Microsoft Flow: Digitally transform your business - BRK2228

All, right, good. Morning, everybody thank you so very, much for waking up and joining us today we, appreciate, you having a, somewhat, early night last night and if not then getting, up in spite of that anyways, to join us my, name is John Levesque I'm joined by my coworker here Kent Weir and this is digitally. Transformed, your business using, Microsoft, Flo so, welcome and thanks for joining us, so. Before, we start I'm gonna ask for a little bit of interaction. And participation this. Is a little survey, a few questions gonna, ask what you guys think, about digital, transformation, since. That term is so overloaded, these days we'd, love to get an idea of what, you guys actually, think it means also, as, a bonus, I'm gonna go ahead and tell you if you participate, you you're, going to want to participate because I have a flow built around this survey, you'll be receiving an email with a bunch of bonus content that, you're not gonna want to miss and if you don't do the survey no one here is gonna forward it to you. Okay. So. Go ahead take a second, guys fill, that out and and. We're gonna let you look, at that and give those answers, real quick. While. You, are doing, that. Kent. Is going to talk to us a little bit about what is digital. Transformation, so. Yeah says John mentioned, digital transformation, is a heavily, overloaded. Term and, so here's an example of, what. Digital transformation, is or perhaps, not as. You can see there's a lot of technical jargon in. This, specific, quote and don't worry there's a slide where Bill redeems himself but. The. Important part here is that digital, transformation, isn't about technology. It's, actually about transforming, businesses, and providing, businesses, with some sort of an outcome, that may, use digital, tools and so, this is what we want to talk about today we want to walk you through how you can set up a digital transformation program, within your organization. Using something called bottom-up. Innovation and, then, on top of that we're going to give. You some examples, of how you can actually build some intelligent, automation processes. That do help your organization transform. And if. You follow, us on Twitter last night we we talked a little bit about. A tool, called power query and power query integration, so, we're also going to demo that at the, end of this session as well as part of one of our our use cases and, you'll. Want to stick around for that so. Digital transformation, the coupling of granular real-time data smartphones connected devices smart appliances, wearables, and so, on and so forth, this, is helpful yes, this is not digital transformation. As. Dementia bill does. Redeem himself and, he, gets a little bit better a digital transformation is an application, of digital capabilities. To, processes. Products. And assets to improve efficiencies, enhanced customer value manage, risk and uncover, new monetization opportunities. So, that's a little bit closer but there's, one thing that's missing. What. Could it be John you. Know I think it's, the people it. Is you're right, without. People there, won't be transformation. Inside an organization. You. Know people are gonna have the biggest impact, on success, in digital transformation, initiatives, and you, have to overcome some, challenges to, get them there right like, suspending, disbelief, overcoming. Someone, else's problem, right we all have that mentality of that's not my issue as someone, else is they're gonna deal with it also, the, ability to absorb and embrace change, right you have to have leadership, bought, in you, have to have initiatives. Going you have to have support so. That you can absorb, and embrace these things because they take a little bit of time often to to go right but it's it's interesting because the. The market has changed right.

Whereas, In years. Past we would think of these, investments. In in these initiatives, as as long-term and, and. And now everyone. Thinks man I need some return in months, right not years, and so, you know that's speaking, to ability to absorb and embrace again and now opportunities. Kent what are some of the opportunities, around so. I think one is really around like driving a data-driven, culture, and, certainly we see this at Microsoft, recently. Like in the last few years we are definitely pushed to, include, data in all of our decisions I know there's another p.m. on another team who, the, a comment, or a quote famous around since, since when do we make better decisions, without data and I, think that is something that is is actually, showing up in businesses, today and we're going to show you some examples of how you can include more, data-driven decision-making, in your, use cases by enabling, or including, flow. Also. Democratizing. Ideation, so today, we're going to talk about bottom-up. Innovation and, so that's not, a very well-known term i-i-i would like to say that actually Kent, is the originator, no no okay. Okay well, and, so democratizing. Ideation, it means that, we, should allow everyone we should enable everyone to be solving the problems right it shouldn't just be I t's job to, solve the problems for the business we should be enabling the business user to have tools and the ability and the empowerment to go and solve these things on their own also. In so then again bottom-up, innovation and, so that, is the frontline worker taking, on that that responsibility. To figure, out what are their problems in their line of work and how can they be the ones to solve them having, IT and and c-level, as supports. Instead. Of driving, the change downward, it's it's going upward. So. Yeah the the question, is like where does bottom-up, innovation occur, and the hint is that it doesn't include or doesn't happen in the c-suite because. The reality is if it did there. Wouldn't be this problem, there wouldn't be this push, for. Driving. Transformation. Opportunities. In businesses, using, technology, and as, John mentioned it does occur closer to where the problem exists, the, people that are closer to the problem have, a better idea of how to solve them but the problem is in the past these, people haven't been, embraced. Or empowered, to sort of be the thinkers, and actually. Come up with the solutions, to the ideas it's, typically driven top-down. Through, large bureaucratic processes, and I think what we want to talk about today is how we can unlock those opportunities, by, using the power platform, and more, specifically, Microsoft. Flow and, so. How can, you, drive, engagement in in your organization, around digital, transformation. Initiatives, and bottom-up, innovation. And. This. Is where we introduced the power platform, and specifically. Today we're going to be talking about that right, inside with Microsoft, flow but.

