Hello, and welcome to another episode of the process and automation podcast with the automation guys. Since we kicked off with this several months ago, Arno and I talk about all things, automation, really intelligent automation, hyperautomation, process mining, low code, RPA, chatbots, and so many other topics in regards to automation.

And yeah, we basically covered all the different automation pillars companies can look at. To get the most out of automation. We have received a lot of that feedback as well. What we receive, we receive so much feedback, which is really great. And we would like, like to really thank you all, on sending, sending the feedback and suggestions in and by the way, your please keep this feedback coming and please rate us on iTunes as well. Five stars. It will be really great. If you like our content really want to make this the place to come back to for everyone looking into automation and get more done with automation, get more done with less in the business.

Yeah. And, uh, Microsoft power platform seems to become very popular with our listeners. And we have been asked to cover, um, recently more and more on that specific topic. So, so this was how our first episode about the power platform and was the request to cover more about the Microsoft platforms. Um, we have been asked to cover.

Also more on the AI and the AI capabilities and some really practical use cases. And, um, yeah, and that’s why we thought it will be really great to combine these two areas as part of the power platform. We now look at the Microsoft AI builder as well, so let’s get started. So I don’t know. Should we give, give our listeners a bit of a background on the power platform for.

Uh, yes, of course. Um, so obviously the, how platform is, uh, the Microsoft platform in this high productivity, low code business app space. Um, so what that means is, um, if you’re a citizen developer and you’re interested in building productivity apps, or either yourself or, um, you know, your department, for example, you could use the power platform.

Um, to build third powerful applications to solve problems and doing so without writing any code, obviously the power platform is, um, nothing new. Um, it has been evolving over the last few years into something that’s actually quite powerful at the minute. Um, we have had quite a lot of requests from our listeners really to, to unpack some of the features.
Um, off the power platform just to sort of demystify what’s what’s inside it. Um, how can you use it? The types of scenarios of business problems you can solve? Um, and I think today is a really good opportunity to kick that off and really to high-level touch on the, the sort of capabilities of the power platform, what what’s in it.

I just at a very high level and what we also want to do specifically today. Really focus on unpacking the Microsoft AI builder, which is a set of artificial intelligence capabilities that ships with the power platform, um, which is really powerful, um, where you have scenarios that you have to say, for example, to an unstructured data, into structured data, plug that into a workflow.
We’ll make that available on the form and mobile device. Do some predictions, some sentiment analysis. Uh, so before we get started on that, um, just a quick Roundup of what the power platform is, obviously it’s part of the office 365 platform. Um, it is licensed as a standalone, uh, by each of the constituent capabilities, which includes power BI power apps, power automate.
And of course our virtual agents, which is the Microsoft equivalent of intelligent chatbots. Um, There’s various licensing models for, for these various capabilities. Uh, so for example, you could license per flow or user, of course, if you have an enterprise license, that’s all, again, in a, in a, in a nutshell, uh, power BI gives us, um, a very, very capable platform to, um, create data-driven reporting inside.

Um, and, you know, visualizations of your data really powerful. It’s been around for quite a while. Um, very mature. And of course, you know, uh, as you would expect with Microsoft, it integrates with the variety of data sources. Um, you know, uh, the data services, so very, very, uh, easy to use. Uh, like I say, quite, quite powerful.

Um, you know, it’s, it’s a no no-code environment as well, so it’s very easy for citizen developers and business process owner. Or, uh, you know, people that’s enthusiastic in inside a company that wants to create reports or to put this together, of course, with Microsoft, the community and all of the support, the help enablement for using our BI is, is quite strong.

So it’s really easy to get started with it. Things like our apps, where you can build custom app. That’s solves business challenges using drag and drop no. And low-code solutions. Um, of course our apps, this is nothing new. Um, you know, it’s, it’s been in the, our platform stack for quite a while. Microsoft has added quite a substantial number of features to it, and also quite a few updates it’s easy to use.

So it is becoming more mainstream. Compared to, um, you know, your enterprise, no, in low-code platforms out there. Um, some people still think they some limitations in, in what it could do. Um, I think it coexist pretty easily with other enterprise low-code platforms. We see coexistence with our apps, um, with other, no known low-code platforms within site, quite a few organizations we’ve got automate, um, and that’s everything to do with.

