Treasury Today Country Profiles in association with Citi

Close to the edge: improving period-end processes (part two)

someone filing a folder in a filing cabinet

The financial close process can be marred by a lack of documentation and too many manual processes. Not only is this inefficient and costly but it can also impact on a company’s ability to make accurate strategic business decisions. Automation may be the answer.

There was a corporate trend some years ago, especially when implementing an ERP system, to pay a consultant to create “the perfect process documentation”, notes Neil Kinson, VP EMEA for automation specialist, Redwood Software. When it comes to mapping out close period processes, quite often that documentation “never left the cupboard in the time since it was expensively produced” and as such it is unlikely to represent what is actually being executed.

In part one of this two part Insight, it was shown that 82% of financial close processes are completely manual and that up to 23% of financial close transactions that are formally documented are never actually executed. Kinson’s assertion seems to be right! The solution to this problem, he said, may lie in automation of some or all of those processes.

However, when tackling an automation project, a certain level of preparedness on the client’s behalf is required. “We have a maturity model that always starts with a ‘reasonable’ level of process documentation,” explains Kinson. “If each close is completely ad hoc or dependent on the knowledge of a small number of people it is potentially a point of failure for the business.” In part one it was explained that the majority of those seeking to automate are businesses which have already embarked upon a finance transformation journey and have indeed reached an acceptable point of departure.

Which comes first: optimise or automate?

It may be possible to automate some simple functions more or less ‘out of the box’, but Kinson feels it is generally not advisable to try to automate existing processes without first optimising them, commenting that “just because you have automated a process does not make it good”. The problem here, he admits, is that the best ideas for optimisation exist mostly within the people who execute the task; so much time and effort is required to execute those tasks by these people that there is rarely enough time for them to consider and develop those ideas.

It is often the case too, Kinson says, that the appetite for undertaking large-scale IT projects to optimise processes may be “somewhat diminished”, particularly where a huge tranche of budget has already been spent implementing an ERP system.

Ultimately, the problem is one of perception: firstly, automation of close processes is not an IT project in the traditional sense in that it is not a one-off ‘big bang’ type of exercise. Instead, firms should be seeking slow and progressive “organic improvements” to their close processes. By moving forward in this way – gaining a few quick wins by automating at a lower level – he argues that an initial level of automation can “free-up bandwidth to allow the business to focus on real process improvement”. A project to automate close processes will thus start with the “relatively low complexity but highly-iterated tasks”, thereafter building on what the company would like to achieve.

Clearly there is a fine balance to be made between ensuring that current business processes are good enough to allow it to start iterating those processes, and looking at how the business can move away from just automation towards delivering true innovation.

Because there is such a fine initial line between putting the effort in and reaping the benefits, Kinson warns the early stages of an automation project have to be executed by the vendor with “relatively little effort” on the part of the client. For Redwood, three key data points are necessary to be able to build an accurate view of the client, these coming from its ERP system, existing process documentation and high-level information related to actual processing costs. The combined information will determine the level of automation potentially achievable, what this means for the organisation, and what the required investment to drive that might be. “We will then ask the customer to tell us their worst case scenarios – the highest volume of processing with the most manual drudgery – which they believe cannot be automated. When we demonstrate that it can be done, it wins their confidence.”

This is just a starting point. The research carried out by Redwood, which is admittedly aimed at proving the benefits of process automation, suggests that typically around 70% of manual processes can be removed. In cash terms, for the average Global 1000 company, this could deliver annual savings of around £2.34m, rising to £3.93m for companies with more than 80 legal entities. Beyond the cash savings, all businesses that are required to produce financial reports must do so with accuracy or face potentially severe consequences. Automation will relieve some of the pressure to deliver; as Kinson says, “the notion of a single version of the truth is heavily dependent on a consistent process application”.

Financial close process automation software is offered by a number of vendors including Redwood, BlackLine, Trintech and Chesapeake as well being offered as applications within ERP packages such as SAP and Oracle.