The Microsoft Power Platform is I'm, sure as you guys have all been hearing, about here, power. Apps power bi and flow which. Is built on top of a sure it is is in support of dynamics, office and also, all of the dynamic standalone, apps it's a new. Powerful, suite of tools for you to be able to build and extend apps outside. Of the box outside of the standalone offerings, and, something that we often talk about with, the power platform, it it is a democratized. Platform, it is a power. Platform, for power users and what this really does is it aligns well with this, notion of bottom-up innovation once. Again going back to having, people that are closer to the problem solving the problem that's, actually what we're actually doing by like. Embracing this power platform, is giving these tools into, a broader set of users that can actually go out and solve these problems, for their business. Ok. And so some of the tasks, that we've kind of identified, around driving engagement is first, of all you have to create some sort of foundational. Knowledge and, so, we'll talk a little bit more about that about an, actual use case where we drove bottom-up, innovation in, an organization, and so we'll kind of touch deeper on each of these but first, you want to create some foundational, knowledge everyone, has to have a basic, understanding of. What's going on and what the tools are what's available to them secondly. You have to define clear success, criteria. And parameters what, does it mean to win what, is a successful initiative. What is not do, we are we looking to save hours are we looking to cut costs we're looking to generate revenue you have to define those things so that people can attack the problems properly, you, have to also establish ownership, if you, want people to drive these initiatives, through give, them clear ownership, but also you. Have to have milestones, measurements. And successes, along with that right those, people who are driving these initiatives forward they should also be recognized, and rewarded for. Taking on such work in trying to drive and change the business one. Of the very important. Parts of this though is that leadership. Is there to unblock not, direct, in bottom-up. Innovation the. C-suite, is there to listen the, frontline workers are there to, be solving the problems, and and the c-level is there to make sure that that happens, and last, but certainly not least you have to accept some failure. Some things aren't gonna work some, owners are not gonna drive them through to success you have to be willing to have a reasonable, failure. Rate as well. So. Now John did you make this all up did, you sit down with marketing, and say hey let's build this crafty plan I did. Not know nobody didn't, and I know you didn't because. We. Actually worked with a customer, on this, they were you, know doing their own transformation so, full credit to them and we were brought in to help support them in terms, of getting some more ideas or, ideation, sessions going to, fill their pipeline, with, new opportunities for their business, so, trans elta is a public, company they're multinational. They run or operates in Canada, the United States and Australia and so, they do power generation, and energy, trading and. So they were well upon this journey of, business. Transformation. That. Leveraged bottom-up, innovation and, so, what we had the opportunity to do is myself and John and a couple of the power apps folks had, a chance to visit them in April and we spent an afternoon with them and actually went through a workshop and this was a cross-functional, workshop, this was not IT you. Know just with IT and I do come from an IT background so this isn't about alienating. Them I think it's more about bringing everyone to the table with, the common purpose of driving and outcome for the business so, as part of this working group we had people from energy trading Human, Resources operations. Supply chain management the transformation, office and IT, were in attendance and what, we did is we like, back to that point around creating that foundational, knowledge we, spent 45, minutes with them walking, them through power ops what is the tool what is it capable of the, art of the possible and. Then I think it was also really important, for this, broader audience to understand that this wasn't a sales pitch they, were already using power apps they have some interesting stories to tell you might, have seen their logo and some of the the power apps slides in terms of customers, so, we wanted them to highlight, to their own business that these are some of the apps that IT had built and some, of the results of that the.

Great Part about that was that you actually saw the room light up and begin to understand, like these things they had been interacting, with they, didn't, know it was a power app they didn't understand, that it was something they could utilize they just thought it was something distribute, that, they were using when, they'd be good began to understand, that these things that they were using were part of this platform and that they too could use it it was amazing. To see some, of the frontline workers begin to, just get ideas and get, excited. Absolutely. So the next thing we did is we did the same thing with flow and we, spent 45, minutes talking about flow, showing. The art of the possible showing, a lot of innovative use cases using cognitive, services some of the things we're gonna see today and then we also handed, the podium over to them and gave them the opportunity to present how they've been using flow within, their own business and I think then like, at the end of these sort of knowledge transfer, sessions, it got super interesting, we, broke out into cross-functional, groups this was something that transfer our transformation. Office took care of and ensured. That there was a good mix of IT and, operations, and trading folks in order, to really come up with these ideas because, I think what we identified. Or what was very obvious, was that the IT folks, had a lot of good ideas in terms of how to solve the problem, but they weren't necessarily aware of all of the challenges that the business was undergoing. So, it really became this collaborative, process of, like hey is this possible, could we do this and, be like yeah sure that sounds reasonable that sounds like a good idea let's mark it down now. Back to the previous, point around accountability, walking. Out of that room there, was someone that was on point for, all, of the ideas that were generated as part of that group they, they needed to put it into the funnel into their transformation, funnel where, they would have to go ahead and create a small business, case the. Small business case is not something, that takes weeks or months this takes days and generally.

