CloudTrade Podcast - Episode 8 - Business process automation
Watch the podcast here, if you’d prefer to read about what was discussed, please read on.
Steve Britton, Director of Sales, talks to CloudTrade’s Vice President of North America, Roger Hatfield, on the topic of business process automation. Roger emphasizes the importance of embracing automation in order to offer leading-edge efficiencies to customers. He also outlines CloudTrade's exciting new BPA development in the pipeline, Project Grandalf.
Steve: Good day, Roger, Lovely to engage with you today. This is number 8 I believe of our CloudTrade TV podcasts, where we are explaining to the market about what we do and why. We have a solution that is very different in the market and I've chosen a subject for us to discuss today: business process automation. Anybody listening will think that's a very broad subject. It covers a whole range of different areas of the business, and I suppose we should put it into context around what CloudTrade do, and typically our focus is around processing of finance documents such as invoices and inbound orders.
So really I would like you to put into context for the listeners your understanding of business process automation and how CloudTrade’s service can help businesses in their journey to delivering full BPA.
Roger: Always great to catch up with you. You and I have a lot of experience in this area and it's one of my favourite subjects because I've spent most of my life helping organizations solve some of their problems with business process automation. And I think we look historically at the way the markets have moved in the last 20-25 years, with the advent of faster, smarter technology, both you and I together have helped many organizations achieve what they see as business process automation.
“… if we as a company can help companies automate those repeatable manual-intensive tasks that revolve around capturing data we're doing the right job.”
BPA means many things to many different people. It can range from as you say, finance documents to order documents, to order acknowledgements, and to processing website orders. For CloudTrade it’s really about helping people achieve straight-through capture of data that needs to be processed in an end system. What I mean by that, Steve, is if we as a company can help companies automate those repeatable manual-intensive tasks that revolve around capturing data we're doing the right job.
“Why do we need business process automation? Isn't this part of the normal world today?”
Steve: OK, that's interesting. So when we think about the modern world that we live in today with smart devices on mobile phones, you and I probably started with Blackberries and so on, but the world has moved on. So I suppose it begs the question, “If I'm trading with somebody, invoices are my life blood and if I don't send my invoice for goods or services, I don't get paid, and I haven’t got cash flow, so why aren't these automatically transacted? Why do we need business process automation? Isn't this part of the normal world today?”
Roger: Great point and I would like to say that, yes it is, or most people think it is, but unfortunately a lot of people use old technology to try and capture that data and one of our greatest one-liners within CloudTrade is we provide data that people can trust.
Roger: People are getting bad data. You put bad data into a process and you get bad data out at the end of the process. And when we look at the areas of business process automation for organizations, they're always looking to gain a little additional benefit from what they're doing. They're doing it either because as a partner of ours they might want to try and enable a bit of technology to provide straight-through capture and processing for their own product, or if it's a direct customer they want to remove all those inefficiencies out of the process.
Both you and I know that the cost to process an invoice, if it's automated, can be as little as two to three bucks. Excuse me talking in the American vernacular there, but we know that other organizations are spending 20, 30, 50 bucks per invoice to process it, and that's because of the inconsistencies, not only in the capture process but also in the automation of that process. And that's where we really add value. Most organizations will make that great commitment to put in a procure-to-pay or a source-to-pay solution or an invoice automation solution. What we find is they put all that effort and invest all that money in doing that, but what lets them down is the capture of the data that needs to go into the process.
Steve: That's really interesting when you talk about the true cost to process a document. You and I have had many debates with people who five, eight, ten years ago, for whom the norm was to say, well, actually, let me take what I'm doing somewhere else for less money. With the CFOs of the world today that doesn't cut it anymore. With all of the challenges that we've had with things like Sarbanes Oxley and Enron and all of the other scandals, you can outsource the process but you can't outsource responsibility.
“… it’s about understanding that the data that's being transacted, being received and processed can be trusted.”
It’s about having the ability to embrace technology, and to address your point, which is really valid, it’s about understanding that the data that's being transacted, being received and processed can be trusted. So when my auditors come in I'm now going to get a clean bill of health. When the tax authorities do a review, I've got a clean bill of health. As a CFO, I can sleep well at night. You can throw robotic process automation at processing invoices, but unfortunately if the data you give to a robot is incorrect you've got a dumb robot, and that causes problems. You just mentioned to me looking at 20, 30, 40, 50 bucks/pounds/euros - whatever the currency is that we're operating in - per document, and I often hear people to say to me that they can have it scanned and keyed in India, Malaysia or China for a few cents, but that's not the end of the process.
