Tech Unknown | Episode 6 | Season 2
Featuring guests Tim Crawford, Isaac Sacolick, Andreas Welsch, and Timo Elliott with host Tamara McCleary
It’s natural for people to be mistrustful of technology, especially when it comes to automation in the workforce.
But let’s ask a simple question: If you had to harvest an acre of wheat in an hour, would you rather use:
a) A sickle?
b) A combine harvester?
Most of us would opt for the latter: do the work you can, and let the machines handle the heavy lifting.
Now, imagine you had to process 50 terabytes of business data for your quarterly report. Would you rather use:
a) A spreadsheet?
b) RPA, machine learning, and AI?
Intelligent automation is poised to do for knowledge work what the combine harvester did for farming. Data-heavy jobs will be less repetitive and more efficient, but will undoubtedly require new skillsets and processes to keep up with the tech.
This episode, we hop in Tamara McCleary’s time-traveling Tesla to explore the history and future of work. We ask the experts how business leaders can prepare for intelligent automation, both on the technological side and the human side.
Listen to learn:
- How to get buy-in across the organization for your intelligent automation initiative
- How intelligent ERP can serve as the backbone for automation
- What businesses can accomplish with intelligent automation in place
- What skills and mindsets to train and hire to prepare for intelligent automation
Discover how intelligent ERP leverages artificial intelligence and automation to transform your business processes – in the cloud or on-premises.
About our guests:
“With intelligent ERP… you get to take advantage of newer technologies that are coming down the road when they become available, which is a huge improvement over the traditional approach, which could mean a 5–7 year lag in adopting new technology.” –Tim Crawford
Timo Elliott is the global innovation evangelist at SAP. He has spent over 30 years presenting to business and IT audiences in over 58 different countries, talking about digital transformation, AI, analytics, and the future of digital marketing. He also blogs at timoelliot.com.
Isaac Sacolick is the president and CIO of StarCIO and the author of Driving Digital: The Leader’s Guide to Business Transformation through Technology.
“The way we used to do things doesn’t work anymore. We need more accurate data, we need to be looking at forecasting more frequently, and the volumes of data we’re looking at are bigger.” –Isaac Sacolick
Andreas Welsch is the head of intelligent processes, SAP S/4HANA Product Management. He holds a PhD in information systems from the Technische Universität Darmstadt.
Did you miss our last episode?
Check out our previous episode with guests Carla Gentry, Iver van de Zand, and Timo Elliott: “Intelligent Analytics: The Search For Hidden Treasure In Your Business Data.” Click here to listen.
Episode 6 Transcript:
Tamara McCleary: Welcome to Tech Unknown, a podcast to prepare your organization for the tech-centered future of business. I’m Tamara McCleary, CEO of Thulium.
Our big umbrella topic this season is data. We’re digging into how sharing data across the organization – collecting, processing, and analyzing it – can increase efficiency, reduce costs, and improve customer service.
This episode, we’re going to dig into intelligent automation. Here’s Tim Crawford, CIO strategic advisor at Avoa, with an overview:
Tim Crawford: A modern ERP solution needs to have a couple of components, one of which is that it’s leveraging some of the newest technology available on the market today. Things like machine learning, artificial intelligence, RPA. All of these really bring together a combination that will really help out companies. They need to look for solutions that will address not just the problems that they have today, but also try and project out and understand what are the problems they’re going to have to address in the future.
Tamara: Before we dive in too deep, let’s start by defining some terms Tim introduced. When we talk about intelligent automation, we’re talking AI, machine learning, and robotic process automation, or RPA. Intelligent automation is unique to “knowledge workers”: that is, people who work with data and information, rather than tangible materials.
With those definitions in mind, we can start exploring how automation will change the job description for “knowledge workers.” The easiest way to do that – provided you have a magical podcast time machine – is to see how physical automation changed things for manual workers.
Let’s just fire up my time machine – it’s my Tesla, not Doc and Marty’s DeLorean. I’ve made just a few modifications to my Model X, and I’m ready to fire up this bad boy!
