Generation CFO's Christopher Argent talks with Henri Taipale, founder and CEO of Qvalia, about the changing world of finance, CFO's new role, the nature of finance data and its implication for automation and RPA. Listen to the podcast here.
Listen to it here:
Here's a full transcript of the interview:
Christopher Argent: I'm really looking forward to talking to you about do what you're doing with Qvalia and kind of the journey to get to where you are today is so many CFO and finance professionals they are probably on a similar journey but you have kind of led the thought leadership.
I think in some of the changes that you want to make in finance operations you know be great to hear your story on that. So to kick off. Where did your career start. And in a nutshell how did you get to the point where you were founding Qvalia.
Henri Taipale: Yeah. It's actually quite a long story I suppose.
First of all I never really had a proper job. I'm coming from an entrepreneurial family. So I've actually been self-employed all my life and it goes back to me being involved in my father's company and he was working with with taxation. And from there I learned that you know people make a lot of mistakes with tax. It's a very complicated field especially international tax as well. And I started to investigate how could we actually start to systematically find these various types of mistakes and did that.
It was a long learning experience and learning how to actually do a data extract from from large ERP systems and then with do statistical analysis identify transactions which could be deviating in some sense and from that perspective then we learned that OK we can actually find decent stakes using these statistical methods and those statistical methods we could actually apply to different fields as well not only taxes but do you know overpayments double payments wrong pricing from suppliers and so forth.
And from there the business actually started going into companies and helping them sorting out these transactions scale it up into the Nordics but actually already probably about five years ago started pondering about that. Hang on a second. This doesn't really make sense. Companies with 80 90 percent electronic invoices they should not be committing these mistakes over and over again.
This again led me to start investigating that. What's wrong in managing transactions in organisations and what we've basically came up with was that you know even if the transaction itself is digital we haven't changed our processes. The processes look exactly the same as they used to be in back in the 80s. We just manage the transaction digitally but we're not the digital rights organisations. And that led in return to us building a platform where we actually manage all of this information in a more secure way.
Christopher Argent: Wow that is a great story even in a nutshell.
So let me just go back a bit because you probably was through. I do know that about 20 years there. If I could but a couple of things that you mentioned there one sort of entrepreneurship to getting tax wrong and then kind of the fundamental shift in I don't know you know what would you call it transaction automation or even transaction. Revolution in a way but the first one you know you came from an entrepreneurial environment.
Given what you know now about sort of core value working in the CFO space. What sort of advice would you give on being more entrepreneurial and what would you think the benefits are if we do do that?
The reason why I ask that is that we see a lot of leadership courses talking about a growth mindset and a lot of the teams that I've worked in even within big companies have effectively been startups and you've had to just you know do everything to get stuff done. It'd just be interesting to get your take on you know starting out in an entrepreneurial way.
Henri Taipale: I mean to be honest I think we have to be entrepreneurial and especially when the information that we manage is digital because everything is getting more and more fast paced which means that whatever information are managing the CFO or in accounting that's actually operational data as well.
It's the same information which means that we don't have time anymore to reflect and think we have to be there with the operation and actually also more customer oriented in order to better engage and run our businesses.
Christopher Argent: Yeah absolutely. You know you remind me of a couple of conversations I had with some of the data scientists that I worked with in the past when we were running sort of analytics projects and they just view what we do in finance completely differently.
It's almost like maybe as accountants you know we're trained to look at things in a particular way you know how we structured data how we structure processes you know what sort of output we provide. But when I spoke to some of the data scientists out there they really challenged why we did things in that way I think the simple answer that I found is we look at things from a compliance perspective whereas they look at things from a sort of data perspective know if they've got that skill set that statistical analytics feels that they look at it from a mathematics perspective. So it's a completely different take on it.
I would have thought you've been on a sort of similar journey where you're asking simple questions that finance people you know can't necessarily answer because they have always done it that way before?
Henri Taipale: Yeah definitely. I mean first of all I might not get popular by saying this I never liked accounting already back at the university. I've really had a hard time with accounting right at it. It's super ironic since I'm working with it now today but I think that was actually a benefit because I was questioning it all the time.