Roger, you've talked about accuracy and the ability to trust data. I started off this debate around the unique approach that CloudTrade has. In the context of business process automation, just talk to us about how we are able to deliver trusted data as opposed to data produced with an OCR capability or by other technologies that are transacting invoices.
“… we are able to deliver trusted data as opposed to data produced with an OCR capability…”
Roger: I've spent some time doing this, and I'll talk about one of our customers who has an OCR-type channel. They are in the BPO space and their OCR channel did capture data, but once they’d captured the data, it had to go into a manual process that was anywhere between 15 and 45 minutes long for a human operative to go in and check the data, change the data, manipulate the data, put it through some revalidation and then submit it again. And that process, depending on the size of the document, could take anywhere between 10 and 45 minutes. By using BPA, with the smart rules that we write within our technology, we are able to take what that human operative does, and then we write rules so it mimics that human operator. We took that time down to less than 6 minutes by just writing and effecting the set of rules.
Now we talk about business process automation. That's what we do. We've been doing it for 10 years. RPA is the new kid on the block. We know with RPA over the last four years that we've seen a steady rise and a steady interest in what it's able to do. Many of the RPA vendors have also created their own capture bots within their technology, which is great. But it can't do the BPA part which is manipulate, develop and write rules, specifically like a human being does, because one rule doesn't fit all for an RPA bot. With CloudTrade we can actually mimic some of those.
“The extraction is the trusted-data part, because we do it with 100% accuracy.”
Steve: Fantastic. Putting that into context, if we receive that human-readable document, we extract the relevant fields, header and line, we apply the rules that you've described, because every receiving application is going to have its own rule set. The extraction is the trusted-data part, because we do it with 100% accuracy. We apply the rules, and therefore we export with no human intervention what we used to aim to deliver in the old days, which was what we call straight-through processing. And that really is true BPA. That's where RPA is great, because with RPA the robot can say it has got the accurate data, and it needs to go off and validate something, or it needs to go off and check and enrich some data for the downstream process. That’s what robots are really good at doing. So CloudTrade in that regard absolutely supports and enhances the RPA journey.
“What business process automation is allowing people to do today is to take a step back, to try and see where efficiencies can be made in their businesses…”
Roger: Yes, certainly. And what I'd like to add there is that most organizations are struggling really hard to keep up with businesses in normal mode, and so they make do with existing processes because they've always done it that way, or they've always wanted to do it that way. What business process automation is allowing people to do today is to take a step back, to try and see where efficiencies can be made in their businesses, and what those efficiencies will deliver to the bottom line.
I would think if you spoke to any CFO or any finance director or finance manager and said, “Do you know how much it's costing you to process your invoices or your orders?”, very few of them would have any idea. But when you present a figure such as “Is it 50 bucks? Is it 40 bucks? Is it 20 bucks?”, and you times that by the volume of documents they've got coming in, that brings into sharp focus that they're leaving money on the table, that they can suddenly make increases in efficiency that's going to save them money, they're going to be able to pay their suppliers faster, and they're probably going to be able to get some payment discounts as a result of that. To my mind, if you're undertaking a change-management program, or you're trying to do an automation agenda within an organization, this is one key area you should go straight into.
The difficulty is, most people don't want to scratch that itch, because their process is working. The biggest challenge for most people is changing what they do currently because it's working. It's either been outsourced or they're doing it well enough in-house. But we know, once we engage and we start mining and discussing it with organizations, we know that there's a huge ROI for our customers and our partners to get as a result.
“Everybody's moved to the digital environment, which is absolutely where CloudTrade sits.”
Steve: Fantastic. We have covered the true automation piece, the reduction in touchpoints, the significant reduction in the cost to process a document, and also the huge benefit, because CloudTrade is a fully managed software-as-a-service model and implementation can be rapid. One of the messages that we give our prospective customers is, “We've worked out what it's costing you today and the delta if you move over to our service, and this is the saving, and every month you delay in making a decision to move is costing you that amount of money.” So there's a compelling reason to move into this environment. There’s a compelling reason because it's embracing leading-edge technology. As we've seen with COVID-19, the pandemic around the world, mail rooms have been closed because there's nobody there to open the envelope and process the document. Everybody's moved to the digital environment, which is absolutely where CloudTrade sits.