Ooooh, this is interesting! Here we are back in the pre-industrial era to see how people worked. Back then… uhhhh, I mean, back now… EVERYTHING was handmade, one at a time. Simple items we take for granted, like a pair of shoes, would take a single shoemaker hours of time to complete.
Cobbler: Here is thy right shoe, Milady. In but two fortnight’s time, I shall complete the left.
Tamara: Thank you, my… uh, lord? Sir Cobbler? Whatever. You get the picture. Having one person handle every tiny task it takes to make a shoe… well, it’s just not scalable. But as technology evolved, we were able to break down shoemaking into its individual tasks, create machines for many of them, and humans just fill in what the machines can’t do. Now we can turn out thousands of pairs of shoes a day, instead of one every two weeks. No offense, Sir Cobbler.
Cobbler: Thousands of shoes a day, you say? How could there be that many feet in the whole of the earth? And what, pray tell, is this black metal book thou hast?
Oh, this? It’s called a laptop computer… it, uh, well, you know what? Never mind, you’re right. It’s a lovely metal book the blacksmith down the road made for me. I just love metal, don’t you? Okay, I have to run now, bye-bye, thanks!
We’d better get out of here before I step on a butterfly or something and screw up our entire existence…
Oh, shoot! Dang it! I left my metal book, I mean my laptop back at the cobbler’s shop. Uh… no worries, it’ll be fine.
Okay, I digress… anywho, the point of that trip through time wasn’t just to freak out a cobbler – it was to highlight just how long a history humanity has with automation. Our latest evolution, using technology to automate knowledge work, is just one more step forward.
Here’s how Timo Elliot, global brand evangelist for SAP, described it in our first episode:
Timo Elliot: The history of computing is absolutely about augmenting human intelligence. Actually forget computing, the history of science. I actually find this fascinating. If you imagine agriculture, a long time ago, people were really restricted to what one person could do in a field. You know, you could try and plow, you had oxen to help you, but now one person with a tractor can do an unimaginable amount of farm work. I think we now have the opportunity to do the same thing but for knowledge workers. Clearly, knowledge work is very valuable, I believe that human beings are the most powerful technology we have. And the ability for these latest technologies, like artificial intelligence, to expand on what an individual can do, I really think we’re at the start of a new golden age for knowledge workers.
Tamara: Even as we acknowledge that intelligent automation is a logical next step, however, it’s natural for people to feel uneasy. Change can be hard, and scary, and nobody wants to think about losing their job to a robot – OR a piece of software.
So, it’s up to business leaders to not only invest in the technology for intelligent automation but also to help people embrace that technology and make the most of it.
Here’s a great tip for helping employees through the process from Isaac Sacolick, president of StarCIO and author of the book Driving Digital.
Isaac Sacolick: When you use the word intelligent automation, I think that works well when we’re talking about it as a platform and as a capability. I think it works terribly inside the organization. I think when the organization hears automation, it triggers two reactions. With the staff, it’s job loss, it’s change, it’s, “I gotta learn a whole new set of skill sets,” it’s, “What am I gonna be doing in the future?” and it’s, “Can this system actually do everything that I know how to do, and all the best practice, and all my expertise, and do it as well as I can?”
So, I think automation is a really bad word for CIOs to be using. And so, I think what you really need to do is focus on where people are needed, right? So, not what you’re not doing anymore, but what you will be doing, where your expertise is welcome, you know, what the new model is gonna look like, and how are you gonna serve customers differently, or how are you gonna be able to do things faster. Or maybe even it’s, you know, how are you gonna have a better work-life balance because you’re gonna be doing a lot more intelligent workarounds processing analytics or doing decisions that the machine can’t do very well, that you’re gonna be part of a learning organization that learns and changes with the advent of new technologies and what their capabilities are. So, I think that’s how you overcome things and start getting more alignment and buy-in and bringing some of the automation technologies into play.
