top of page
  • Writer's pictureDan Beck

4 Questions to Ask Before Starting a Big Data Project: An Accounting Perspective

Updated: Feb 1, 2021

I received an email recently from Harvard Business Review, regarding the four questions to ask before starting a big data project (Use This Framework to Predict the Success of Your Big Data Project,” by Carsten Lund Pedersen and Thomas Ritter).

These questions were:

1. Is your data valuable and rare?

2. Can employees use the data to create solutions on their own?

3. Can your technology actually deliver the solution?

4. Is your solution compliant with laws and ethics?

I thought these were fair questions to ask, but wanted to add a few of my own thoughts as they apply to the accounting industry.

Is your data valuable and rare?

For most firms, you have more than a decade of data. This includes timesheet data, account data, billing data, payment data, tax data, ledger data, CRM data… the list goes on!

The problem is that only a handful of firms actually know what they have! As accountants, we use data to help our clients, but we don’t use our own data to analyse our own businesses??

Most firms can calculate their profitability, but can they do it per client or per job? Do they know how long it takes for clients to return queries? How many times jobs are picked up and put down? Is work being done at the right levels? Would you make different decisions about resource allocation if you had this information at your fingertips?

Most firms have the data to be able to analyse this, but there are gaps in capabilities. The firms who do take the time and effort to analyse are able to make better decisions, and more timely!

Is your data valuable? Yep, but is it rare? On the macro level, probably not, but in relation to a particular client, team or team member, absolutely! Who has a better history of a client, their behaviours and habits than their accountant?

Can employees use the data to create solutions on their own?

This is a great one, and I think important to consider. Intelligence tools that are only usable by a select few create a very reactive process for internal and client improvement. Think about it: you need to change a report to make it specific to a client or a process - if you need to wait a week to get changes made, are you going to use the platform long-term or just find work arounds?

Any Business Intelligence (BI) solution should come with a side of training, and not training on how to use the final platform, but training on how to enhance the solution, how to build and augment. We’ve been using this strategy for the last couple of years - help develop the platform and then provide training on how to augment as part of the project, as this ensures it can continue to be used and improved internally with a guiding hand as necessary.

Can your technology actually deliver the solution?

There are so many BI platforms offering different levels of functional complexity, and the solutions required by clients have varying levels of difficulty. This makes it hard to obtain a working demonstration of your specific solution and not just a Proof of Concept (POC). However, most reputable providers will be able to demonstrate how they have worked with similar situations and apply them to your own.

Another issue is the longevity of the potential technology. Is it custom-built? Is it built from off-the-shelf components? Are there other third parties that can help you with the solution to minimise risk?

Is your solution compliant with laws and ethics?

If you want to understand why this is important, read this article about Target and the use of business intelligence: How Target Figured Out A Teen Girl Was Pregnant Before Her Father Did - Forbes

Just because you can, doesn’t mean you should, and the article linked above makes it really clear why this should be a consideration for every project.

So how do you get started?

Always start with strategy. What is your intent, and what is the ROI of the project? What is the payoff and the impact for your business?

The easiest and most comprehensive way is for PT 2.0 to help your team with the strategy, planning and ROI, and if it all stacks up we can then help you build the solution and train your team on how to use and extend the solution themselves. We work with dozens of firms across Australia and New Zealand on reporting, analytics and data integrity, so please reach out if there is anything we can assist you with.


bottom of page