What They're talking about is what, is the return what is the return to the company by, making this investment what. Are the timelines there was some clear parameters around the, duration, like you did there wasn't like five year payback they're interested in less, than two your payback sort. Of another like key, part of the Power Platform is you can address those time to value conversations. So. Beyond that what. Were the results what. Was the outcome we, had 60, plus ideas. That were identified, so it was great we had HR, people and IT people and in operations. And c-level, leadership, in in the same room and and, it was amazing to see the front line workers rattling, off all these ideas and the c-level people just sitting, there stunned have no idea that these workers were encountering, half of these problems, so, they identified, 60 ideas 15, of them got added, to the transformation, initiatives. With owners, and and. Within two months they, actually landed. One there in the process of landing several more and, so they took their whole Kent what is sem what was the process they moved to LA and management, contract, process there it was and they so they moved that fully over to flow as the workflow. Engine and. There was another example had the opportunity, to share the stage with the trans Alta folks at the power. User, group last week in Calgary, and they talked about another app that they built a power app that, they built as part of this transformation, initiative. And it had to do with de-icing, of wind, turbines and in, the past what's happened is they, operate in eastern Canada, you, get some ice on wind turbine blades you have to stop them it's both a safety concern and a maintenance concern, which is a very expensive problem, if you try to run these things too far so, naturally, with like, slower, access to data they would be run more conservative, in terms of how far they would run these turbines well using the power platform, they built an app where, they could actually run the turbines further, and that, allowed them to actually generate additional revenue. And revenue. In excess of multiple, millions of dollars within a few months not insignificant, I'm glad you could share this one this time yes the public. And. Then in addition to that they said that they it cost them about 30k, to actually build this app over the course of you know a month month and a half, so, if you want to talk about like ROI I. Challenge. You to come up with another platform that can deliver on on better, returns, than that and. Be able to do that with just a small team this, was not a huge development project, with like 10, 15 people on it it was really someone from IT working, with someone closely in the business and being able to pull all of these different data sources both, real-time from the SCADA Network from, the you, know whether whether Canada, historic. Weather data or weather data pricing, data and, actually give them a tool that, would do all of this trending and analysis and give them real-time, alerts so, there's a really interesting. And. Real, story that really sort of highlights, the opportunities, when people embrace this type of technology. All. Right and so now let's talk a little bit about Microsoft, flow so before. We begin who. Here, would, say they are an expert in Microsoft. Flow let's get you guys out of the way an expert who. Won. A mod, he's. A flow MVP he better raise his hand. Okay. I've used, flow and I'm somewhat, familiar. Okay. I've never touched it at all oh, good. Welcome. Thank you for coming I'm so happy you guys especially are here no offense to the rest you guys but the newbies I love the newbies okay. So what. Is Microsoft, flow it's a tool that allows you to work less and do more that's. A marketing line that we threw in there for you but, what does it do it helps people work smarter by automating, processes across. Apps and data flow. Works great, in the platform, it connects everything, it is the glue and so. We have a couple different kinds of flows we're only going to talk about one today if you, come to the flow session tomorrow, at. 12:15, for me and David Blythe we're going to talk about business process flows there if you want to hear about that today, we're gonna focus on automated, flows in, an automated flow is simple it has a trigger and a series of actions it can have conditional, logic it can branch off and do parallel things but, ideally, it's a trigger in a series of actions. So. What, are the characteristics, like I said it starts with an event or trigger performs. Any number of actions wow, I just said this whole slide I didn't even know what was coming can I have conditional logic and it can actually call other workflows.

So We have this idea of nested, flows where a flow can call a flow and you'll actually see a little bit of that today. Integration. And connectivity, it's one of our biggest strengths, so, Microsoft has three, clouds, they. Are office, dynamics. And Azure and and. Lucky for us we, are the beautiful glue, that sits between them all we, have connections, to over, 230. Products, we have all of the first party Microsoft, stack that you would expect as well, as many of your favorite third party products there as well I would challenge to say that each of you could, look at this list and probably find five things that you use right, now also. If, you have on-premise. Data, we have a gateway that will allow you to connect to certain data sources on-premise, and then, if you have a service with a restful api you, can actually build a custom connector to, connect your service into flow so that your customers can begin to use the, value of flow and all of the other connectors, as well. Now. Templates, so for you newbies this is an, excellent, place to get started, templates, are not only a great way to just jump in and like have a flow, kick off and start running for you and doing something but. Also if you open them and click the edit button you get to see how a flow is built so, when you read the template that says okapi Dynamics, accounts, two common, data service, customer, organizations. You. Can then see oh okay I'm gonna edit that here's, my trigger which is when a customer is added and oh, I see the next action takes, this and and says move to common. Data service and so it's a great way to learn, and see how things are built to understand, the logic of of flow, and workflow. Alright. And now I'm gonna do a little bit of a click through especially. Since we have so, many new faces, here, and we're. On number 6. All. Right. There you go perfect. Thanks. Kent, alright, so. I'm. Gonna take you through this site a little bit so I can give you an idea of what everything is so, that when you land on the page you're, not too, too, confused, right so first, things first I want to teach you guys how to get support and understand. Flow so, the first thing to know is the learn tab the, learn tab obviously. Is where we're gonna get started learning and so I would suggest the first thing you do is go to guided learning there's, 14, videos there that walk you through the basics, of how to build flows, that, it teaches you what is a flow it's each's you various, different types of flows and approvals, and things it's a great introduction. Now. From, there if you really want to start jumping in you can go over to our documentation, and we've really tried to make this simple if you're a beginner there's a tab for a beginner if you're intermediate I don't, like how it says I'm an intermediate. That's terrible grammar but, if you're intermediate flow, there's a tab for you and expert and admin so on and so forth a very, good place to get started now if you. Want to start asking questions and, connecting, with experts. And getting beyond the simple, I would. Recommend jumping into our community, so there are over 100,000. Monthly, active users in our community, our MVPs, have a heavy presence there we actually have a support, staff that helps make sure that questions, get answered, and so it is a great place for you guys to jump in ask questions, start, learning and connecting with other people that are building flows I, cannot. Recommend enough how. Important, it is to be a part of this community it is such an added value to. Have the supports, and the resources, that go along with this it really really, really will add to your, overall journey, I promise. So. Now let's go back to the flow site and I, want to point out a couple of things on the homepage here, we, have featured, template, collections, in the center here and so, they're based on job role and so if you are. A software, developer, if you're in marketing if, you're a teacher, if, you're an HR there, are collections for you and we are continuously, trying to add them and also, add to them, now.

