AI in practice: Practical applications for accountancy

Technology business concept
istock_joztar_aw
Tom Herbert
Acting Editor
AccountingWEB
Share this content

While the wider media is awash with breathless articles proclaiming how artificial intelligence (AI) will take over the world of work, the concept has only just begun to permeate the accounting space outside the Big Four.

This ‘AI future’ can seem quite nebulous and far away to your average high street firm, but according to one leading practitioner currently using AI, there are practical, affordable solutions available now for accountants to apply, using techniques they may already be employing.

Speaking ahead of the Alternative AI for professional services event Becky Shields, partner at top 20 firm Kingston Smith, told AccountingWEB that it was important to cut through the hype and demystify what AI is currently doing for the profession, and what it could do in the future.

For Shields, the millions of pounds spent by the Big Four firms on AI solutions has skewed accountants’ perspectives, and there are reasonably priced, flexible tools firms can adopt now that could make a difference in the way they work.

Sample selector

One practical example of how Kingston Smith is using AI is in their audit function. The firm has partnered with North American AI provider MindBridge to use the tool as a sample selector.

“Rather than our trainees using random number sampling or a sample based on what they deem to be risky transactions, MindBridge ranks all of your transactions on a risk rating, and then depending on whatever sample size is you pick it gives you, say, the top ten riskiest transactions,” said Shields.

“It picks a sample that stands up to scrutiny by the regulators. If someone asks me why we have audited a particular sample, I can explain the computer-based technique which is a lot more robust than saying ‘one of my trainees picked ten transactions’.”

Machine learning

One of the things that has impressed Shields about the MindBridge tool is its learning capability.

“Over time MindBridge will improve its ability to identify risk transactions,” she said. “Using machine-based learning techniques, the software will learn more about what a ‘normal transaction’ looks like and get even better at spotting the risky ones.”

“Mindbridge will also learn from the weighting applied to each risk factor by its users, again helping to assess the riskiness of each transaction more accurately.”

Client conversations

While that is one of the practical applications of how the tool is currently used, the application has also given Shields a lot more to talk about with clients.

“Being able to look at an entire population of transactions and explaining ‘those transactions that are deemed risky and why’ is quite a useful conversation to have with our clients,” said Shields.

“HMRC is exploring these tools as well, so it’s useful to highlight any issues early on and say ‘this is the sort of thing that might get picked up if anybody else does a deep dive of your transactions and to advise our clients on how to improve the quality of the data’”.

Benefits

With a client base made up predominantly of small and medium-sized enterprises, for Shields and Kingston Smith some of the added value of AI comes in being able to bring big data analytics to the table for small businesses.

“For SMEs, AI and big data doesn’t necessarily exist because they’re filing abridged accounts and not publishing their statements, so benchmarking data is limited.

“Some of our clients don’t want or need an audit, but do see the value in us performing a deep dive into their audit trail, so we’ve actually sold that as part of the solution as well. Equally, AI comes up in tenders. People ask specifically about computer-based audit techniques, so the benefit is that we’ve got a real solution we can talk about”.

AI for everyone?

For Shields the message is simple. Accounting firms don’t have to spend as much as the Big Four on AI to compete.

“Terms like AI and big data analytics are banded about, but some of it is stuff we’ve always done as accountants,” said Shields. “It’s affordable to implement, the training is simple, and in terms of cost, we are talking about less than an hours’ time for a partner which we haven’t had any problems passing on as it stands – in fact, it has almost been a negative cost implication for us so far.

“I’d say to people don’t be afraid of it, it’s not going to cost you a fortune and it’s not difficult to implement”.

 

To learn more about how AI is being applied in the accountancy sector, register to attend the AI for professional services event.

About TomHerbert

About TomHerbert

Tom is acting editor at AccountingWEB, responsible for all editorial content on the site. If you have any comments or suggestions for us get in touch.

Replies

Please login or register to join the discussion.

avatar
22nd Sep 2017 11:09

"If someone asks me why we have audited a particular sample, I can explain the computer-based technique which is a lot more robust than saying ‘one of my trainees picked ten transactions’"
It is still a random sample. Yes, it is generated by a computer programme, not human.

Thanks (0)
22nd Sep 2017 20:13

I hope this is the first in a series as the article seems focused on the AI applications related to audits. The vast majority of uk accountants don't do audits any more.

Thanks (0)
to bookmarklee
25th Sep 2017 10:26

Fear not Mark - there is plenty of non-audit AI content in the pipeline.

Thanks (0)
By DJKL
22nd Sep 2017 22:48

The great advantage of random sampling is surely that it is random.

I can see the point of targeted sampling re some substantive testing, monetary unit sampling of debtors was a targeted sample, with larger debtor balances having greater probability of selection , but for compliance testing I would get worried that I was being "gamed", especially if my use of AI for such a purpose was being discussed with clients.

Having said the above I have not worked in audit since 1999, and even then they were getting scarce re the firm's clients and tended to be more industry specific, so I likely am not the best versed in suggesting concerns, nor am I versed in modern audit approaches.

In similar vein it is even longer since I studied statistics, but I presume if a sample is not random then normal calculations re sample size and confidence go out the window and the AI amends the sample size to compensate for the fact that the sample is not random?

Thanks (0)
24th Sep 2017 21:41

An interesting article.

AI and machine learning in the accountancy industry runs the risk of remaining nebulous. For accounting software to truly move into the realms of AI there has to be rich and well defined syntax and semantics to the data input into and generated by the system.

As a software engineer who cut their teeth in the specification driven communications/networking industry and who is now working in the accountancy profession, I find very little evidence of that being the case.

In general software needs strong contacts/interfaces and clearly defined data sets/data vocabulary upon which which meaningful decisions can be made by computing algorithms. Then the mass adoption of those software and data contracts allows large populations of users to benefit from the collective advancement of best practice.

The concept of double entry bookkeeping which is the fundamental principle which all current market leading accountancy products revolve around, is probably not enough to facilitate the introduction of AI technology to any meaningful degree.

A syntactic and semantic layers needs to be developed on top of double entry book keeping and this needs to be defined and published as an industry wide standard. Then software companies can build tools that might possibly give us what we think AI means.

Thanks (0)
to carlaccounting
26th Sep 2017 09:36

"The concept of double entry bookkeeping which is the fundamental principle which all current market leading accountancy products revolve around, is probably not enough to facilitate the introduction of AI technology to any meaningful degree."

So for most small businesses with very simple transaction processes the development of AI in an accounting context is meaningless?

Thanks (0)