How data helps manage the balkanisation of tax functions

Like many of us, I’ve been watching the Euros. The 24-team format does not sit well with me. I was fairly sure that when I first started watching football (I recall the doomed 1988 campaign, Van Basten volley and all, rather too well) there were not even 24 countries in Europe.

As with many memories, this contains a grain of truth. I came across an old atlas the other day. The map of Europe today looks very different from my schooldays. The vast monolith of the USSR, beaten finalists in 1988, no longer dominates all of the right hand page of Europe; bottom centre, Yugoslavia (eliminated by England in the qualifiers) has vanished. In 1988, these supra-national states were on the verge of disintegrating even while the tensions between the vastly different groups they contained were smoothed out by the block colours of the map.

While I love a good analogy, it might be stretching a point too far to use this as a parallel for the tax function in a large business. But what on the outside looks to be a single functioning group is often, seen from within, broken out into ever more specialist domains, each with its own ‘language’ and culture.

But (and this is perhaps stretching the analogy to breaking point) turning the page in that atlas is a map showing the road and rail networks, connecting all the states in a single communication web. In the world of tax and finance, there is a similar connecting function in organisational data – accounting systems, commercial platforms, operational databases. All of which are used, to varying degrees, across the different tax and finance teams.

Often, the consistency of data used between the teams is poor. Each will pull the data it needs, knock it into shape, use and store it – oblivious to the very similar requirements in another team. Transaction data, for example, is integral to VAT compliance. But in a slightly different form it might also underpin the calculations for digital services tax… or, cut differently, capital allowance calculations.

Many organisations are starting to realise the value in consolidating these various approaches to data. One business, for example, is starting with creating a data store for the various sources used for Indirect Taxes – including the esoteric information needed for intrastats and currency conversions as well as the invoice data. But this has been built with future requirements in mind. Not just Digital services tax, which shares a similar dataset, but also for operationalising transfer pricing adjustments – and for creating the tax provisioning calculations. This has been achieved by keeping the data structures flexible and building in sufficient capacity for change.

Obviously, this approach also has down-sides. Gathering and consolidating these requirements can be time-consuming; and tech builds are often prone to delay or misinterpretation. Most organisations can tell horror stories about over-spec’d IT builds that fail to deliver on their game-changing promises.

New technology helps avoid these pitfalls. The cloud provides inherent flexibility in data capacity; APIs enable different ‘building blocks’ of technology to talk to each other; machine learning and AI enable technology to make ever more efficient use of our data.

For example – I have seen cases where organisations have used VAT transactional data in areas as varied as consolidating vendors as part of a procurement process, and reviewing fixed asset spend. We are seeing this approach in action at ARKK, where businesses are extending the for:sight platform they use for VAT compliance to other parts of the wider finance function.

To be clear, consistent use of the data that underlies a tax function will not see the different specialisms of tax and finance dissolving into a single team – any more than we will see the re-emergence of the supra-national states. After all, it is a truism to state that regulations get ever more complex; and this argues for more specialisation, not less!

But working from a common data set means that many of the inefficiencies can be ironed out. Less manual data manipulation ultimately means more time on the more interesting advisory and expertise focussed work. After all. The atlas may now be more complex (and provide many more teams to slip up against in the Euros). But it is infinitely more interesting.