I'm Gonna take you into, the flow, editor. Here is what we call it and I'm. About to spoil the surprise this. Looks different than what you're going to see this, is preview, you guys can sign up for the preview tenant right now and and. You'll see this new flow experience the rest of you in a few weeks will inherit, this experience, so what, we see here is we have a tab for my flows these are the flows that I create, that I run for myself that I have not yet shared, with anyone now. If it's, a very very important, flow let's just get this out of the way if it is a production, flow that, is running something important, please, never be the only owner for, those of you guys that are new in software, development if you make code if you make things in production you're never the only owner of those, things you create and so, with flows it should always be the same you don't want to be the only owner because then when you leave your credentials, die everyone's, in a panic how do we fix this flow so, that's, where we, see team, flows and so. These are the flows that I create, and then I share with others and now, it's a team flow when I leave Kent, could then go ahead and put his credentials, in and just keep that flow running. Also. Separate to that said. We're not going to talk too much about it here I'm gonna show you this business process, flows you. Need a P to plan to. Use B P F's and, and. Like I said if you want to hear more about that come tomorrow and, now. I, also. Want to touch on the, approvals, Center, so in flow, we, have a first-class, approvals, experience, I like, to call them lazy approvals, because they they're everywhere, so, when I get when I send an approval out it goes to my email it, goes to my mobile device if I have the flow mobile app it, also goes here into the approvals, Center. And. So. If I had any pending, approvals, you'd actually see cards show up here with all the requesters, information, asking. Me if I approve or not ah, let's. See so in this environment we haven't sent any approvals, yes so but, that is a perfect, place now the phone mobile app I touched on that for just a second guys, that's an awesome experience, it empowers, flow buttons, and buttons. Are manual, input flows you just press the button and then something happens that's the trigger press the button and so, what, what I actually did was I created a button that, would collect site, requests, so like team site requests, if someone requested it I had a button that would press and it, would gather up all those week's requests, and send me a report right. Also, another button example, that's fantastic. It's isn't it's the simplest one I've ever heard of but I love it we. Have a customer and they, have a warehouse and every, day at the warehouse the truck is supposed to arrive at 11:00 but sometimes. It arrives at 10:30, and sometimes, it arrives at noon and oftentimes. There's a lot of guys standing around for a lot of time waiting for a truck that didn't, show up and so, what they did was they, put a flow button by the door that sends, a text message to the whole team when the truck arrives so the first guy to see the truck hits the button everybody, gets a text they show up saved, him thousands, of dollars for guys just not standing around waiting for a truck and it's literally a two-step, flow push, the button send text they created it in 30 seconds. With. That said I'm actually gonna go ahead here into flow and I'm gonna just show you real quick how to construct, a simple, two-step flow. And. So. I click new and create from blank and I. Am dropped here into the flow editor now, this is where I am asked to select my trigger I'm gonna go ahead and start with the flow button for mobile. And. So. Then I'm gonna add an input and a text field and I'm gonna say, message. If. I. Could type, message. Now. In the next step I am. Going to add the info. Bip service. And. I'm gonna say send a text message and. So. Then actually what. I'm gonna do here is in my next. Step I'm going to add another input and I'm gonna put phone number. And. So. Now what, I'm gonna do is I'm going to open up my action. Which, is the send a text message from info bip and what, you can see happen is there's this drawer that kind of pops out from the side we, call this dynamic.

Content, Dynamic. Content is all of the information, that's fed to you from the previous, step in flow. As you build all the information, from the top steps becomes available in the subsequent, steps and so, you can see here, manually. Trigger a flow that's, my trigger and these, are all the data points that are now available to me from that button press we give you a lot of things the, date it was pressed the address, of where you are latitude. Longitude the, country this city, now. You can see I put into custom, inputs message, and phone number and those are available to me and so, I go ahead and I click right here into recipients, phone number and I just clicked that phone number box now, whatever I put into the first step is just gonna get handed into the second step and then, I'm gonna go ahead and say what do I want the message to be and. I. Enter that there and that. Literally is how. You make a flow that can send someone a text message and so. It took me all of two seconds, I pass my values, through I hit, the button it's gonna open up and show me two boxes you. Know what let's run it real quick. It's. Let's, let's test, the flames oh. You. Got to do it Kent never all right we're gonna skip that for now and get into some better demos actually so so, here's one thing I want to do with you real quick I want. To refresh this, and. So. Here. Is our digital, transformation. Survey that. You guys took, earlier. And so. When we said how, equipped, are your customers, for change by, far everybody. Said about a three not. Bad not, bad that's that's way better than a one those. Of you who said one in there I'm sorry that's that's, rough. Count, of what is holding them back mostly so it's a traditional, and stuck mindset, people. Are stuck, in the past and, then. When, what. Does digital transformation. Mean to you it looks like a fairly even split between. Seamless. Customer experience, across devices and channels and engaging. Customers, via, smart connected services, it's. Interesting so, what. I why I do this is a couple different reasons one I'm going to show you the flow behind the scenes and how this is made but, to. It just goes to show how easy it is to do a little bit of digital transformation, right. I now, have, great insight, to understand, how a crowd of 600. People thinks. Of digital transformation, and now when. I want to go back home and create my next marketing campaign, to promote this material, I now. Know what to target, so. Thank you guys for that and, also, for those of you that didn't participate, in the survey, those of you that did you're actually we're talking about all this bottom-up innovation stuff, today and what we did at trans Alta those of you that did the survey received an email with a link to that material so you can actually lead the bottom-up, innovation sessions. As well with your customers, and your organizations. Excuse. Me. Let's. Go ahead and show. You how that flow, is built and then I'm gonna hand it over to Kent to, show you guys the really cool stuff. All. Right so what I did was I first went and I built a form that's, the easy part, I'm not gonna walk you guys how to through how to build a form but, then what I did was I went to flow and my trigger was when a new response, is submitted, and so. I took that trigger and I said connect it to my form, over here then. I went and I added my first action, which is get response, details, also from Microsoft, forms so the first one collects all of the surveys lets me know there's a new survey the second one goes in and pulls all of the actual responses, next. I created, a streaming data set and power bi because, flow is amazing, and can create anything. Turn anything into a streaming data set so if you guys are power bi fans you're gonna you're gonna love flow and.