Um, automating processes. So this is workflow automations using flow and also the, the RPA side of things where you could use, um, attended and unattended robots. That’s within cyber power Waterman. To automate some of those mundane, repetitive tasks out there. This integrates well with your power apps. So if there’s a handoff from a power app application to a flow, which is another RPA sequence or a workflow, everything as you could expect with Microsoft is fairly tight.
Ticketed graded, very easy to use, very drag and drop. And there are some. You know, capabilities. If you compare the, let’s say the RPA side with something like UI path and blue prism automation, there is some limitations. Of course the platform itself is not, not that mature, but I think the target audience for this, at this present point in time, as, as really sort of that citizen developer that wants to create productivity apps, some have automations, you know, email comes in with a.
Um, perhaps the invoice gets downloaded from machete inbox. Uh, invoice gets, um, OCR for example. Um, and then the data gets pushed in a database somewhere or surface somewhere on a power power app, um, user interface. So it’s these, these types of automations that you can speak to that, of course, the virtual agent, um, you know, that’s a place that you can use.
Build your chatbots to, to engage in conversations with your customers and your employees. And it’s got pretty strong, natural language processing capabilities built inside it at the detect intense. And of course it could also, um, trigger, uh, power automate, um, processes, including RPA and workflow processes.

Um, so, so fairly, fairly, um, solid offering from Microsoft, um, as you can, um, it’s becoming a lot more popular with, with, with, um, businesses and enterprises out there. Um, you know, for, like I said earlier, sort of citizen developer type automations, of course, nothing stops you from, from, um, you know, doing large enterprise grade, um, deployments as well.

Um, but of course they are setting limitations still in, in certain, certain, certain capabilities, um, of, of, you know, compared to, so let’s say enterprise low-code vendors. So that’s sort of a, I guess, a very high level. Um, you know, what is, uh, what is inside the, the power plant?
And it seems like lots of companies, especially the bigger corporations they can make, make good use of it already because it’s lots of that stuff is part of their big corporate license agreement anyway. And they, they have sort of, lots of, lots of credits cloud credits available when it comes to the. To the sort of consumption of this, uh, this services, the AI and everything was just volume based and, uh, Azure based.

So, so you had to have plenty of, um, capabilities. Um, and yeah, I think that’s, that’s the Y also becomes quite popular. There’s good. There’s good. Um, there’s a good framework. There is a good set of tools there and. It is easy to get started with. So you don’t have to go to another sort of procurement layer layer to get something really substantial substantially, um, um, capable, built.
Isn’t that? So that’s why it’s popular. Yeah. And I think one of the reasons why we want to touch on. Uh, Microsoft AI builder today is that if you look at the power platform, including the power apps, power automate, and then obviously the virtual agents, I think the Microsoft AI builders is really a differentiator here where it is.

It’s adding a lot of value to the platform, um, because Microsoft AI builder is artificial intelligence platform and it’s. Um, stack and it’s, it’s really, um, it takes a low code approach, um, to develop. And introducing artificial intelligence into your projects, um, without the need for thoughtful coding experts, of course, there are some limitations in what you can do, but at a whole, I think what it’s trying to target is sort of, it’s a really large proportion of problems, common problems that exist with inside the business.

Um, since we’re really, I think what we want to do today is sort of diving deeper into what that. Microsoft AI builder is. And I think, you know, the best thing to start really is what can AI builder do. And again, I think, you know, I build a, provide you with the ability to optimize your business processes further using artificial intelligence.

I think that’s in a nutshell, you’re making your, um, uh, processes and your process automation more intently. Um, you know, so you could use AI to automate processes and gain really good insights from your, from your data. Um, and then combine that with the power of power automate to obviously create automations and then the power apps to create some of the user interfaces.
Um, now of course, the, um, Microsoft AI builder, all of these are built on the Azure AI capabilities, which is a standalone services as well. Um, and the basic idea is that you could train models and build models to enhance the intelligence of your business apps. Um, you know, uh, using data you’ve got, for example, in dynamics 365, um, and your Microsoft data version, which is of course, a repository and data repository, a singular repository for all of your data objects.
Um, so, so, so, so really with, with our builder, you can build custom models, um, tailored to your needs, and you can also choose pre-built models, which is ready to use for common business scenarios. Um, so a couple of examples we can, we can touch on. Yes. Yes. I think you’re very popular is, is, is always the, um, making sense out of, um, documents, um, in letters, invoices, which are not necessarily, uh, available in electronic format, like a PDF.