I did not believe in set standards and ways of working but I was questioning it all the time because I said I didn't really like it.
This has led me to a couple of revelations along the way and made it back to what you said there as well. Okay. Why do we do accounting and this is something I've asked a lot of CFO was and I've sort of like double checked my own opinion on it. And we do accounting for a few reasons.
One is because the law requires us to do it. And the second one is for some kind of internal control. We need to know our revenues and our costs and in some sense we need to categorise it now you know if you're closing of accounts you don't really need that many accounts in your chart of accounts that you do actually for their internal control part. And when we talk about what is accounting it's a data categorisation. It's nothing else.
It's actually I don't know if the term accounting is even getting obsolete because it's only data. It's nothing else. It's gross and stables and we're just categorising this is really important for people to understand it's not special data. It's just data.
Christopher Argent: You know I recognise what you're talking about there and I don't think it's too controversial to say that you didn't like accounting and you're trying to kind of improve the role or the learning that you were having because I talked to a lot of people and there's almost like this emerging theme of reluctant accountants and it's almost like the reluctant accountants are the ones who are saying look I don't want to be doing low value tasks in this role. I don't necessarily want to spend two years in every finance operations role before I eventually start talking to the business and being commercial. I think it's people like yourself who don't like accountancy and are trying to think of new ways to go about what we do which is reporting whether it's from a compliance perspective or a management perspective and adding value and partnering with the business.
I think it's people like yourself are absolutely in the driving seat for the future. I've encouraged a lot of reluctant accountants I suppose to speak at the various things that I do because I think their voice is really important. And one example of that right.
You mentioned tax and starting out in the tax journey. There's a lot of tax accountants out there and there's a there's a lot of bookkeepers out there who do tax returns. You know there's still growing in a number of cases you know you see a lot of people complaining about the return that could have been better if we'd have done something in a particular way or it's just plain wrong.
You know the tax accountant has made an error and I think that's sort of driving us towards a more automated environment. And there are companies out there who are providing that offering. Did you learn something specific about the tax problem that you took forward?
Henri Taipale: Well actually I did. I think you know tax as a consequence and a lot of the systems today they're trying to make that tax sort of like you know in the system I would explain this sort of like for me when you're managing the information it's like who's the buyer who is the seller? Where is the product coming from where is it going? How is it being used?
That creates the premise for how the tax will be decided if he will be the VAT or something else. Now it's not on the transaction where that should be determined. Today we send the invoice with the VAT on an invoice. But the problem there is that okay; is that correct?
The VAT you know you have to determine all of these factors in order to understand that value added tax on that invoice.
Now that means that for systems to understand this they need to understand who is the buyer who's the seller who is is. Are they registered? Or is the product going from and where it going? In order to automate the tax.
Now the problem if you don't do that to data management properly from the start or from the get go you know we'll have problems in automation so it's always about the quality of the data the fallout of the structured data that helps the automation part and that's why I think a lot of automation initiatives they just take a slice and they sort of like some optimise some certain function within the larger process.
But what people need to realise is that hang on a second this whole process might need to change in time.
Christopher Argent: You know this leads us perfectly on to the third point that you made around process before come on to that you know get your take on sort of there's lots of buzzwords out there and automation is one of them.
But looking at a capability rather than a buzzwords in a robotic process automation RPA which for listeners who don't know it's a software that basically sits in your environment and it performs the key strokes that a human would would make to carry out a task that task within a process an A/P process or sales process you know to your point around maybe the process being the problem and these RPA projects not delivering as much value as you think they should you know what would be your advice to people who essentially looking at that RPA project because with have capabilities maturing and we're starting to see more use cases.
Great but we're also starting to hear about how not to do it and the failures around RPA projects said have you got a point of view on that?
Henri Taipale: Yeah, I mean first of all RPA, I mean, wonderful technology but to me it's a bit like duct tape. You know it's a quick fix for something and it can be great. I mean duct tape can also be really great but it's not something that you do normally if you do something; "I need duct tape".