“… we're a SaaS model and implementation can be very quick…”
Roger we've talked about where CloudTrade can provide automation, reducing the costs, improving the time frame, because with us we're a SaaS model and implementation can be very quick. This is where we are today in terms of our capability. But the market moves on rapidly. We talked about Blackberries and how technology is growing at an ever-faster pace, and how CloudTrade as a vendor in this market needs to keep abreast of that. We need to keep ahead of our competition. So, aren't all businesses automatically connected electronically today? Well, the reality is they are not. So what are CloudTrade’s plans to help that move forward? We've got 20,000 sending parties and probably 1000 receiving parties on our network today. Can you explain to the listeners what our plans are to stay ahead of technology. What are we doing in terms of additional artificial intelligence, if that's the right term to use these days, to drive our product forward?
Roger: Just to reiterate, COVID-19 has given us the opportunity with many organizations to engage with them, but it's also made them look internally at themselves to look at the BPA agenda, and that's made us look at ourselves too as an organization. We've had to. And the agility of products out there, and the speed to market. of those products, means that we must move with the field. And what we've done as an organization, traditionally, is offer a SaaS model, we do all of the work, services are provided by us. But the beauty is, we've got a very exciting development called Project Grandalf.
Steve: OK, The wizard!
Roger: Yes, it’s the wizard with a big fat top hat! Project Grandalf, son of Gramatica, or Grandfather of Gramatica, I'm not sure which at the moment. But it's really truly exciting.
Steve: Sorry, just for the listeners, Gramatica is our core product. Our CTO Richard Develyn created the application, the service we deliver, and he named it Gramatica, so that's our product’s internal name.
“The plan is for a remote service that the end user can use …”
Roger: We've been working on a project with the brains trust of CloudTrade development team and it's very exciting. The plan is for a remote service that the end user can use that will take in an invoice, it will be mapped to a vast historical library that CloudTrade keeps of invoice sets and invoice documents and other documents that we keep, and it will find the best match on that document for what you're trying to capture. So you might be looking for, let's just say, 14 standard key areas and it would capture invoice number, invoice address, supplier name, maybe some of the products, and what it does is the system itself will go away, it will capture the data, and then it will present back to the user a set of questions through a workflow that we've got in our product saying, “I think this is invoice date, can you confirm it?” and you say, yes or no. “I think this is invoice number, can you confirm it?” Yes or no. If it automatically captures the data well and it can validate it, you won't see that question.
“What this allows us to do is put the onus on the enablement of the technology back on the customer or the partner that's actually using it.”
What this allows us to do is put the onus on the enablement of the technology back on the customer or the partner that's actually using it. So the example there is, we talked about someone in accounts payable. He has an invoice that has come in. It's got to be paid now and he needs to put it into the system. But it's a three-page invoice that he has to put in. So someone's got to type that in. That's going to take someone 15-20 minutes. With our technology, they will be able to send it into our platform, it will return back to them within less than 15 minutes. They just need to answer a few questions. It will go back into the system. The system will remember it, and process it, and send it back to them as validated data. Very, very quick. Once they have done it, they do not have to do it again.
Steve: Right, so that model leverages our 100% accurate data extraction. We may need, say, 14 or 44 different fields off a document, and we extract what we can in first pass, present it to a user. They confirm or they add in the missing information. It learns behind the scenes. It creates the rule – i.e. the mapping that we would historically do manually. It’s now doing that automatically and it exports the required data. The next time an invoice like that comes in, it utilizes the rule that the user has set up, automatically. It’s amazing.
Roger: Yes, and if the supplier changes the invoice, if it comes in and it’s changed, we will pick it up and will send another email back saying, we think something's changed on this invoice, can you just confirm that? They say, yes, the rule remembers that and it goes into the AI logic, and it's there, it's done.
Steve: Fantastic - AI on steroids! Sounds really exciting. I know I've been involved behind the scenes on some of the work that we’ve been doing on that and we are really looking forward to it. It is going to be a game changer and as I mentioned before, it's going to keep us ahead of the curve for our competition. We are ahead of the curve with the header and line capture and the fact that we guarantee 100% accuracy. Nobody else in the market today does that so this is going to take it to the next level.
Really exciting, Roger, and I think we have probably come to the end of our time slot for today. I think the underlying message really is don't hesitate listeners, if you've got a need to process inbound human-readable documents and you want data you can trust, come to CloudTrade and we will give you a rapid ROI and the service that's absolutely best in class and leading edge.
Roger, it’s been a delight talking to you today. Thanks for your time and will speak again soon.
Roger: Good to see you, Steve. Thanks.
Steve: Take care, bye.