So the text recognition capability of, uh, The Microsoft AI builder is potentially a very popular one. Um, and this was, this was a model, a model of text recognition. Um, you can automatically process texts from, from images, basically scanned images and, um, there’s so much document, so many documents still out there.

And. Effectively you use, you use a optical character recognition to, um, to find. This text blocks and read the texts within the, as well as the photographs and, uh, yeah. And all sorts of scan documents. So that the key here is as well to, to convert that unstructured data, which sits in lots of databases, um, and convert an infrastructure data.

So very, very important when it comes to working with RPA. Um, you could say. Uh, text recognition or your OCR, um, is sort of, uh, sort of the eyes of a robot. So without, without that structured data robot can’t really do do much, or the workflow can’t be as effective as, uh, as if you have lots of structure data available.

So this is why it became, um, again, a really, really important. Discipline when it comes to the skill, um, how it’s called these days as well. And when it comes to, um, to automation to make sense out of documents. And extract information. Like it could be a very simple as an invoice. Um, identify the, the VAT number, the invoice amount, the, um, the supplier name, all that kind of stuff.
If it’s not electronically available, um, uh, this is what you would look at and, um, or documents, contracts, medical, uh, documents. It could be all sorts of documents. And by identifying specific words, um, which are just relevant for your business. Um, yeah, you can, you can make it highly effective, uh, over time.

So when you train a couple of documents, um, was very specific terms. Um, you can really, um, yeah. Gain lots of efficiency here. Yeah. And it’s all back to, I guess, um, the old school OCRM, but this just ships out of the box. So if you do have an application form that needs to be. B, um, OCR. Um, this allows you to upload, let’s say five examples or 10 examples of typical forms that get submitted.

Let’s say it’s a mortgage application or any type of form where a data is in a predicted place. Um, the AI model will then consume. The actual documents itself, or it could be images, it could be PDFs. Um, it doesn’t really matter. Scan, scan images, uh, and it allows you then to pinpoint certain areas in the document that’s of interest.

Um, so it could be input fields that you want to map to specific entities and fields and the model itself. There’s a training, um, routine that you train it, you provide it. The mapping effectively to extract the data. And, uh, once the model is trained, um, you can publish it and it’s ready to be used in, uh, power apps or a power automate workflow, for example.

So it’s a very easy to use. And of course, if it changes your existing model is still available, um, you can then retrain it based on, uh, changes to the form. Um, and it would recognize those changes. If there’s additional mappings for it, for example, and you know, it makes all of this available without any coding, which is just really powerful.

I think that’s the appeal with all of these things, um, is a minimum of five, five documents. That’s a requirement, but of course, if you add more models, um, uh, accurate, um, but that’s, that’s basically text recognition. Uh, the next thing it’s useful with is object recognition. And, you know, this allows you to train our AI model to recognize specific objects, what types of objects.
So for example, a, it could be used in inventory management to, to detect, uh, equipment. That is, that is otherwise very difficult to identify by sight. So you could take pictures of products, specific products. You can train the model to, um, predict that these pictures actually belongs to a specific. Um, again, very useful for inventory management type scenarios or in manufacturing, for example, a website ordering.

So you can take a picture of a specific product uploaded to a power app, uh, application, and it would use the object recognition to understand what the product is. So I could recognize that as something you wish to. Um, so again, very, very powerful feature. No, no code required to do so. All you would need to do is just train the model to index these pictures or these products, for example, and it could be anything really, it could be, uh, damages to a car.
For example, if you have insurance claim, uh, you could, um, uh, create a model that predict a damages, for example, So, so there’s, there’s a lot of use cases for this type of object recognition within sight, the AI builder for real world.