Tamara: As Isaac is saying, it’s crucial to get executive and employee buy-in on the benefits of intelligent… you know… automation. Fortunately, there are plenty of potential benefits to help you make the case. Here’s Andreas Welsch, head of intelligent processes, SAP S/4HANA product management:
Andreas Welsch: So I think if we look at ERP historically, what we’ve seen that it has been and still is the backbone of many companies in their operations, right? Whether it’s planning, operations itself, or finance. And traditionally these areas are fairly labor-intensive, right? Finance, for example, if you think about a process called cash application in the area of accounts receivable, you may receive incoming payments and you need to match them to open invoices, it’s a pretty labor-intensive and tedious task that somebody needs to do on a daily basis. So through technologies, which are machine learning, AI, and robotics process automation, we’re able to increase automation in key business processes, end-to-end. And that’s where you see the key value of intelligent and emerging technologies being able to accomplish that.
Tamara: Perhaps one of the easiest ways to illustrate the value of next-generation ERP and automation is to see what happens when you don’t have it. Here’s Isaac with a cautionary Winter’s Tale:
Isaac: I walked into a CIO role a number of years ago. It was December. We had just finished our board meeting. We had projected a fairly good year. I walked in the next month for our first management meeting and heard the entire discussion around that essentially our forecasts were off, and instead of it being a plus year, it actually had been a down year.
And so, as the CFO was going through the story and try to understand the analytics around it, it was a mix of poor forecasting, poor data coming in at the wrong time, the effect of currencies. We were a global company, and there were currency issues that were miscalculated. And a lot of calculations done outside of the ERP and spreadsheets that just had bad formulas and bad mechanics around them. You know, that’s a telltale way of saying the way we used to do things doesn’t work anymore; we need more accurate data, we need to be looking at forecasting more often, and more frequently, and more accurately, the volumes of data that we’re looking at are bigger.
Tamara: Isaac clearly illustrates the problems we’re trying to solve here: The complexity of data flowing through the organization, the potential for human error, wasted effort… and above all, the sheer <yells> VOLUME…
Sorry, the sheer volume of data involved.
Here’s Andreas with an example of how intelligent automation can meet these challenges.
Andreas: So these technologies work best when they’re intertwined, when they are combined. And one example that I can give you is, for example, we’ve been working with a customer in life science for quite some time now. Specifically on this process called cash application. And for them, it was key to get more automation out of their existing process where they already had some rules that they have managed and maintained for some time but they were only getting so far.
So to get the last mile, so to speak, they were looking, inquiring a new way of doing that with machine learning and artificial intelligence. And in that example, we are training a machine learning model based on historic information of payments that have been received and how they were matched to open invoices. And we applied this to new payments that are coming in to give a recommendation of how these payments should be matched. And if this machine learning model has such a high confidence that this lump sum payment should be assigned to these five payments, well, then let’s go ahead and let’s have the system automatically clear and apply this payment. And then the cash application analysts only need to look at those payments that the system was not able to clear automatically. So that’s one point.
Extending from that, and extending to a robotics process automation, in that process customers receive so-called payment advices, where their customer tells them, “Hey, next week I’m going to pay you $100,000 for these five invoices.” Now, today, somebody needs to go to a portal, download this PDF, upload it into ERP, and then kick off the next step in that process. Using RPA, we’re able to log into this portal automatically, download the invoice, upload it to ERP, and then kick off the cash application process. So driving more end-to-end automation through a combination of different technologies.
So in the example of this customer, what it translated to was that they were able to accelerate their quarter-end close in one of their key markets by more than 50%. So talk about that as success and benefits of automation. Secondly, what they found was that for some payments, where it has taken the accountants over 100-120 days to make these matches correctly, it’s taken the machine learning-based system cash application only a single day and one run to do that. So where that has helped our customer specifically is by reducing the amount of working capital needed and by optimizing one of the key KPIs in finance.
Tamara: Andreas’ example shows just how much potential this technology has for businesses who fully implement it. 120-day processes reduced to a SINGLE day! 50% faster quarter-end closes!
But intelligent automation is about more than just time saved. It’s about what businesses can DO with that saved time. Automation on a next-generation ERP makes businesses more agile, more flexible, more capable to invest their time exploring innovative new offerings and lines of business.