First of all you should understand your whole process and you could understand that. Okay I need to fix this maybe with RPA from the beginning but I'm actually happy with this whole process. What is my long term goal in changing my data management on my processes. What's my long term goal here and you need to understand that long term goal and when you have sets that long term goal then you can come back today okay. What can I do today to just fix my process. What do I need to start changing in my processes or data management in order to drive me forward towards my goal?
People don't really take that time to do that and I understand everybody's super busy but you should not to go out to the market and say I need automation I need RPA you should start first by looking at okay what information on my managing and why.
Christopher Argent: Just to challenge your thinking on that one you know if I'm a CFO and I've heard about this capability that could potentially replace a person you know why do I have to look at the whole process or this longer term vision if say within two or three weeks I can put a bolt on my desk you know automate some of this low value work that say, you know, report create type role is doing, or an A/P manager is doing. What's the sort of duct tape scenario vs. this sort of wholesale process change?
Henri Taipale: Yeah. Well first of all if you won't do it your competitors will. We're all here in order to provide customer value, that should be the driving force for everybody.
And I think a lot of people should actually understand that even if they're just a small cog in the wheel in large organisations they're still there for the customer. And I think that's also true that we spoke earlier about the entrepreneurial part that is being entrepreneurial. Understanding that my function here it is vital it is important but maybe in some instances it's actually not important anymore. It's actually for the better for the customer. And then I need to prove it.
Then back to that duct tape idea I mean well first of all the managing of data in organisations that's for the leadership so a strategic question for every organisation and it's actually trickling down from their CFO to that actually everybody in the organisation.
So I think whether you start changing the processes I understand there will be a quite a big reluctance to change in a lot of organisations. "I'm already a leader" because they might be looking at their point result whereas again you need to do this because you need to survive on a five year term basis. Days go by.
An example that I always use is Nokia, in five years they went from almost 60 percent market share to 3 percent. That's extraordinary. I don't know if that has happened in history before but at least not in recent history I believe it's a tremendous change and these things do happen.
Christopher Argent: Yeah absolutely. It's a fascinating sort of journey that we're all on.
I think you know you having started this sort of so long ago you know makes it very interesting to see your point of view but what would you say is this sort of significant change because if 20 years ago or probably not long 15-10 years ago you were working in this of analytic space dare I say machine learning in a statistical modeling space. Why is it becoming such an important capability for finance now, the CFO is now, and where it wasn't 15 years ago?
Henri Taipale: I think the biggest change is probably the pace how information is flowing today.
I mean we have been in payment terms 30 days 60 days or whatnot because information was slow before. Now when everything is becoming digital, it's becoming more and more fast-paced.
We have the Internet of Things. We're starting to build digital ecosystems which means that from a CFO perspective you need to understand this. You need to understand what the product is how the information flows are going. What will be my pricing for my products in my organisation? What happens? How does that pricing drive customer behaviour like this as well? I mean it's all entangled. And the more digital it becomes the more entangled it is. Yeah. And it's not only about my organisation and it's not only about the customer. It's also about my suppliers, right?
Christopher Argent: Interesting when you talk about sort of suppliers and supplier relationships. I've got quite a strong view on that. So I worked for a company in the U.K. That was very customer focused but also very supplier focused.
Would you say that your approach to supplier management changes with this capability and is it a good or a bad thing? So I must feel that customers are number one, but suppliers are only a sort of one point one now rather than low down the list.
Henri Taipale: I'm sure there are differences between different industries. I'm generalising quite heavily but what I think is that we're all sort of like in a vertical.
My suppliers, I need whatever products or services I'm buying in order to package it in some sense in order to sell it forward to my customers. And the better this vertical is working the cheaper I can probably provide this to my customers. If I have a good relationship with my suppliers where I can actually automate all my purchases and administration.
Therefore I think the suppliers can provide their services and products for a cheaper price and I'm sort of like now going out guessing here but, have a look at Tesla. I mean what did they do with creating their own factories for batteries? Why is that? Because probably they can create their vertical and sort of like embedding it in their whole process, in order to provide the product to the customer.
Christopher Argent: Yeah, and I suppose in a manufacturing environment your sales and operations go hand in hand.