Yeah, prediction is a, is another really good one. So if you have, um, lots of data available, um, um, you can, you can use the prediction capability or the prediction skill, uh, was a Microsoft AI, build it to. To, um, to identify patterns, um, which otherwise will be very difficult to identify manually. So you can’t go through 1 billion rows of excess spreadsheets.

Yeah, you can, but it will take you forever. Um, but it was this, um, was this skill. You can identify these patterns and use that knowledge to predict the future. Um, and it can answer questions about your data very specifically, like, uh, making comeback. Yeah, so no two-folds, um, pass or fail kind of, uh, output.

So it’s, it’s very, very useful for, um, so one really good example. We use it for predictive analytics, um, uh, to maybe figure out on sorts of IOT data captured in, in trains. For example, it will, uh, we’ll monitor it. All the, the sense of data and push it all into this huge data set, um, catch captures, um, every minute, um, what is sort of, um, in these IOT devices, it could be sensors in a train, all that kind of stuff.

So it all flows in there based on, based on your prediction model. You can then, um, look at, for example, uh, under what circumstances maybe a train might go into maintenance, uh, because of some sensor data says. Uh, it’s basically out of the social, that kind of stuff. Um, so if you, if you have something like this running, you can prevent potentially a maintenance outage, um, by, by just knowing the future.

Um, and that can then trigger again, a workflow, maybe maintenance workflow where we’re, um, some, some engineers looking at a specific problem. So combining all these things together will then, then really, uh, produce a powerful. Yeah, I think for prediction, another common scenario is in manufacturing where you have to predict based on your sales orders, what orders you need to play by play place for parts, for example.

And a lot of companies do have a lot of historic data, but that’s still a quite a tedious process. And it’s probably for a lot of manufacturing scenarios. We’ve seen that that’s still, yeah. Very very driven by Excel for a lot of businesses because it’s so complicated. And, uh, you know, people want to have flexibility to, to, you know, put in some formulas to, to make some predictions when products needs to be, uh, or components needs to be ordered.

Um, with this, you could pull in that historic data historic decision-making and it could make predictions based on your sales, what type of components you need to. Uh, to fulfill your sales orders, for example. Um, so again, a very, very powerful, no code, uh, solution for prediction. Um, I think, uh, the simplicity here is, is, is, is really powerful.

It’s easy to get started. And like you alluded to Sascha, you can obviously plug this into other workflows, um, for art, for instance, if you need to place an order for, for some parts, all of that can then be, uh, streamline. Um, another powerful capability. Um, and this is very similar to the text. Recognition is form processing.

Now the difference with form processing is that it’s more advanced in the fact that it can predict or extract tablet, data repeating, repeating sections. Um, so for instance, uh, if you have an application form, perhaps there’s just some, uh, data in there which is, it’s not very complicated with the form processing.

Um, it actually detects what detect, uh, repeating sections, like invoice line items. Um, very powerful way to train them on. To, um, you know, detect, you know, very complicated data with insight, PDF documents, for example. Um, and again, turning that unstructured data into structured data that could then be, um, you know, used downstream to be keyed into a financial system, for example, using a robot or, um, you know, plugged into approval process where somebody can actually.

Um, look at the data, um, ensure that it is actually correct. For example, if it’s an order or invoice, that’s above a certain threshold, for example, you can go through approval workflow. So there’s, there’s endless opportunities. Um, and of course what we’ve removed. Um, the sort of form processing AI capabilities, the fact that we’ve already extracted data, there’s no re-keying of information.

It’s a very, very powerful feature. The really, really good application for that within insightful. Yeah, nowadays more and more of these documents come through electronically. Isn’t empty PDF documents. Uh, um, yeah, so, so this is a very wide use case. Um, you and the next, a really big, um, big element of Microsoft AI builders classification.

So thinking about, um, uh, texts coming in an email, um, was all sorts of content. Um, it might be a customer customer request, um, but was tech was classification. Um, you can actually figure that out. Is it actually a customer request? Is it their complained? Um, so yeah.

These patterns was in your language and text and classify as content. Um, can also tack something which is potentially a, um, an angry customer. So what’s the sentiment analysis that figures out, uh, someone is not very happy. And then you could take that insight, that knowledge, and maybe rule to specific EMA going through, or customer requests coming through to a, to a specific agent which deals with, uh, maybe angry customers.