Here’s Isaac again on the transformative possibilities:
Isaac: I wrote my book Driving Digital all about how organizations need to think about transformation and what’s the process around it, what’s the collaboration around it. And so, you know, I’ll go through a number of the things here. You know, starting with just transforming the business model, right? So, how you’re selling products today, what the pricing is around it, how they’re structured, what are you bundling, what are you selling separately, thinking about subscription models or even tiering models in terms of consumption.
These are all things… you know, I ran a lot of data businesses for those. I was the CIO in data businesses. And these were very difficult things to do, you know, to have skews for products that had multiple configurations and multiple ways of going to market and different ways of doing subscriptions around. It was really hard to do five, 10 years ago, and now they’re more commonplace. So, we’re starting with, you know, how do you actually transform the business model so that you can get recurring revenue streams and the consumption sort of matches up with your spend, and so that you, you know, customers have a better idea of what they’re buying. So, I think that’s, you know, one big thing that the endpoint looks like.
I like looking at customer experience, and more specifically, you know, what new markets companies want to get into. And so, people talk to me about transformation, and it always looks like you’re adding more, you’re doing more things than you’ve ever done before. You’re adding more technologies, you gotta have more capabilities. You also have to look at, you know, what new markets you want to go into and what markets and products you wanna exit out of.
And, again, I think the ERP is the heart of, you know, what types of businesses are you profitable in or showing growth that you have a market to go after to sell into. And transformation is really a practice of doing that on a very recurring basis. Being agile and experimental. So, how do you throw lightweight ideas out there and bring them to market quickly, seeing what kind of revenue you’re driving from it, what kind of customer feedback you’re getting from it? So, I think that’s a big part of it.
Tamara: All of this sounds great, right? But I can almost hear your objections already.
Voice: But Tamara, my organization is stuck with a legacy ERP and we just don’t have the support to make these types of major changes!
Tamara: Wow, yeah, did you hear that or was that just my suggestible imagination? Clearly I’ve been working alone and remotely for too long!
Voice: If only we had an expert who could explain how to get started with a digital transformation!
Tamara: Yeah, if only… someone like… Andreas, perhaps?
Andreas: So I think what’s really key here is, as with many of the emerging technology projects, start small, you know, do your pilot, your proof of concept, get a solid understanding of how the technology, how the product works, get some quick wins, and then from there on [you] expand into surrounding areas.
You’re not going to become an intelligent enterprise overnight. It takes some work. It takes some transformation. So starting with very well-defined steps, that you can also quantify, where you can quantify the benefits and then looking for things surrounding them, is key.
Tamara: Starting small and racking up quick wins makes a lot of sense. But there’s more to it than picking the right pilot program. When you move to a next-gen ERP, you have to really reexamine your relationship with the ERP.
ERP voice: Tamara, we need to talk. It’s not you, it’s me. I hope we can still be friends…
Tamara: Uh, yeah, NOT what I meant, ERP. Here’s Tim to explain.
Tim Crawford: One of the common questions that comes up today is: how do I go from a traditional ERP approach or thinking in that framework and move to a cloud-based approach with ERP systems? And that question can be a little overwhelming and a little daunting. One of the things that I found as a success factor for that move from traditional ERP to cloud-based ERP is with traditional ERP, the mentality was we changed the software to be customized to our business processes. Those that are finding success with cloud-based ERP solutions are actually flipping that on its head, and instead are saying, how do we use this as much as possible out of the box and customize it as little as possible.
Now that in itself may sound simplistic to those that have been initiated in the process of actually implementing ERP solutions. But it’s actually easier said than done because you’re talking about changing business processes to match that of a common software product. The benefit here is you then get to take advantage of newer technologies that are coming down the road when they become available, which is a huge improvement over the traditional approach, which could mean as much as a five- to seven-year lag between when new technology becomes available. And when it actually shows up in the enterprise available for users to take advantage of. It is huge.
