Somehow the lack of attention on the modelling appears to be justified by how poorly the estimates (assumptions) are done.

I agree that standards are important and really appreciate the work that Navigator is doing in promoting better practices in financial modelling.

I have also seen the results of an effort to implement standards across multiple government departments and my view is that the issue is people related.

1) Decision makers don’t understand financial modelling, the complexities or the need to do it better. Consequently they don’t fund it.

2) Financial Modelling is often seen as the dirty job that you have to do when you enter industry, which you then get past.

3) Clients are often reluctant to pay more than the rate of a first or second year graduate when hiring one of the big firm. This not only reinforces the second point above, but means that mastery of the full set of skills is rare, and often not valued (unless a PI claim is lodged).

4) On the client side, many of the people involved barely understand basic finance concepts well, and the analysts may be barely hanging in there. Getting them to understand the core financial principles can be hard given the project’s time constraints and when you try to get them to move to best practice you can lose them completely – they think you are being a nitpicking perfectionist and lock you out.

5) Many clients don’t like paying for financial modelling full stop. If you think a job will take 200 hours to do properly, then they may give you 20-30 hours if you are lucky. I can’t tell you how many times I have a $500m plus project give me a few days to deliver something to take the client the next step. Do my models follow core modelling principles – yes, do I have time to follow a prescriptive standard – not a chance, the client won’t pay for the extra time and I’m out of business.

6) Simply following a standard won’t produce a commercially sensible model. I have reviewed and fixed up models built to standards by most of the main players that were overcomplicated and nonsensical in terms of commerciality. I am doing such a job for the next fortnight. The model was of a high standard, but is almost completely unusable by the client.

So, to take a leaf out of the software development industry, a financial model should always be designed with the end user in mind – they may be using it for years. Standards that solely promote ease of use and audit by other professional modellers will by default fail.

What other methodologies have you evaluated ?

Nick

Having devoted the early part of my professional life to the topic my observations, from the point of view of a transaction professional are:

1) Any ‘modelling methodology’ needs to be able to be written on 1 side of A4 – otherwise it gets too technical, boring and easily forgotten in the heat of a transaction.

2) Presentation + Audibility = Confidence

3) ‘Speed to build’ doesn’t rank in the top 10….

The main commentators in the field come at it from different angles, all of which have different objectives. An interested, non-specialised observer should probably take a bit from each…

Nick

Some practical differences that may occur may be the degree of application of the above principles e.g.
1) reference to blank cells (common for opening balances in greenfield projections) – do we block it out (shade grey etc) or not?
2) what is considered as hardcodes? Is (“1″ considered a hardcode in a period counter formula? Do we have to name range “1″ to “One”?

I favour a more “principles based” than a “prescriptive based” approach and on principles/spirit there are not much difference.

On another note, I do appreciate the efforts of Navigator in advocating its modelling techniques to the general public (client and non-client) through its website. This does help in raising the general level of good modelling practices.

Would definitely welcome the opportunity to see some of BPM and FAST models being presented or made downloadable.

You seem to have misunderstood my comment as I was only referring to the _presentation_ of the FAST standard – not the overall impression. From a model auditor’s perspective (which is the prime focus of Corality) I completely agree that the robustness of the FAST standard way exceeds market average. However, from a decision maker’s perspective I am of the view that FAST makes calculations sheets look overly technical compared to for example the Navigator methodology. From a pure formula calculation point of view I would say that both methodologies have equally strong focus on transparency, simplicity and flexibility.

We’ve been using the standard in our bank for a while now and I have to say it beats any approach to modelling I’ve seen elsewhere. I only wish that more of the models we receive in the bank were built using the standard then it wouldn’t take us so long to read and understand them and it would make it much easier to change them. I look forward to the day when more companies adopt it.

thank you for your comments around financial modelling standards. I agree that FAST is a good public initiative but it is also worth highlighting that all participants in the panel have very similar standards that they are using. Arguably the Navigator ‘standard’ is the most well-known of these standards and we at Corality share that methodology when it comes to development of new financial models as it has (in my view) the best compromise of productivity and presentation. The FAST standard in my view has a higher productivity factor but the presentation is not as neat and professional.

http://www.navigatorpf.com/resources

and indeed there are standards followed in the finanicial modelling society.

Please visit:

http://www.fast-standard.org/

I really like what the FAST standard is trying to do – http://www.FASTstandard.org – several firms that I do business with are implementing the standard and a standardised approach seems to have a big productivity impact.