What can you produce as much as what can you sell and it's normally a supply chain that's the limiting factor there. So the more you can control your supply chain the more you're going to be able to sell.
I just think as a sort of particularly from the experience that I've had to learn partnership in the UK the you know the supplier was treated as fairly as the customer. And when it came to supplier transactions and dealing with queries and payment terms and things like that it was always done in a sort of fair way. You weren't always looking to sort of extend terms or slow payments to suppliers, as you know some companies may create a working capital.
Henri Taipale: It comes down to also friction. I mean you don't want friction because friction is the cost.
Christopher Argent: Yeah, absolutely, so you know that's a good intro into Qvalia itself.
I think we've touched on the process but you're almost suggesting that the process is sort of valueless, almost detracting from what we're trying to do. And I understand that the core value is presenting a whole new sort of set of rules around the transactional process.
Tell me a bit more about Qvalia specifically the problem you're trying to solve with it?
Henri Taipale: Yeah, I mean what we're doing on our platform is that we're still looking for the whole either on procurement or sale side and looking at consolidating the information that we're managing in that process.
So just a quick example; today when we're doing procurement, we're buying, we're ordering items or products and then we actually match it with the delivery and then we'll get an invoice.
Now you have three check points in that process, you could actually have more of those as well. The problem being from a data management perspective is that it's actually the same line items that you're managing all the time, regardless if it's the the order object or the delivery object or the invoice object and so forth.
Now you're multiplying the same information many times over and quite often in many different systems. We ask; why should we actually do this as the information we're managing is the same?
So we're consolidating that information management and also looking at the functions of the persons in the process now. Quite short, this is not actually the reality but a bit of characterisation of how I view it about the people in the process. We have people who are performers you know a person who is actually digging a hole. You have people who are information managers people telling that "OK you need to dig a hole over there" and then you have decision makers people deciding that the hole should be dug over there.
Most of us in organisations today, we're just information managers. We're just conveying information from one place to another which means it can be automated. The functionalities are actually the approval points in the process and the approval points or are of course the order and the delivery. But what else am I doing in that process because I'm basically just pressing an OK bottom. Simple as that.
But if you have a look at the other systems today they can be extremely complex. Make a lot of time for the people to enter in the correct data and so forth. And what we believe in is that this can actually be consolidated and simplified to the extreme. You know we just scratched the surface on this so far, on the market.
Christopher Argent: It's really interesting and I'm going to mention the B word here and I'm curious about your thoughts on this and it's not Brexit because we don't talk about Brexit, it's blockchain. Now the solution that you've talked about there, I've also heard blockchain consultants talk about where they're saying that you know we can create a frictionless transaction you know down our supply chain, and each transaction has to be agreed 100 percent at each stage. And that will be managed in a blockchain to eliminate all of these sort of supplier queries and the friction that you talk about. What's your take on blockchain and how is it different to what you're talking about with Qvalia?
Henri Taipale: Yeah, hopefully I'm not have to eat up my words here as it's on recording. Well, first of all blockchain is a really really good technology. Tremendous technology has definitely its use cases but. But first of all I'm probably an analogy again if I want to build a house. I'm not going to ask the builder if they're using a hammer or what type of a hammer they're using. That's quite uninteresting for me. I mean technology can be an enabler but when it comes to blockchain I think it has its place but centralised data is extremely good because it's efficient.
I mean have a look at Google. We all go to Google to do our searches because it's centralised and there's the value for that, and there's speed involved in having it centralised. So I think, you know, organisations need to understand where to use a certain type of technology.
But don't just run after the next buzzword because that won't work. You need to understand what tools will you be using for which function or a solution.
Christopher Argent: I like your take on that. And I think you know there's lots of people out there who are sitting on the fence when it comes to blockchain. I still don't see a use case for it in the CFO space. I still have the sense that it's the sort of solution trying to find a problem and if we are going to work differently as you sort of say fixed the process, well, do you need a blockchain to do that? We're talking about a bigger problem here and it actually comes around in the analytics as well and data management.
It's like if you really want to provide, you know, real time valuable analytics to the business then you need a steady flow of data. How does that happen? Well, it's probably fixing your processes and automating your processes, as much as it is around bringing in the data analytics and data management capability. So, yeah I kind of with you on that.