Tamara: Essentially, it’s about letting the software’s capabilities evolve your business processes – not trying to make the software fit your existing workflows. With a next-generation ERP, you’re not buying a set configuration that matches where your business is now. You’re investing in a living, continually developed solution that can grow and change with you. Instead of customizing before you buy, you can do it on the fly with add-ons, plug-ins, and API connectors.
Here’s Andreas again to wrap up what we’ve been talking about… wait… shoot, that audio clip is on my laptop. Which I left in the 1400s. Let’s pop back and grab it real quick.
Here we are back in our pre-industrial… wait… was that factory there last time? I hope I didn’t mess up the timeline too much.
Cobbler: Ah, Milady, it is good to see you again! I hath finished your right shoe! And these 1,000 additional pairs of shoes!
Tamara: Sir Cobbler! This is amazing! How did you scale up your production?
Cobbler: Well, I didst use the metal book you left here and discovered the principles of “intelligent automation.” Verily, I hath a complete picture of my supply chain, I hath automated my manual tasks, empowered my knowledge workers, and maximized my efficiency!
Tamara: Well, geez – uh, yeah… that’s just great. Say, I didn’t catch your name… what do they call you?
Cobbler: Well, my family trade is tailoring, and my Christian name is Charles, so I am known as Chuck of the Tailors!
Tamara: Congratulations, Chuck of the Tailors. And hey, one more hint, since we’re meddling in the time-space continuum… leather shoes are great, but have you thought about canvas?
Cobbler: Hmmm… sturdy and breathable… canvas is a good material for shoes, yes! And we could make bright colors and emblazon them with a star.
Tamara: Absolutely. Good luck, Chuck of the Tailors. You can keep the metal book, just let me play this wrap-up from Andreas Welsch:
Andreas: So if we look at the full scope of opportunity, and we’ve talked about that a bit earlier on as well, it’s really that combination of different technologies into a concerted effort of driving end-to-end automation. So, for example, you can have your RPA bot that is checking your inbox for incoming payment advices, to stick with a finance example. When you do receive a payment advice, the bot automatically downloads it to your desktop, stores it there temporarily, opens up your screen for ERP, uploads the document there. That’s the RPA part. The second part that then kicks in is machine learning, for example, whereas with a form of intelligent OCR, if you will, OCR plus machine learning, we’re processing that PDF document, for example, extracting information from it, such as the payee, the amount, the date, currency, line item information, and storing this structured data in ERP.
Once it’s there, the next component using machine learning is triggered on a regular basis. In this example, a cash application where you match incoming payments to open invoices and augment this information with what we just extracted from our payment advices. Now, if there are any changes because somebody created a dispute and one of the invoices is no longer valid, we can alert an analyst to that fact, that there needs to be a dispute to be created. If in that process then, also we can kick off another bot that will send out an email automatically to this customer asking for additional information or a justification for the dispute. And lastly, monitor the inbox again for any incoming communication that we alert the analyst to for dispute resolution. So really, expediting many of these steps through a concerted effort of different intelligent technologies along an end-to-end process. That’s where I see the key value that we can bring in ERP.
Tamara: Intelligent automation isn’t a single technology or process – it’s a suite of solutions that all work together. First, there’s next-generation, intelligent ERP – that’s the foundational layer that makes everything possible. With the ERP as the backbone, you can add tools that use AI, ML, and RPA to automate processes, reduce human error, and increase efficiency. All of which leaves the humans free to continue innovating, using the ERP’s data-analyzing capabilities to drive that exploration.
And from there it’s experimentation, optimization, identifying new lines of revenue, new business models, and TOTAL GLOBAL DOMINATION! [echo]
Well, at the very least, you’ll have a digitally transformed organization that is ready to outpace the competition and shape the future of business.
Thanks for listening to Tech Unknown. And thanks to my guests Isaac Sacolick, Tim Crawford, Andreas Welsch, and Timo Elliott. Please subscribe on iTunes, Google Play, or wherever you listen to podcasts.
I’m Tamara McCleary and until next time: Stay sharp, stay curious, and keep exploring the unknown.