Uh, or you can classify a specific request coming in as spam, um, it all sort of stuff, or you can identify, this is their name change request. When it comes for example, to insurances, um, um, people got married and then they need to have a name change. Um, and suddenly that needs to. It needs to process some name changes in the backend system and combining classification.
And then again was a, was the power automate was the RPA capabilities. You might be able to go into backend systems and do those things start directly or with specific classification, um, you will trigger some other, um, uh, workflows may be triggered, uh, the virtual chatbots, the assistant. Um, so yeah, once, once we know what it is, We can do all sorts of stuff with it and trigger all the relevant processes to achieve more end to end processing.

So you have very, very important capability and, um, and it also learns over time. Isn’t it. So more and more texts come through. At some point you, you have the system classifying all sorts of stuff in your, in your company. And, um, yeah, I think maybe as a roundup, we could just cover a few business scenarios.

Um, of course we talked about the Greek classification processing prediction, object recognition, and also pixel recognition. Um, but I think it’s, it would be good if our listeners, um, if we. Just provide a bit of a summary on the business scenarios and the type of model you would typically use with inside.

Um, AI builder, sort of first business scenario I want to touch on is, um, you have a automation to automate, I don’t know, a customer application processing and, uh, The model you’ll use for that is your form processing engine, because that is something that you want to extract information from a form. You want to turn that the infrastructure data, and you want to potentially plug that into a backend process.

For example, um, you want to automate an expense report and of course, with expense reports, there’s some receipts. That’s typically associated with that. So you’ll use the receipt processing model within AI builder, because that has got a very, very good capability to detect any type of receipt and, um, extract the structured data from that.

And it’s really powerful. Um, I’ve seen examples where receipts are scanned, they were crumpled and it’s, it’s really detecting, um, you know, the values within sight. Um, another one and I’ll give you a chance as well, to run through a few of your own sessions where you pass one to categorize user feedback and what the type of focus is of the feedback.

Are they angry? Are they happy? Are they ecstatic? And of course we could use the category classification because we could, we could look at the, the text itself and the AI builder would be able to categorize the, um, the sentiment within. Um, the, the user, uh, feedback, for example, um, and also if you want to extract insights from your product reviews, when people typing free texts on your products and they again, just in natural English, they want to just leave a feedback.

If you want to extract, um, that into, uh, structured entities, you could use the entity extraction, for example. Yeah, you mentioned some really, really good use cases. Then we touched on briefly on the, uh, um, on, on some where we went through the, the different skills of AI builder. Um, but a very, still a very common use case is in depth, uh, in indeed sort of accounts payable where an email comes in with an invoice attached where you, we have that example for so many and so many places, isn’t it.

Um, so, uh, But it is still a very common use case. Um, and sometimes you get, um, you get the PDF, uh, like Arno mentioned, you can then use that extraction there, but sometimes you still, um, you still get stuff in, um, from, from, from, from external, um, where there’s still a normal sort of. Is, it really is a scanned invoice.

Um, some, um, government places that they’re, they’re not sending him emails with invoices, they they’re sending, they’re sending really scans and then maybe an outsourced partner. They, they, they bring that stuff into sort of an email was a, was a scan. And for that, we in very, very many cases, we, um, we use OCR.

OCR is really sexy again. Um, To, to really extract that information, um, to allow that further automation, otherwise someone has to look through it and code and manually into the finance system, and this is not the right way. So, so this is, this is really helping, uh, accounts payable departments. Uh, it could be as well, a purchase, a purchase order, which is coming, um, coming through, um, in the old fashioned way.

Um, Yeah, not every company is, is modern, unfortunately. So that’s, that’s a really, really big use case I’m still to tackle and, um, and was making it that easy. Um, from the Microsoft side, uh, it is, it is really worth a look, um, uh, otherwise, um, you know, you, you have a really old OCR software. It doesn’t give you that, that kind of learning isn’t it.

So it’s, it’s usually very rigid and very strict. You, you need very strict documents and if your new document comes through, you need to re restructure it. So in this, um, this capability here, uh, gives you more and more, more flexibility. It learns and, um, is dynamic. Um, and this is something you really need.

