DATA STORIES | ACCOUNTING | KNIME ANALYTICS PLATFORM
Hard To Mess Up Those Numbers With Automation: The Accountant’s Perspective
My Data Guest — An Interview with Ali Igram
It is my privilege to welcome an accountant and a finance expert in this episode of my My Data Guest interview. Ali revealed the accounting and finance tools he uses, how useful and practical automation is and how to increase awareness for new advancements in the industry.
Ali Igram is a Certified Public Accountant and finance expert who is passionate about data science. His expertise lies in finance transformation and he leads such projects at Lineage Logistics, a company with generations of cold storage experience that benefits from cutting edge technology and innovation. Ali works for both worlds of data science and finance as he helps people to solve financial data problems and save time using low code tools by automating manual processes.
Rosaria: Tell us more about your professional tasks and responsibilities.
Ali: For many years, I’ve worked in traditional accounting and finance roles. I have since transitioned into transformation and helped lead finance transformation projects. For example, we conduct workshops to help map out our current processes and understand the pain points which prevent us from being our best. We provide project management to make sure we’re delivering on our projects and we keep everybody on task. There is a lot of communication involved in this role as well. We deliver training to upskill the organization on new tools including KNIME Analytics Platform. We also do some hands-on solutioning within finance and accounting using low code tools which is where KNIME comes into play as a favorite.
Rosaria: What kind of tools does an accountant use?
Ali: By far, our dear friend Excel remains the dominant tool in the accounting and finance space. It’s really being used for everything and the funny thing about that is that we spend so much time working with data but we traditionally use the same set of tools. We also use other Microsoft Office products like Powerpoint for reporting. Power BI and other visualization tools are increasingly gaining some traction for financial reporting as they are providing some benefits.
Rosaria: Excel always gets into a heated discussion. It’s a love and hate relationship. Some people love it, some hate it. Where are you on this love/hate scale?
Ali: It all really comes down to what we’re using it for. I used to believe that Excel was the answer to every problem and I think this is the prevalent approach in the accounting and finance world up until today. I was actually very fiercely defensive of my spreadsheets even though they were complicated and difficult to follow. A colleague of mine in a previous role confronted me about a complicated spreadsheet he didn’t understand. I defended the spreadsheet, but it eventually crashed and burned under the pressure of very large datasets. In the early years of my career, Excel got me through some things but also caused me a lot of pain. When I added KNIME Analytics Platform to my skill set, a lot of the pain points from Excel were addressed and new possibilities and capabilities opened up.
Rosaria: Which pain points do people usually encounter when they work with Excel?
Ali: Let’s face it, spreadsheets are kind of a chaotic free for all. The University of Hawaii did a study indicating 88% of spreadsheets containing errors. We can all relate to this: they are like a smorgasbord of hard coded values, formulas that are impossible to follow. There’s funny highlighting that you don’t really understand. There are leftover comments that are disconnected from each other. It’s really hard to figure out what’s going on and there’s no audit trail so you end up having to rely on what someone said that they did. And despite Microsoft’s incremental improvements, our data woes continue. Many surveys indicate that accounting and finance professionals spend between 40 and 70% of their time on data. It’s because we continue to work in the same way. I strongly believe that it’s time for a change. The heavy bulk of the work should not be in spreadsheets.
Rosaria: Some of your memes on LinkedIn are really attention grabbing. They both address the work of an accountant and pain points of using Excel. There is one specific meme that best summarizes why after a certain level of complexity, Excel sheet handling might become a problem. Does this meme come from a real situation/experience?
Ali: Yes. A lot of times in my current and past roles, my colleagues and I struggled with this. You needed an organized, structured step by step process. This one struck a chord with people as it’s not an uncommon situation. Excel is a flexible tool but it can lead to chaos and disorganization, especially with numerous data sources, nested if-statements and complex index-match and lookup formulas. You can’t really make heads or tails of what’s going on. It’s very bad from a control’s perspective for accountants/auditors and mentally draining to work with these files. I don’t think Excel was ever the right tool to handle a certain level of complexity. We had no idea what’s going on with these spreadsheets, and we had to ‘hope’ that everything was ok. Once you structure your data and give it an audit trail like in KNIME Analytics Platform, it really helps to solve those problems.
Rosaria: In the meme, you show a chaotic and confusing Excel part and the clean and nice KNIME part. Is KNIME really such a good alternative? What are the KNIME features or nodes that best come to the rescue in solving an accounting problem?
Ali: I think it is a good alternative to more complex spreadsheets because it forces you to be intentional and think scientifically about what you do. You’re selecting from these bundles of nodes, which are like puzzle pieces. With intention, you place each piece of the puzzle in place until the problem is solved. One step leads to the next until the workflow is completed, so there are no questions about data origin or what steps you took. It’s a different way of working that KNIME Analytics Platform promotes — the automation-first mentality.
