in which nobody spoke the same language. And I still believe this is how most teams using spreadsheets operate today.

The answer to this problem was, and remains, both obvious and simple - a global spreadsheet modeling standard which governs how complex spreadsheets are built, without limiting what they are used for.

Back then, we were far too naive to appreciate that most financial modeling organizations would fight to the death rather than work together, so we created the Spreadsheet Standards Review Board (SSRB), wrote a comprehensive standard as a starting point, and invited every competitor and advisor we knew to join us to mould it into a single global standard.

This is when we first experienced resistance to evolution, and it was swift and brutal. Everyone we met agreed with the concept face-to-face, before vanishing, only to return a few months later with their own competing 'standard' or 'approach'. Ironically, our attempt to create a collaborative improvement to financial modeling became a key driver in the sector now containing a complete mess of differing approaches, ultimately making things even more confusing for the businesses that rely on us to make things better.

This experience changed me as a person and as a businessman, because it taught me that there are such deep conflicts on interest in the business world that there are parties out there motivated to fight evolution even when doing so is detrimental to their clients.

I remain chairman of the SSRB, but for the past 5 years we have opted out of the messy online debates about which 'standard' is best, because I am convinced that few of the parties that have created a 'standard' or 'approach' are truly interested in collaborating to create a single global standard. I know this because I've asked them, and continue to ask them, and their position never changes, only their opaque justifications for not doing so.

To put it another way, many model auditors would rather convince their clients to conform to their specific gauge of railway track, which is intentionally different to that of all of their competitors, as this differentiation limits their clients' ability to stop using them without significant cost and risk.

For the past decade, I've repeatedly heard the originators of competing approaches state that "it's better to follow an approach than not have any approach at all", which is true, but this statement is nothing but a diversion from the fact that they are refusing to work together for the benefit of their clients.

In a perfect world in which model auditors genuinely want to make things better for their clients, a global standard would be something on which they would all be proactively collaborating, to the benefit of everyone in the financial modeling sector.

Are you embracing new technology which aims to reduce financial model build time, cost and risks?

The second tell-tale sign that your model auditor isn't acting in your best interests is a lack of interest in embracing technology which aims to reduce financial modeling time, cost and risks.

It is an undeniable fact that a significant part of the time involved in building, maintaining and reviewing financial models is the repetitive and manual nature of editing static spreadsheets. So an obvious starting point to resolve this issue is removing high-risk repetition using technology. Yet the last 20 years have revealed that the model auditing sector as a whole has little interest in doing so.

Personally, I find it absolutely unacceptable that Spreadsheet Detective has remained one of the most common spreadsheet review tools. It's a nifty piece of software, but it's barely changed in over a decade and really should have represented a starting point for more advanced technology than becoming a stalwart of the model auditing sector.

I wrote a LinkedIn post in February this year about the huge potential benefits of using modular spreadsheet development to create and disseminate pre-audited pieces of financial models, thereby significantly reducing the risks associated with building financials models. However, most model auditing businesses remain disinterested in exploring this potential.

Perhaps the most personal example I've experience of a model auditor resisting technology was after BPM released bpmModules back in 2005. bpmModules was the world's first modular content management and sharing system for Microsoft Excel, significantly reducing the time taken to build integrated 3-way financial models. But it used heaps of range names and entered unnecessarily complex formulas, which were quickly used a basis for publicly disparaging the system by one particular Australian model auditor.

I still recall receiving an email from one of BPM's clients, asking me if I'd seen the document that was being emailed around by this model auditor, warning them against trying bpmModules. It even contained the phrase "The idea makes sense in theory, but in reality it's not possible to systemize the development of bespoke financial models", before providing a comprehensive list of the limitations of the system along with a dire "use it at your own risk" warning.

I'm the first to admit that bpmModules had some limitations. These limitations took us over a decade to resolve when developing Modano. My issue is, and still remains, with the covert and malicious manner in which a highly-respected model auditor went about disparaging software which provided a benefit to their clients, albeit a first release.

Fast-forward to 2017, with Modano having been used to build models worth over $1 trillion in 20 countries - in less time and with less risk - and I still regularly get frustrated emails from colleagues and clients informing me of ongoing scaremongering by model auditing firms that have never tried and/or don't understand Modano. Only it's now starting to reflect more poorly on them than Modano, as the platform unquestionably reduces model build time, cost and risks.

We're better off friends

Model auditors are a fundamental part of the financial modeling sector, providing important third party review and assurance services so that financial models can be relied upon when making important decisions. They're the mash to my bangers, for want of a better phrase, so there's no upside in us not getting along.