So I created a streaming data set I wire everything, right in and then, I send you this nicely formatted email I grab, your email address from the forum I grab your name from the forum I format, it with some nice links, you, get tons of useful information and a thank you and a reminder to evaluate. This session literally. Guys I put this together last night in my hotel room in an hour, the forum the flow the power bi dashboard, everything. And actually it wasn't my room it was a mods room. All. Right Kent. Over. To you ma'am so I think what's also interesting is we talked about like developing, a data-driven culture. And that's, just a very quick example, of how you can actually go ahead and do that just, by simply collecting some information, from, customers. From your end-users and how you can actually then drive you, know some sort of insight so I think that's pretty cool and it's, something that you can all go ahead and do today without. A whole lot of friction. So. Next we're going to talk a little bit more about, some other, technologies. And some other examples, now, who here, by show of hands has heard of something called robotic, process automation, RPA. And. So RPA is a type, of software that mimics, the activity, of a human being and carrying out a task within a process it. Can do repetitive stuff more quickly accurately and tirelessly, than, humans, freeing them up to do other tasks requiring, human, strengths, now, we're not positioning, Microsoft, flow is a complete, rpa solution, that's not what not why we're here today but. There's many different scenarios that. Actually can be addressed by using Microsoft. Flow and, probably, more interestingly is, that, you can do a lot of this without any additional, costs, you know beyond what you already have entitled to through. Office 365, so, we wanted to show, some some use cases around this and I think it's also, important. To understand. Like that there's a new sort of class of automation. That's emerging, called, intelligent. Automation. KPMG. Recently, did a study that says. That this market is currently at 12 billion dollars and, they expect it to become a, 232. Billion dollar industry, in five years in 2022. And/or. 2023, rather and so, rather so what this means is that this. Is not going away this notion of automating, but then also being able to automate more, intelligently, by, taking advantage of cognitive, services artificial.

Intelligence And machine learning and certainly, that's the theme of what, we're gonna be talking about here shortly and I think what's really interesting is, that these, are scenarios that would have been cost prohibitive, to develop just to three years ago you, would need a lot of developers, with like a strong understanding of a, lot of these these technologies, like machine, learning artificial intelligence. Microsoft, has really democratized, a lot of this through. Using flow but also through using the azure cognitive, suite. So. What are some common use cases around our PA or intelligent, automation so. Certainly customer, service dealing, with status requests, requests for information support. This, is going to be focus on some, of the demos that I'm going to show you here shortly, certainly, finance, accounts, payables, where, you need to be able to inject. Or ingest an invoice, and be able to pull it apart and, actually feed an ERP. System in order to provide, some processing, historically, that would be achieved through swivel-chair, integration. Where you're in one screen typing away and then you copy it and then you swivel around and then you paste it into the other system that's a great term switch, air on it I didn't create that one either. So. Procurement. Requisition, approvals, we've talked about that where we can use Microsoft. Flows first party approval capabilities. Maintenance. Human, Resources timesheets, onboarding, benefits, we've, seen many customers, use technology, for onboarding whether, it's provisioning. New azure ad accounts. Adding, users to groups provisioning. Teams for provisioning, sites and then, certainly information. Technology, the IT department, can also leverage this technology for, dealing with password, resets, diagnose. And, fix technical, issues we, have one customer that's been able to you, know automate, the creation of a ad account, and sync them with Dynamics, 365, accounts, something, that saves them you, know half an hour every single day but it reduces the mundane, repetitive manual, work and actually, allows them to automate it so the, first thing I want to talk about is automating, customer, service and. Yeah. Actually. Sure we'll fill the video well show it again all right so so yeah go ahead show the video. All. Right. You. Got a second sure okay. So. I got a customer and they're, wondering, if we could build a, solution, getting, inundated with emails and, so, what they would like to do is set up like a database, of questions and answers where. They, could have commonly, asked things and then, let's say that an email came in and they already answered that question before, what. If an email automatically. Went out to that person with that answer that they already had do, you think that's. Something like that is possible so. Yes. So. Recently, there's a service. Yes. So, we, could likely leverage that okay, and catalog, all of these questions related, answers and then we'll actually understand. What, the question is, line up the answer and email. So, a question will get asked, something, will happen and answer, will automatically. Yes. All. Right and so what, we have there is quite, a regular, occurrence. Kent. I often. Times bust. Into Kent's office with, these random ideas, like Kent, can, you build this and it's, amazing, because every. Time he's, like yeah, yeah I could build that and. And, so you're gonna actually get to see this I I literally, tricked Kent into automating. My email inbox in, the process, though we he, created, quite an amazing solution, so Kent, man show these guys how you made this hmm, oh. Uh let, me put you back number six. Yeah. Now we share an office so he doesn't have to even bust him that's right that's, right they put Kent and I in the same office now I think, cuz we were across the hall and I kept yelling at him so they're just like we're gonna put him with Kent so, now he can't escape me at all, so. This this flow is a little bit more advanced and not all of them are this advanced but I think it was an opportunity, to sort of demonstrate a capability, we haven't talked a lot about before, and, that's this ability to have, a flow call another flow and so, what we have here is this Q, a worker flow that sits in the middle and what.

It Does is it exposes an end points that can be called from other flows so, as you can see there's a few different icons up here where we've, got emailed, which John mentioned but. It's like kind of like why would we restrict this just to email what if we included, Twitter. As another input channel, and then what if we include the at bot you. Know that's a chat. BOTS connector. Framework that's a part of our, ecosystem as, well so, it's like why don't we just sort of inject all of these different entry. Points and then, what we'll do is we'll have this worker, that'll go ahead and process these, requests, call-outs. To QA maker and see. If there is an answer for. That specific, question now. The notion is that sounds really scary like oh you don't want to automatically, email people that's gonna send wrong information no I do I want to automate I want to have all my emails automatically, answered yes, but you want to be able to do that in a reliable fashion master, so that's true we wanted to build some more intelligence, into this process so, what this means that you can have it when, it's really confident over 90 percent, confidence that, it's found the right answer, to go ahead and automatically, send those emails out but, what if there's a gray area what if it's like sixty percent confidence. Don't, you want to have the ability to actually sort, of review that before the email gets sent out and perhaps creates. An embarrassing situation as, my least favorite thing is being embarrassed so. What this allows you to do is to use approvals, to, get in the way to say hey this is the proposed response, that's coming out of our knowledgebase do. You want to go ahead and send, this out and mortgage. King go ahead and say yes a proof that sounds good go, ahead and send it out now, in the case where there's clearly not a good answer then, you can actually reject, it and actually, take the inputs and feed the Machine so. That's what's represented in, this update flow, in the upper Henry right that corner update flow is when you reject you actually put a comment in of the answer you want to send and that actually. Sends that answer through it sends the answer back to the request address also updates, Q&A, maker and actually, republishes. The knowledge base so that the next time that that question is asked, I don't have to answer it exactly, this is a sort, of an opportunity to build, a really intelligent solution, with, all of these components without, writing any code, and guys. He literally built this in like one day I told I asked him about it in the morning and by the afternoon he, had a demo ready for me he's like like this like, just. Like that. So. What we're gonna do is we're gonna go ahead and send. Off a question. And. You. Know we'll keep it fairly simple but like, I said when there's situations where there are more complex you, have the opportunity, to inject and update so, the question we're going to ask is what is the pricing, model for Microsoft flow I'm going to send this email to. My work, account and I'm going to include this. Subject a flow question, because. When I have the the email processor, it's going to be looking for that specific subject.