Um, yeah, and I’m, I’m really, I’m really a big fan of this, uh, building, building this prediction model. Um, I think he companies was the, was the data. They should get it in, in, into this, uh, system and really let the, that the algorithms work on it, um, for, for specific use cases. Um, and it really. Bring so much, um, insights.

Um, and, uh, it was all this IOT data. I think we don’t do much enough. Um, so this was the example I had earlier. Um, there’s so much stuff, um, in, in building management where so much IOT data is in there. So you could, you could prevent so many issues, maintenance issues, and I really want to reiterate on this one, um, by.

By really analyzing that data. Uh, it could be as financial, financial departments as well. They look at data and they can actually predict, um, uh, certain, certain outcomes. And you can, as a business, maybe get ready for, for certain, um, events, um, before, before they actually happen. Um, And, um, some, some, some tools bring that already in, in, within that tool.

But some tools are still very old fashioned. We working with very, very old ERP systems as well to pluck that on, on, uh, sort of on it, uh, will immediately. Really put your old ERP system or your other old finance system sort of, yeah. It’s sort of a finance system on steroids, um, by putting, putting those kinds of tools and models on front of, in front of it.

Yeah, exactly. And I think, you know, the prediction element of this is, is quite relevant because we are generating a lot of data. In an exponential fashion these days. And you know, if, if you, for instance, look at a example where you want to detect fraudulent transactions, it is a AI capability that you need.

It will be very hard to spot fraud with insight, massive data sets. Uh, so again, with the prediction model, Um, with, with AI, you know, that’s with inside the reach of organizations. Um, and of course with AI builder, it doesn’t just stop with detecting things. It’s also again on the, uh, the text side of things.

Um, it can translate texting to, uh, Lang uh, you know, any other language. I think it’s about 90 languages that support. Um, you know, it’s got things like, um, uh, extracting, uh, that sort of texts from a photo and saving that into a database, uh, object detection would obviously help you to automate inventory tracking.

Like we mentioned earlier, um, you know, there’s even things like, uh, business card readers, just to give you an example of the sort of ad hoc nature that read texts, but there’s model. Uh, although I personally think business cards is a bit antiquated, but again, it gives you an example of the fact that, you know, it, it is very intelligent in terms of a variety of types of information, and it can detect and then save in a database for example, and in key phrase extraction, very powerful one, um, where you might monitor it, you might monitor social media feeds and you want to detect whenever your brand.

Reference. So key phrase extraction would help you do that and then created the required automations to deal with that. Of course, that’s very hard to monitor that manually.
Lots of use cases in this one. Um, I think, um, it will be useful for obviously all our listeners out there. If there’s any additional ones that you might have on your mind and run it past us more than happy to. To, uh, discuss that with yourselves. Um, the power platform is, is in itself quite a broad topic.

Um, I think it’s high to high time. We actually dive a bit deeper into this because there’s a lot of noise up there and a lot more people that is asking, um, you know, what is it? Should we use it? If we’ve already got office 365 now state, uh, you know, how does it coexist with other low-code platforms? For example, So definitely a topic we’ll cover in our future podcasts as well.

The pursuit of covering anything automation. Absolutely. Yeah. Um, yeah. Thank you very much for, for everyone listening again. And, Arno I mentioned, you know, we, we cover everything, all things automation, and yeah, if you have further suggestions, please visit us on our website, theautomationguys.net, there you will find more, more resources on all things, automation, and you can leave your suggestions there.

We will be very happy to, to pick that up in the, in the upcoming episodes and the off. Um, make sure you’re subscribed to our podcast and to our newsletter on theautomationguys.net and that, that will give you, um, uh, sort of all the, all the information when we have events coming up from special webinars, um, special demo sessions, hand-eye hands-on labs, that kind of stuff. What we’re planning for the podcast community now, our loyal listeners.  Thank you very much. And let’s automate.

Unfortunately, that’s it again with this episode of the process and automation podcast. If you liked this episode, please give us a five-star rating and don’t forget to subscribe to this podcast. So you don’t miss any upcoming episode. We hope you will tune in next time and until then let’s automate it.