The reason I'm writing this article is that I believe model auditors fighting to maintain the status quo are, ironically, creating significant risks for everyone in the spreadsheet-based financial modeling sector. Fewer businesses can afford high-quality financial models, while more financial models contain errors, thereby motivating businesses running on spreadsheets to look for non-spreadsheet based alternatives. And then we all lose.

A global spreadsheet modeling standard and technological advancements may reduce the number of hours charged by traditional sweat shop-style model auditing businesses, but they would also make high-quality financial models and model audits more accessible to more businesses, resulting in a much larger financial modeling market. And then we all win.

I really hope we can all put our egos aside, stop bickering over manufactured trivialities, and collaborate to move our sector forwards into the 21st century.

Onwards

Michael Hutchens

CEO | Modano

When you are talking about "standards" there are two broad types that need to be distinguished. There are low level standards that concern formula lengths and which columns to use, and there are high level ones about how to go about modelling complex systems. The former are not very interesting and the latter tend to either be vague or very complex. Template driven systems like Modano are excellent if you can find a template that closely matches what you are trying to model. But different models tend to be interested in different things. In order to cover most of them the templates tend to become very complex and cover many things that are not of interest in a particular situation. And most spreadsheet errors are introduced when modifying models rather than creating them. I actually think the problem is Excel itself, and its simplistic structure that comes straight from Visicalc. But figuring out what the replacement should really be and then implementing it to incorporate the polish of Excel is a big ask. I was involved with one failed effort years ago (after I started the Detective), there have been a few, Quantrix is still around. One local effort that I rather liked was Sumwise, but like many others it has proven to be too difficult. Given that, the Spreadsheet Detective does a pretty good job of addressing the basic flaws in Excel. I believe that a tool like it is useful even with a highly templated approach. Although the Detective has been around almost as long as Excel(!) it is continually improved, and I just shipped a new version this week. But it is focused on working with Excel's traditional model. I think at the end of the day modelling, like programming, is just difficult to get right. Programming has not fundamentally changed in 30 years either (although systems have become much more complex). There are better tools and best practices, but it will always require good people to produce effective models, regardless of which tools they use.

Hi all, This topic is something that has been spinnig around within each financial modelling team in any kind of company which develops/needs financial models. I have been working for 10 years with various companies, in different sectors, with different kind of business and in different countries throughout the world, and trying to create "a single global spreadsheet modeling standard" in financial modelling seems a very bad solution. Financial modelling itselfs is a dinamic tool, with its own DNA based in its developer, the business unit within the company where he is working at and the type of company. I have met engineers, mathematicians, accountants, "tax-guys", financiers and other graduates with different perspectives, and each of them have their own needs when discussing about standards and they need their own "audit". And these needs are in evolution with an unique constante: cash flows. After working with all of them I came up with an easy solution: standard framework within the spreadsheets (such a simple set of four colours and a simple structure of the columns and rows within the sheets), with a simple summary, and no more rules. No limitations in formulas, no limitation in number of rows, no limitation in breakdown, no limitation in nothing... The reason behind this is because each of them need their own information, and in case a thir party would try to limit them, the Financial Model will only be valid for the Financial Modeller who limits this tool. I also have engaged some investment banks and big4 consultants to develop Financial Models for Public Tenders and at the end I need to create my own sheets within their Financial Model, due to there are many internal procedures which companies do not want to disclose. Said that, my suggestion for all those model auditors is what I have done all my life when working with financial models that are not mine. I do "extract" those inputs which origin the business itself from the financial model, I do introduce them in my financial model, I do check the differences, correct them in my model and finally check them in the Cash Flow Statement. If both provide similar results the Financial Model has my approval. So, sorry Michael Hutchens, I have a different vision, anyhow, I do encorage you to do your best in order to obtain that standards within the sector, but I will always prefer not having limitations. Regards, Fernando

Alfred Griffioen 🌏

International marketing & commercial distribution expert

7y

That's a new profession for me: Financial modelling auditor. Where can I register for the course? Perhaps good to give an overview of Financial modelling first...

Emmanuel de Villeneuve Esclapon

Managing Director I Deal Execution Services I Securing Transactions I Securing Payments I Digital M&A @The Deal Execution Group

7y

We will review any financial model built by anyone using any standard or tools. But Model Review (as usually performed) will remain a commodity, with very low prices, very high risks, a rather low cooperation between clients / modeler / auditor, and as a result very little time / cash to devote to improvements. The key is to do it differently.

To view or add a comment, sign in

More articles by Michael Hutchens

Insights from the community

Others also viewed

Explore topics