Some of the most common nodes that will be beneficial are the Joiner node, the filtering nodes and the Rule Engine node. Loops and components present unique automation opportunities too. That’s why a low-code platform like KNIME Analytics Platform is so effective at automating that spreadsheet work.
Rosaria: When we talk about loops and components, we are basically moving onto the next level of abstraction: encapsulating nodes in other nodes or implementing loops, something that a normal data spreadsheet can’t do.
Ali: I think that’s where you get more creative. What do people want to do when they come to work? They want to be creative. They don’t want to do the same thing every month. That repetitive work becomes a drain on a lot of people. When you’re given these powerful tools, you can start to think about ideas that you didn’t think were possible. A tool like KNIME enables that for a lot of people.
Rosaria: Tell us something more about other accounting use cases where KNIME has been a little helper for the solution.
Ali: Journal entry preparation is a big one. It’s a pure data exercise where you’re moving data around and trying to get data into your system, and your ERP — whether that’s Quickbooks, Oracle, SAP or Netsuite. KNIME’s flexibility allows you to work with a variety of formats, systems, or data heavy internal controls like reconciliations where companies have the checks and balances in place. There’s a lot of different use cases within audit procedures. Financial statement validation and finance use cases like capital budgeting, investment analysis, modeling and forecasting are very common ones.
Rosaria: Can you also use KNIME Analytics Platform for auditing then?
Ali: You absolutely can. One of the greatest opportunities for low-code automation in the accounting world is the automation of audit procedures. You can use KNIME to set up the audit by creating lead sheets from the trial balance and save a ton of time. Instead of processing the files manually, you standardize that work to be a one-time effort. You can even mine texts with KNIME’s advanced document and image processing capabilities.
Rosaria: What about some use cases in finance where KNIME Analytics Platform could easily be applied to save time and generate more money?
Ali: I think data heavy exercises that are done manually in Excel are the easiest ones to go after. Manual journal entries are ripe for the picking. They are so problematic because they are high risk, time-consuming and prevalent. The financial analysts know their areas so well that if they can learn how to use a tool like KNIME, they would be able to put that domain expertise into action in their workflow. Reconciliation internal controls are also quick wins, because these can be handled using the Joiner node. KNIME can do a great job at getting that data for you with the Tika Parser and Regex Extractor node. If you can get a certain level of predictability, you could really supplement your contract and review.
Rosaria: How prominent is the use of text processing techniques in your sector?
Ali: It’s not prominent enough and we’re still buried with work. The power of text processing techniques could really help. We work with text all the time. We’re actually generating text: creating content to explain variances in financial reports. Natural language processing is an untapped opportunity and we ought to spend some more time investigating it.
Note. Do you know that you can now prove your knowledge of text processing with KNIME with the new L4 certification and complete the NLP learning path?
Rosaria: Do you only use ETL operations or do you sometimes build workflows including machine learning models?
Ali: Not really, but there are a lot of use cases for machine learning in accounting and finance. Finance is more well known for forecasting but accountants also try to use predictions in order to estimate the future payouts. Time series analysis could really help us with long term planning and more accurate accruals, but we don’t see too many projects of that nature in the industry currently.
Rosaria: Another meme that you posted, showed the same accounting chaos in 1999 and in 2022. In 20 years nothing has changed? What is missing in accounting to progress forward?
Ali: Everyone is overwhelmed. Overtime is par for the course in the field and we all feel the pain or have felt it in the past, yet we continue to work using the same tools we’ve always used. There is a lot of fear and anxiety around change, because of the criticality of accounting for companies and the far reaching implications of errors, but what we fail to realize is that being stagnant carries risk as well. Our problems continue to increase in complexity, but the tools we use to solve them have remained stagnant. It’s not a matter of capability because we have the capability. It’s a matter of recognizing the opportunity and executing on it. KNIME Analytics Platform is a key enabler by making programming easier for people. We need to evolve our skill set and be open to new ways of working if we’re going to gain control of our peak times and our own sanity.
Rosaria: Basically KNIME Analytics Platform could help you keep your sanity?
Ali: Exactly. A lot of the things we do are predictable and we do them on a recurring basis. So you can build those workflows to handle the recurrence.
Rosaria: So you need automation and control?
Ali: Yes, awareness of these opportunities and the willingness to take the dive and get started on learning.
Rosaria: You also recently wrote an article for the KNIME blog, where you played Batman. We never had a data superhero as a guest in this interview series. Would you like to describe to us the horror story in the article and what Batman had to do with it?