So, We're gonna go ahead and click, send so, that's gonna go off and be working in the background but while it's doing that what. I do want to show you is Q&A maker so, this is Q&A maker you can access it at QA maker a I, it. Is an azure service there is a cost for it but it is a consumption, based service, it's where the jeans yeah, it's like a cup of coffee a day generally, and. What it allows you to do is to create these knowledge. Bases in this case it is going to be. You. Know intelligent, email routing and what, we can do is we can go ahead and provide. Questions, and we'll, say Kenneth for example can I you, restore a deleted flow, and the, answer is yes. You can with, a support ticket now. That's not the only way that you can ask that question that, kind of question, could be you ask many different ways so what you actually do here is you go ahead and add and you can add additional ways, that you would ask that question and what, happens is it actually uses, natural. Language understanding based. Upon the Luas technology, and other Microsoft technology, to try to understand, your intent, and then what it'll do is match your intent, with this specific answer, so. That's that's the core of the service let's, go see how we're doing right now. Okay. So cool it is running, and. So what's happened is we've, now picked up the email from me it might inbox and we'll it'll match that subject, criteria. Of flow question, it's then going to convert the, HTML, email naturally a lot of emails these days are in HTML format QA. Maker could care less about CSS. And markup in HTML and all that goop so, we're just going to extract the text out of it and then, what we're gonna do is we're gonna call a web, hook and what this is doing is it's actually calling, that worker flow that flow in the center I talked about and basically registering. With it saying hey when you get the answer that you need call. Me back and what, this allows us to do is avoid some timeouts that are just inherent, in using HTTP. So. If we flip over to that worker flow. We. Should see that we have a flow. That's in flight. And we. Do and so. What's happening, is it's actually gone out called. The Q and a maker service, and actually. Got a response, back and we. Can take a quick look at this so. We talked a little bit about flows calling, flows that's what you're seeing here the first email flow calling. Into what we're calling the QA worker, flow and so, here we have the question that was asked do you have any pricing information and, then. What we'll see is that'll, give a response, back, so. The response is actually hello, there the flow pricing page can be found here go, ahead and click this link and actually, there's a link to a podcast as well, but. In, this, case what we're wanted to do is we want to be able to approve that so, in this case what would what. John would receive is a notification. To, go ahead and approve this specific. Request and so, here we can see and there's some HTML markup here to make it look a little bit prettier we. Can see the proposed answer and if, we're happy with that proposed answer we can just go ahead and click, approve, so. We'll go ahead and do that and, then what will happen is that, our flow will. Complete, running we, can see now it's it's completed and it'll. Then return that information. Back to the, calling flow, and we can now see that this has, been completed. As well so we've gone ahead and sent, an email back out to the originator, which. Was that admin appear, the account I was using and saying, you know hello there this is where you can find the flow price, pricing. Information so. If we head back over to our email and hit, our inbox we. Can see that the email has arrived, and we can see the, the, response, and what. The original question was as well so, obviously you can go ahead and tailor this customize, it as you see fit and, go. Ahead and update this I see a lot of you guys taking pictures of these flows if you want to see how this is built stay.

At The end I can share the YouTube channel link with you guys Kenton, I make videos regularly where you can see us doing this stuff all the time and. Now we don't have time to run through the other scenarios I want to get to the other demos. But I will just show you the logic in terms of how you would deal with an unknown, question, now. In this case what we're going to do is once we call the QA maker service, we, can go ahead and a value to score and we can check to see if this score is above 90, if, it's above 90 we actually will automatically, send that answer, back without. Doing an approval but. Otherwise we will start an approval, and we're. Going to go ahead and you saw that like here's the here's, the question, here's, the proposed answer, and if. It's approved will automatically, go back and send it back to the the email publisher, but, otherwise if, we're going to say reject this is where we'll provide the, new feedback the, new answer and we're going to update a SharePoint, knowledge, KB. And the. Reason for doing this is we don't want to have to give everyone access to the, QA maker service, so this also allows other people to go ahead into. SharePoint, and add their own questions, and answers and so, we've sort of decoupled, this process and created another flow that, subscribes, to these types of events and will then go ahead and pull. From SharePoint and that's this flow here we'll pull these new records from SharePoint we'll, go ahead and call the. Q&A. Maker service, to update the knowledgebase and then we can go ahead and, publish. The knowledgebase so that the next time that question is asked it'll, automatically, be answered and so what this means is by next, ignite, I shouldn't. Have to answer a simple email. So. The so, that's a demo one. Now. What I want to talk about is, is. Automating. Forms, processing. So I got to telling the story of how this came to be so, Kent and I are trying to figure out a demo we want to do we're like we got to we, need to add one more thing what's, it gonna be and we're. Riding in an elevator I don't even remember where there's a random guy with us and I was like hey guy what. Could you have in your life that that if you could have it automated, what would it be and he's like oh like. I have apartments, and and I have to deal with applications. Like crazy if I could somehow, automate. That process it'd be great I was like Kent that's it lissa applications, and so, again Kent was like I could do that hey and here we are and so now you're gonna get to see the, elevator idea, lease, application, processing. Yeah, and so the like. A core capability. Or core use case inside of our PA or intelligent, automation is the ability. Rape information, you, really want to reduce that manual, interactions, of copying and pasting information, so what if you had the ability to take a PDF, document and have, a process, that can go in and pull out specific fields. That you're interested in and then use it to update other systems, so, what we've done here is we've actually you. Know have a PDF where we'll actually send it in via email flow, will pick it up it'll use a third party connector, called doc parser, which allows us to define parsing.