So just looking forward kind of trying to summarise that if, you know, if everyone goes on that journey that you're thinking with Qvalia and maybe not with proper train and then the other capabilities that's out there you know in 10 years time what does the finance team, finance operations team, look like and what's the sort of impact on the people who are working there now?
Henri Taipale: Yeah, I think that change would be quite tremendous especially on a 10 year scale.
Well, first of all if you look at what we're trying to do on our platform is that we can provide all necessary information on the transaction, forces the invoice, so that when it comes into accountancy you can just automate it. So basically information just magically appears.
We've talked about spend analytics that you have the line item information on exactly how many items did you buy last year who bought them or what price from different or which vendors, all of that information will be available but also making rules that if I'm buying iPhones, how do I want to have it in my accountancy? So, of course from that perspective, it will change and it will also change through automation. Markets will change in vertical integration as well.
I mean today's CFO's need to take a lot of different decisions on, you know, various types of solutions. And I think the more automated everything becomes, the more one supplier can actually provide it to the organisation, for the customers as well.
And then when we have a look at the CFO and what they're doing in their day to day work, it will be a lot more data-oriented. It will be a lot more closer to the customer, closer to the operational business. And in fact it is the operational business. There is no operational parts and administrative part. It all merges together because we're managing the same information.
Christopher Argent: That's a really good one and I think it's one of the messages I suppose I have to the finance team that you touched, and in operations not being sort of separated from admin, and potentially in my view, even finance.
So if operations have this capability first, where does it leave finance teams? Because if you fix the operations of the company and you fix the operational processes then the value add that we know comes in finance today will probably be automated out of that operational process.
And what I mean by that is embedding analytics in your operational processes. You know you wouldn't need a finance team to create a report for you on the same processes and you know it sounds a bit bleak but it's not if the operations capability gets there first then what does it mean for finance? And that's one of the reasons why I say to finance people we need to start being more data aware data literate, data savvy, and learning about the art of the possible in this space because it's in our absolute interest to do so we've got the most to gain and the most to lose.
Henri Taipale: Also I mean it goes back to universities and sports as well.
Know it's a long time ago since I was in sport but you know you can't just pop out accountant professionals and expect that market to exist. You need to adapt and change. And I said also going out of course and focusing more on data, we're going to need people to manage rules. We're going to need people to train and so forth. And we've got any CFO expertise in order to have the right pricing for the market.
I think we're going to have, you know, a and b test pricing on a totally different level. So we've been able to do it before because we have direct information of the effect of what we're doing with our customers.
Christopher Argent: Yeah I think as much as I'm saying you know we need to move into this space and it will significantly impact what we do if somebody within the company gets there first.
I always believe that there's a business partner role. You have to apply the business context you need the industry knowledge that domain expertise to help build the A.I. and to help work with the data. And that is a significant role. And you won't be touching any code even by doing that role. So yeah there's definitely a role for us.
You know there's no wholesale change here and I think some of the things that we've talked about today just to be sort of ultra clear is it's about the transactional process and the tasks within those transactional processes. It's not about the strategic plan or the like, you say this sort of strategic analysis that we might do on a whole industry or pricing, you know that's that's a sort of long way away when it comes to use in a or statistical algorithm.
Well, thank you very much for your opinion today on what you do is great. Hearing about your entrepreneurial past. I think everyone should take a leaf out of your book there to become a bit more of a reluctant accountant even if you love the numbers just to sort of have that slightly different mindset about what we do.
You know I love your sort of take on how we need to sort of radically change fundamental processes and starting with tax. I think you would have learned so much because like I say you know you still see problems in that space even though it's in a highly structured and arguably sort of rules based.
So it's ideal for automation and your sort of thought leadership on changing processes and how you a capability like value could help you with that. So all the very best with your journey. And thank you for being on the podcast today.
Henri Taipale: Yeah. Thank you Chris. Thanks for having me. And also thank you for a really good podcast. Thank you.
Interested in learning more about finance automation and how to transform your processes? Download our ebook CFO's Guide to Business Automation.