Ali: Well, it’s about an overwhelmed and stressed out accountant whose Excel files keep crashing. He is learning a new skill and saving the day for himself and his company by opening up new possibilities and automating journal entries. This is what’s happening nowadays. I’m a big promoter of empowerment for people to let them access more powerful tools and innovate. People want to be creative and solve their own problems. They don’t want to lose hours of their lives on repetitive, manual work. This is the situation in the accounting and finance world. This type of work is not very fulfilling but most people don’t realize what they could accomplish with the right tools and knowledge.
Rosaria: Can you name 3 nodes or features of KNIME Software you could not do without?
Ali: That’s a tough one, because I probably have about 20 that I couldn’t do without. I love the Rule Engine node. Nested if-statements were always a headache in Excel so the Rule Engine node is a nice organized way to have a complex ruleset that’s easy to follow and understand. Then there is the Joiner node, which I use all the time in my workflows. The joiner node is really a savior for accountants, as often we are combining data together from consolidation and sub-ledger systems, or trying to bring two datasets together that we want to reconcile. Since this is an accounting and finance focused episode of My Data Guest, we have to give love to the Math Formula (Multi Column) node. Doing math across all columns in your table in one shot can save a lot of time. It is great for avoiding tedious work and repetition, because it applies a formula across all columns that you select.
Rosaria: What about components? Do you use them in your work?
Ali: Yes, I use components, they are one of the best automation opportunities because they allow you to eliminate duplication of efforts. Let’s say that you develop a way to clean up a P&L transaction report: you could capture that workflow within a component and then use that component across other workflows that use that data so you or someone else doesn’t need to reinvent the wheel. Components also support user input, by allowing you to group several nodes together in one nice configurable window. For example, we can have a date configuration for journal entries date, an integer or double configuration for journal entry numbers, and a selection box for account all in one nice window.
Rosaria: It seems that in the name of preserving known running processes, HIPPOs in companies are often okay with a convoluted and clunky tool like Excel that lacks many capabilities: from automation to error handling and ML. As a domain expert in the accounting sector, what’s the best way to talk to HIPPOs into the adoption of visual programming-based tools?
Ali: Accountants are very conservative and there’s a principle in the accounting handbook called ‘the principle of conservatism’. Fear is the biggest obstacle to overcome. Accountants tend to believe that if something bad can happen, it will happen. But the complacency with the tools leads to many fire drills because the environment keeps changing. We need to emphasize that complacency is the real risk, and the current state chaotic spreadsheet free for all is not sustainable. The American Institute of CPAs estimates that 75% of the CPA (certified public accountant) population will retire in the next 15 years. So what will we do then? We will not be able to continue using spreadsheets for larger and larger datasets. We have to emphasize the risk of staying complacent, and provide some comfort that future ways of working are actually less risky with lower fire drills and error, and that people can be trained to use next-gen tools.
Rosaria: How much accounting do you need to know to build a KNIME solution for your accounting department? Shall we all start learning some accounting?
Ali: The domain knowledge definitely helps for many projects but there are some common themes for accounting projects that can help to simplify the requirements. Accounting workflows typically need a lot of checks and balances to validate the integrity of the workflow. Also, the requirements for accounting and auditing comes from accounting and auditing standards. There is no harm in brushing up on those areas if they are relevant to your project. Companies also have policies in place where they interpret those standards and apply them to a specific context. Those policies or manuals can be helpful. Often, you don’t need to do all that though, you could have simple exercises that are just joining and summarizing or reconciling data so some projects don’t require as much domain expertise.
Rosaria: The information and data you usually work with in accounting and auditing are very sensitive. GDPR in Europe and similar regulations in the US about data protection and privacy have accelerated best practices in data governance. What are the advantages of using a visual programming-based tool like KNIME Analytics Platform in that sense?
Ali: Having that organized and a structured way of working with the audit trail really helps. Everything can be very easily documented and structured in a KNIME workflow so you know what happens where. From a data privacy perspective, KNIME Business Hub provides some good access management features for workflows that could help along with more secure storage options. This is another problem area with spreadsheets, as data can be easily lost or transmitted outside of the organization.
Rosaria: We are reaching the end of our conversation. Before we say goodbye, could you give us an anticipation of your next meme?
Ali: I wanted to show a crowd of accountants that have adopted low code programming but I haven’t found that image yet. I am sure it’s out there somewhere. Sometimes these images come into your head but they don’t exist in real life and that’s the problem.
Rosaria: How can people in our audience get in contact with you and your work?
Ali: Feel free to connect with me or follow me on LinkedIn. Send me a message, happy to make new friends. I post regularly about automation with KNIME in finance and accounting. I’ll also be releasing a low-code automation course this month catered to the accounting, auditing and finance audience. So please do check that out if you are interested.
Watch the original interview with Ali Igram.