Rules Where we can go out and pick out specific data elements so for example first name last, name age years, at last residents, the, you, know salary and, we can use this information and pull it out to actually create. Contacts. Inside of Dynamics 365, if, the contact already exists, we're going to go ahead and update it if the contact is new we'll go ahead and add it then, what we'll do is we'll create an opportunity and then. The opportunity, can be managed by you. Know whoever is managing, that process whether that's a, line. Worker or a manager, now, in addition to this and John showed a little bit of this is you know how, do we actually forecast, some sort of demand like how do we know whether or not we've got enough. Properties, that we can actually go ahead and rent so, we can actually see, what this upcoming demand is by, including a power bi a real-time, streaming data set so, now we've got instant, insight in terms of the quality of our candidates, plus what are they looking for like, is everyone looking for a two-bedroom two-bathroom or is people looking for a one-bedroom one-bathroom how do we quickly, understand, that demand and so that's the the point of this next demo where we're actually going to actually solve this problem so. Here, is the, PDF, that I'm about to send in there's, a few interesting fields that we want to capture one is the property name in this case it's a fictitious, place called Bellevue Manor we, can see the name curt, the. Last name of Steven's date of birth or email address which will be interesting in order, to understand, whether or not they exist or not where, they currently reside the. Company they work for their job role their, salary, whether. Or not they want a one-bedroom, a two-bedroom, and if they have pets so. We're going to go ahead and, send. That in. And while that's processing, I'm quickly going to flip over to doc parcel so this is a third party service it, is a connector, that we do support and you, can actually use this to create a parsing. Rules so what you would do is you would upload documents, much, like the one I just showed you before and then you can go ahead and create these parsing, rules so.

For Example we can go ahead and look at first name so, first name was as simple as going, on to the form itself and identifying, the location of, first. Name so, we would go ahead and do this to the rest of the document, where, we're interested, in pulling out that, information now. You might be asking yourself well what if it's not accurate what if it guesses wrong well you can also leverage this that notion of like predictability and, probability, whereas, if it's not you, know confident, that it's giving you the right answer it's going to lower that probability. And then you can actually key off of that include, some logic to say hey now I'm going to go ahead and off-board this to a human process and get a human - to actually get in the way because a computer, or machine wasn't. Able to figure it out so. Let's go ahead and we'll check out our, our, flow, here and so. There's two flows there's one flow that's going to be reading. From the Inbox pulling, out attachments. And uploading. It to the doc parser service and, then we have another flow, that's going to be listening for doc, parse or publish events and that's when we're going to actually going to use all of this data to talk, to dynamics, so, here if we go ahead and take. A look at this flow it's very simple much like the flow John showed before it's it has two steps when. A new email arrives we're, gonna go ahead and see, if the subject, filter is rental application, and if, it is we're, gonna actually go ahead and get that email and for every attachment we're, going to go ahead and upload it to the stock parser service so, that's completed successfully and then, what will happen is we'll head over to our other. Flow. Which, is going to be, pulling. Down to these new, documents, from, the doc parser service and then, doing all of this work with dynamics, and power bi so. Here what we can see is it's once again not over. The complex flow but. We're going to get this new document, now this used to be a PDF but. What's happened after we've used doc parser, is now, we have access to all of these individuals, so first name property. Name last, name middle, name etc we, talked about making dynamic, content, and so what he did here is all those fields that he put boxes around all those fields now get turned into dynamic, content so he can use them in the other steps so, exactly so for, example when we want to list records to see if this contact exists, we're gonna use the email address that was passed in so, this is completely, dynamic then. What we'll see if there, is a record if it does exist we, will go ahead and we will just update that contact, information once. Again we're going to use all of this dynamic content. That, was ingested. From the doc parser service where we can actually go ahead and create the the, contact or update the contact provide the ID include. Their last name their city their phone number etc. In. The event that it is a new contact we're going to do something similarly, the difference is we're going to create the contact not update the contact. Next. What we'll do is we will go ahead and create the opportunity, and we will link the opportunity, back to that contact so now we've got a tight, relational. Relationship. Between the, contact, and the opportunity, and, lastly.

What We'll go I guess not lastly what we'll do next is we'll go ahead and add, rows to real-time streaming data set so, we would go ahead and define all of the different fields in, the power bi dashboard that we want to automatically. Update and will, once again use all of that dynamic content, from the previous steps lastly. Because we don't want to leave the customer hanging did they get Micah applications, did they not get my application we're going to send them an email that. Includes their, opportunity. Number, or basically their identifier indicating, here's your ticket and, so, let's go ahead and check out. Dynamics. First. So. If we head over to. Contacts. We. Will look for Kurt. Kurt. Stevens this, was the the, PDF that we had sent in and, we. Can see their phone number where. They live and their job title so that that, was created. Successfully. Next. What we can go ahead then now do, is see opportunities, and we, should see there's our opportunity. For Bellevue manner so that was something we were able to pull out of that PDF, document, and we, should be able to go ahead and see. Some. Additional information about, what they're looking for in, this case they're looking for a two-bedroom, they, have pets, this. Is their current address, so. Let's say that you would historically, be a very manual process where, you're copying and pasting and doing. A lot of like low, value work in order to solve that business problem when you could actually automate, a lot of that and then, focus on on other, activities that do require human. Strengths, oh. Let's. Treat your power bi as well. Okay. I want to see this dashboard so here's the power bi dashboard that I built which, is going to give us some insight in terms of you, know what are our candidates, look like here we can see that we've got 23. Pending, applications, the average income is about 90 K we, can see the distribution of bedrooms, you know one bedroom two bedrooms, and also, bathrooms, as well so. Now if we're looking for forecasting, one understand, like you know hey maybe we can charge more for, a two. Bedroom because there's greater demand now we have that insight we, can also go ahead and much, like other power bi capabilities, slice and dice and look by different property name and, we can also check check out our moving dates so. If we want to just focus on specific. Time frames we can see that on December first there's five people that want to move in and then as we move outwards. To month we have one person that wants to move in on the 13th and one, person who wants to move in on 15th so now we've can visualize, all of this data in terms of instead of trying to manage it on a spreadsheet copying and pasting and doing, those type of activities.

