ExcelCheck FAQ

What are the benefits?

Who is the ExcelCheck service for?

How are spreadsheet errors found?

How does model audit work?

Is shadow modelling a tried-and-tested method for finding spreadsheet errors?

Why is ExcelCheck different from a model audit by an accountancy firm?

How are errors classified?

What are the most common errors?

What are the most common errors in financial models?

What have we learned from academic research into spreadsheet errors?

What are the benefits?

  • Waste less time
  • Find peace of mind
  • Focus better

Who is the ExcelCheck service for?

You have an Excel model or spreadsheet with important results that you are relying on – maybe others are also relying on it.

So:

  • The results of your spreadsheet have important implications
  • You think there are errors but you don’t know exactly where
  • You are not dedicating enough time to checking the accuracy of your spreadsheet
  • Your reputation depends on the accuracy of your spreadsheet model
  • The reputation of your company depends on the accuracy of your spreadsheet model
  • Your spreadsheet model contains complicated business and financial calculations
  • You don’t think someone else will understand your spreadsheet model
  • You have inherited a spreadsheet model built by someone else
  • You are happy with the structure of your spreadsheet model but want reassurance there are no errors or potential errors
  • You are really not sure whether your spreadsheet model is fit for purpose

How are spreadsheet errors found?

All spreadsheet audit work uses a mixture of software tools and human analysis.

This is because computers are better at mapping and looking at patterns across hundreds if not thousands of spreadsheet cells.   But humans are better at understanding the objectives and judging the effectiveness or otherwise of particular logical choices.

Humans also have the advantage of experience and intuition.

ExcelCheck uses software tools during the review but is mainly based around the spreadsheet rebuild carried out by a human modeller.

To initiate this process, we ask the model owner important questions about the spreadsheet objectives and context.

 

How does model audit work?

Essentially there are two ways of finding errors in a spreadsheet model:

  1. 'Cell-by-cell' model audit
  2. 'Re-performance' model audit (also called shadow modelling)

In our view the re-performance method is creatively positive as opposed to cell-by-cell which is slow and very tedious.   This is important, because there is more risk of errors in slow and tedious work.

We also believe that the cell-by-cell review is technically flawed because it is too aligned with the initial thought processes.   Re-performance poses the same initial commercial questions but allows for an alternative/independent set of thought processes.

ExcelCheck uses the re-performance audit method – this is also the less costly alternative.

Is shadow modelling a tried-and-tested method for finding spreadsheet errors?

Re-performance (or shadow) model reviews are commonplace for most or all of the auditors.

The first time we heard of the technique being used was when the UK government sold the first tranche of student loans – this was in 1998/99.(see gov.uk/government/news/sale-of-mortgage-style-student-loan-book-completed).

The financial model describing this package of loans was subject to a re-performance model review prior to sale.

Why is ExcelCheck different from a model audit by an accountancy firm?

The big accountancy firms and select others provide model audit services.   These are mainly aimed at giving written assurance to lenders at transaction events such as financial close, merger or acquisition.   These are expensive services because effectively they are insurance policies.

ExcelCheck is not an insurance policy and does not seek to offer that same level of expensive assurance.

ExcelCheck is what is called a "agreed upon procedure review" and the agreed upon procedures are to check the spreadsheet is "logically and arithmetically correct".

ExcelCheck re-performs your model and reports all the same results to you without the expense of a written assurance letter.   In most cases this is perfectly sufficient.

So you get all the practical real assurance on your model without paying for the expensive insurance policy.

Here are some examples of ‘big four’ accountancy firms offering model audit:

And here are some examples of other accountancy firms that also offer model audit:

And finally, a smaller independent:

How are errors classified?

The expert in the field of the taxonomy of errors is Ray Panko of the University of Hawaii.   Panko has produced a holistic and generalised taxonomy, shown here:

 Taxonomy of spreadsheet errors, by Ray Panko of the University of Hawaii

Taxonomy of spreadsheet errors, by Ray Panko of the University of Hawaii

In our experience of looking for spreadsheet errors through model audit of various forms we differentiate between 'errors' and 'potential errors'.   Using Panko's taxonomy above we can see this described as Quantitative Errors and Qualitative Errors.

In practice Quantitative Errors, as described by Panko, or simply 'errors' for us, alter the results in some way.   What Panko calls 'qualitative errors', and we describe as 'potential errors', may not immediately alter the results, but could do as the model changes.

An example of a potential error is a hard-coded number in a formula.   The results may be correct at the time but not necessarily later.   Similarly for a formula using an incorrect cell reference which in the first instance may have the same value as the correct cell.

What are the most common errors?

According to Ray Panko some of the following common errors yield either immediate errors (quanitative errors) or potential errors (qualitative errors).

Notwithstanding, here is a list of common spreadsheet errors as defined in research carried out by Powell, Lawson and Baker (*):

  • Logic error - a formula is used incorrectly, leading to an incorrect result

  • Reference error - a formula contains one or more incorrect references to other cells

  • Hard-coding numbers in a formula - one or more numbers appear in formulas and the practice is sufficiently dangerous

  • Copy/Paste error - a formula is wrong due to inaccurate use of copy/paste

  • Data input error - an incorrect data input is used

  • Omission error - a formula is wrong because one or more of its input cells is blank

What are the most common errors in financial models?

Hera is a list of common errors from spreadsheets used as financial models:

  • Incorrect summation formula
  • Incorrect interest rate calculations
  • Non-balancing balance sheet
  • Circular references caused by use of carried forward balances
  • Incorrect debt balance calculations
  • Negative cash balances
  • Debt principal repaid greater than debt amount borrowed
  • Sources of funding not equal to uses of funding
  • Incorrect depreciation calculations
  • Incorrect tax calculations caused by:
    • adding rather than subtracting deductibles
    • deducting accounting depreciation rather than tax depreciation
  • Incorrectly calculated effective interest rates on debt 

What have we learned from academic research into spreadsheet errors?

This is the subject of a specialist research document by Dominic Robertson and contains some very interesting and profound results.

Free quote service

Lazuli Solutions offers a free quote service.

Simply upload your spreadsheet model and we will give you a quick call to discuss it with you.  Then we will go about getting you a quote and delivery estimate.

 

Alternatively, use this form to get in touch.

ExcelCheck contact form

Name *
Name