Recently I wrote a post about the accounting profession and its future in light of artificial intelligence (AI) and software vendors’ eventual and unavoidable intention to apply it to problem areas that accounting professionals usually deal with. My post certainly got people thinking and talking.
One stream of thought came from an accountant situated in Australia who has built a small but effective practice off the back of Xero’s product. Said accountant was adamant that I was totally wrong. Basically his angle was that software vendors have a history of overstating just how disruptive their products actually are and that businesses will always need a trusted human adviser who can give them advice after analyzing and synthesizing a client's data.
+ Also on Computerworld: Automation and the end of accounting +
I gave several example of where technology has been used to automate formerly manual processes, but said respondent pushed back against that and continued to assert the belief (deluded, in my view) that somehow accountants are miraculously exempt from this trend.
I note that there has been a move in recent years for accountants to introduce fixed-fee and purportedly “value-add” offerings to their clients. My experience with these approaches is that there is very little value-add, and all they are doing is changing the revenue model—the fundamental service being offered remains unchanged. The fact that software vendors are increasingly able to bite off even greater chunks of core accountant business would seem to undermine both the standard billing models and fixed-price ones.
3 classes of accountants
My assessment to that person was that there are probably three classes of accountants. Firstly, those who simply want to continue doing what they’ve always done. They're comfortable, and they don’t want to change who they work for or what they do.
Then there are those who are leveraging the new platforms and using them to build what they suggest are fundamentally new approaches to accounting. As mentioned previously, often this is really just lipstick on the pig and doesn’t mitigate the risks they face.
The final group, and in my assessment it’s a small group, are those who really see the writing on the wall and are fundamentally changing what they do. They’re the rare accountants who are actually entrepreneurial and can offer strategic advice, governance-level service and real partnership to their clients.
I had an unsolicited email from another accountant. As I opened the email, I figured it would be another defensive note bemoaning my comments. Nothing could be further from the truth:
I believe there are large swathes of what traditional accountants do that is ripe for disruption. And yes the main accounting software providers are ideally placed to exploit this. Albeit they walk a tricky tightrope at present not wanting to disrupt their cash cow too much before they can jump to the new and monetize that sufficiently as a replacement.
While probably considered heresy by most in my profession, I also think you are bang on re your comments re the types of accountants. My observation over (too?) many years is that most accountants suffer from a affliction called being “comfortable enough”. They are smart enough to have made themselves a good living but precisely because of that and being comfortable they are not alert enough to change and motivated to take action.
That comment was largely focused on small accounting practices, so it was really interesting to get a note from someone who spent years building a cloud practice for one of the large “Big Four” accounting firms. As he explained it;
In the BIG accounting firms the cash cow and bulk of revenue stream comes from the Audit services. This is where they use the pyramid approach with one partner on top and lots of minions at the lower levels to manually bill tons of hours. ... Many functions like Sarbanes-Oxley depend on the integrity of this process.
Enter Cloud Computing. Now armed with my knowledge of the process definition and the power of cloud computing I have infinite computing resources available that I didn’t before and could run through the transactions for an entire period (week, month, even a year) in very short timeframe and know with a high degree of certainty if the numbers being reported are accurate. There would be no need for the minions of human resources working and charging for the tons of hours and armed with Cloud Computing brute force could actually have better ability to either dispute or attest to the financials being reported.
That certainly sounds disruptive. So, of course, the question needs to be asked whether this has already happened and the business of the Big Four accounting firms has already changed. Not so, says my correspondent:
The large accounting firms so far have not pursued this automation, as they don’t want to jeopardize the huge revenue streams. Each of these accounting firms have over 100,000+ employees in 140 countries. There would be a massive impact on the tens of thousands of clients that utilize their services—huge changes that would be immediate in their ramifications.
And so the status quo—both for the big accounting firms and for the smaller ones—exists. But for how long?
Is there a bait and switch coming?
Let’s get this clear. Once the big accounting software vendors get sufficient scale, and that scale sees the data residing in the cloud, they are ideally positioned to apply machine learning to that massive pool of data. And by doing so, they can create highly accurate algorithms that, over time, will deliver advice that betters that which accountants can give.
It is similar to the speech recognition journey over time. Speech recognition by computer has existed for decades, and over that time the accuracy rate has shown exponential increases to the point where, today, most of the good speech recognition platforms offer higher accuracy than humans. Slowly, but inexorable, we’ve seen machine beat humans at their own game.
So, if we assume that this type of self-learning, incrementally improving and machine-generated learning can be applied to the problem of giving small businesses advice, then we must be close to the point where vendors have enough data to do so. Xero has a million customers in the cloud, and Intuit has well over 2 million—that’s an immense amount of data that they can churn away on in the background. Of course, no accounting software vendor is ever going to come out and admit that accountants are, after all, one of their key routes to market, but you’d be a fool to think they’re not thinking about this stuff.
How long will this take?
Bill Gates is (in)famously quoted as saying that, in the technology industry, we overestimate what we can achieve in one year, but underestimate what we can achieve in 10 years. This situation exists in this case as well, but the issue here is less about quantity of data and quality of the algorithms that can be created and more about when vendors will have sufficient scale or enough alternative route to market to forego accountants as a channel.
Imagine Microsoft or Google acquiring Xero and applying its own number-crunching chops to Xero’s mass data and then delivering business-facing advice. Google and Microsoft already have the channels to market, and arguably, there would be little to stop them from circumventing, in one fell swoop, much of an entire profession. It’s an intriguing thought—and a chilling one for accountants.
So, what’s the likely scenario?
It’s not a zero-sum situation. Accountants will continue to exist for the foreseeable future, but I see a number of changes slowly occurring:
- Accounting software vendors will introduce these higher-level advisory offerings gradually, but consistently.
- The generation of accountants that either won’t or can’t change will retire, sell their practices or gradually move to focusing only on the clients that simply won’t use an accounting vendor.
- Those who remain will have to, very rapidly, develop the ability to think innovatively and entrepreneurially such that they can take those vendor-generated pieces of advice and contextualize them for their clients. The jury is out on how big, and how lucrative that opportunity will be.
Reflecting upon this, my contact with the history at the big four accounting firm had a pretty dire prognosis for the industry:
Basically, the disruption is imminent, as the technological capabilities are already in place for either one of the Big Four or a new Uber-type player to disrupt this market by automating a significant portion of the auditing function of these large accounting firms. And the impact and changes will be dramatic and immediate due to displacing the lower level repetitive accounting/auditing functions and the sizable associated volume of billable hours, thus reducing this lucrative revenue stream that these firms rely on.
All food for thought, and it’s going to make for a fascinating journey.