Okay. So the next thing I want to talk about is. Pretty. Exciting so who here has heard of power query. Quite. A few that's cool I'm. A big, fan of power query I've used this before, in previous, life and also at Microsoft a fair bit but power query is a data connection technology. That enables you to discover, combine and refine data, sources to meet your analysis, and needs. Or analysis, needs so, this is not, a new technology this. Technology, is used by power bi and by excel so, this is something that is sort of been, battle tested and, is, obviously used by, millions. Of users and really, what it allows you to do is it to ingest data from all of these different data sources and then perform some level of transformations. Or mashups, and then you can use this mashup to then actually feed other systems. So. As of this week we are announcing a, preview. Of power, query for, Microsoft. Flow and this. Is available on top of the sequel connector. Now. The in. Terms of why the sequel, connector so there's a few things that we need to take care of one is data, loss prevention policies. So. DLP, is there to enforce, which. Connectors, you can use within a single flow or a single, power app and. Naturally. Power query has a lot of other sort. Of connectors, or data sources that it can talk to some, of them aren't actually in flow so, we don't want to create any sort of security gaps by, enabling, that once. We have the opportunity, to plug that into DLP, that's, when we'll see more and more data. Sources being available and. So this was very much an, opportunity, for us to sort of build this like, foundation, for power query and flow and then we can certainly have the ability to build on top of it so, this is going to leverage the power query online service, so if you are using Excel or if you are using. Power. Bi, desktop. That's. Using like a local. Power query service, so there are going to be some differences between p/q online and sort, of the core power query service itself we. Will do some throttling on you we, will throttle you to hours a day or 10 hours a week so, what that means is when you have power query online you have this service, that is actually doing all of this processing, for you so if you have really long queries, you, will eat into that entitlement, certainly. Interested in hearing feedback if that's not, enough we'd. Encourage you to reach out because I'd really like to know that so we can work, with the power query team on finding the right balance and right. Now we are providing, single, query support, but. I will show you sort of how you can get around this in. Order to build. Out sort of more comprehensive, mashups. So. Why did we build this. Number. One we wanted a more approachable, solution. For filtering and shaping data. And this is really an alternative to Oh data so, we've had some of. Our users they haven't been big fans of Oh data it. Can be a little bit gnarly at times in terms of the understanding the correct syntax so, this is a much more approachable way. Of filtering out data this. Unlocks, future, power query integration, so I'd love to hear from you if there's, other ways that you could envision using flow with power query so, that we can actually unlock those use cases and, really. I think the future of this and this is why I get excited is that today, we don't really have like a data mapper like you might see in some other sort. Of enterprise integration, tools but. Using, power query actually allows us the ability to mash up different, data sources so I can envision, a time this isn't a commitment, to roadmap commitment but a conversion a time from, like a vision perspective of say taking, in flat files taking in JSON files and being able to mash them up with some of these other data sources like sequel in order to really feed, downstream. Systems. Ok. So let's do a walk-through and then I'll show you a demo. So. I'm just going to go ahead and create from blank. We'll. Just use a simple button flow. As our trigger. Then. What we'll do is we'll go ahead and type, power, query. Then. What will happen is we'll actually see this action called sequel server transform. Data using, power query so I'm gonna go ahead and click on this now. What's going to happen here behind the scenes is I, need a connection, to that sequel server so this is part, of the reason why we need to include it as part of our existing DLP policies, is because, this is an existing, concept for us this. Will work just like any other new connection, to sequel server where, you would actually have to provide like the database name the, database server username, password, so, in this case I already have my connection, established, but I can now go ahead and click, create query. At. This point you're now talking to the power query online, service, and what.

I'm Gonna have the ability to do is select different, tables, that I can now use in my mashup so, here I'm going to go ahead and select apartments. Crip. Not, crypto tray its demographics. And customers. So. These are all tables, that are now going to be in my context. Next. I'm going to you, know go to the next screen. Where I now can build my mash up I can, use these different queries and actually. Aggregate. Them or merge them together so as I mentioned before will, only support you running one query at a time but you can get around this if you have multiple, tables by, going ahead and combining tables and merge queries, as new so, what I can do is a join I can take my, work order data which, will have a reference to a customer, and then, I can enrich it with actual customer, data so, I don't want to have like all of my customer, name/address the, demographic, info on my work order table I can now actually go ahead and essentially. Join these tables, so I'm gonna go ahead and, choose. So, I chose the wrong table but let's. For. The sake of argument I'll go ahead and say choose apartments, and I, could go ahead and choose customers, and then, what I can do is I can go ahead and link those two using. A, left, join click. OK and then. What will happen is now I have access to this. Dynamic content, and I'll actually show you more of this in an. Actual demo. So. Let's just explain that demo. So. This is it here now sort of in this same line. Of thinking with, customer, service so. We've got this person that's managing, apartments, and people, create work orders the elevators, broke in the, AC is not working the gym treadmills broken, feel like we got to find this apartment, guy okay, dude here's your flows. So, now what we want to do is actually process those work orders but, we also want to understand, like how happy, our customers, are they about to leave how. Can we get some additional insights. Into, that so what we can actually do is plug, in another one of these Azure cognitive, services and, in this case we're, going to go ahead and perform some. Sentiment, analysis on, top of it so, what we'll do is we're going to have a flow that will run on a schedule and what. It'll do is pull, out all of these outstanding work orders check, for the customer, comments, run, it through the sentiment analysis engine get, an indicator, back from 0 to 1 indicating. How happy or how dissatisfied they, are, and then what we'll do is we'll publish this into Microsoft, teams as a, digest, now does it have to be Microsoft teams it could be em

2018-10-04 04:28

Other news