Few in number will be the businesses that have never used a pro forma document, that key part of all types of business management for which Cut and Paste in Microsoft Word might have been specifically invented. It also seems like the type of job for which the next round of new business tools, such as Augmented Intelligence, Machine Learning and Robotic Process Automation (RPA) should find ready application.
One of the companies looking seriously in this direction is KPMG, the global accountancy, auditing and business management consultancy firm, which has more than a passing interest in the subject. In fact, much more. It is not unreasonable to assume that the company lives or dies by its skills in producing and managing its own version of the pro forma document.
Within the company these are known as Engagement Letters, and each one sets out the precise and detailed terms under which any part of KPMG, in any country where it has partnerships in operation, interacts with its clients and customers.
As such, they carry the details of every nuance of the relationship between the company and a client and because it is a global business, if any relationship extends beyond national boundaries, all the vagaries of different national business and tax law, business regulation, best practice compliance and a host of other areas of legal intervention have to be accommodated.
This is then compounded by the fact that KPMG is a global operation of individual partnerships, sometimes described as similar in style to a McDonalds franchise. Each partnership in each country has its own channels and tracks, and each has to align with national practices, and then also try to align and present a united front in that country, and then globally.
Within the framework, these are letters that can literally be worth millions, and they have to be got right, for every participating partnership, every time.
Cut and paste construction still has a minor part to play, but as Nick Maxey, who heads up the on-shore development team in the UK for KPMG, explains, the company is now looking towards Appian, and its low-code BPM/RPA applications development environment to provide the tools with which to compose and manage the production of these important documents. He and his team are already starting to look at other areas where the low-code approach may find a use:
We have been using Appian internally for the last year to provide some internal audit information, in particular trying to optimise things like engagement letter creation and quality. There’s a lot of scrutiny going on at the moment around audit quality in the UK.
Get that letter right
This is something of a hot button for KPMG at the moment, particularly when it comes to engagements for accounts auditing, where the accounting community in general is coming under increasingly forensic scrutiny. According to Maxey, the company is doing all it can internally to address perceived gaps to ensure quality is there, improve turnaround time, and optimise contact time between key staff, customers and clients. In addition, he aims to take some of the monotony and rigour out of achieving such compliance.
The start of this work was rolling out a confirmations automation workflow, where customers and clients are asked to confirm details of the commercial engagement. This was extended so as to gather the inventory counts, look at stocks and warehouses, look for particular papers and encapsulate that information and upload it into a system and compile it into the audit engagement.
From this came the creation of Engagement Letters, the actual contracts between KPMG and its clients. Some 90,000 of these are produced each year, and to date there has been an element of human gauging of what the levels, the terms, might be within a certain parameter set:
What we’re trying to drive is the narrower definition of what an Engagement Letter would actually contain. That’s been the next big step and that’s taken quite a lot of work to do to bring the business in and get the buy-in. Now, there’s higher level interest, they want the streamlining, the efficiency, the quality. It’s just trying to get everything lined up so that you’re not trying to push an engagement letter that doesn’t match a particular engagement.
There are a significant number of quality gates to pass through following initial construction and passing in draft form. They are then vetted, verified and signed off by the Offices of the Counsel before going out or signed by the partner. It never hits the stage of being anything other than a working paper until those quality gates have been passed. To date it has been, Maxey observes, quite a manual process.
Compared to the manual construction of an Engagement Letter, a click-through process with a couple of drop-down options can save significant amounts of time. He reckons it is, typically, in the region of from three weeks work to around three hours, a saving which brings with it the obvious cost savings.
What’s been quite inspiring is how the business has bought into this and wanted to drive it faster and go deeper and go quicker. When they can see things turning around much faster than for example a custom code platform can do, which we’ve seen extensively with Appian, they already start thinking about more requirements, and then you start to have to really get into that project software delivery lifecycle discipline.
And here comes RPA
The company is also working with work flow practices for client engagement and managed services-type engagements. It is already delivering applications such as due diligence solutions and quality check processes to two major banks. A third stream that is now being developed with Appian is RPA orchestration, says Maxey:
We’ve got some relationships that we’re starting to develop and build out now with (Appian partner) Blue Prism. We have some UI path in the estate already. About two years ago we had a number of tools across workflow automation and RPA. We’ve been trying to find the best of breed and consolidate down on those, and then build out our delivery capability and excellence around those. We’re seeing the focus now on Appian as one of our primary tool sets and then extending into Blue Prism as and when appropriate.
Maxey said that one area where RPA may find a role is in improving the quality of data coming into the channel and ensure that anything generated meets required standards, regularising the way that the information is gathered, orchestrated and controlled so that the loop-holes are at least smaller. He acknowledged that this leads directly to discussions on when/if you can take the human out of the loop. His view is that this will not happen in the near future, though it may come sooner than he envisions.
He also sees possibilities for using RPA tools even earlier in the process, pushing back into the accounting processes themselves:
It’s an interesting area and I think the bit that’s going to catch people out. The ultimate decision point is where does the liability start and stop? Ultimately you have a partner who signs off on an engagement, code or report that is generated. At what point does that signature need to be there? At what point does someone need to take ownership for that bot or how those processes work? That is still the 80/20. We can probably do 80% of the work already through RPA and intelligent automation, but you are still going to require human oversight that says, ok, I’m comfortable putting my name against that reputationally.
Paradoxically, however, he still sees such developments having a positive effect on employment prospects, especially in the managed services business area, where there is growth in the need for human contact:
Those kinds of skills and traits are probably going to be more prevalent. It’s the cubicle type of workflow stuff, you might see less of that, but you’ll see more interaction services.