Citizen-Assisted Development with Mike Fitzmaurice

As part of the #CollabTalk Podcast, I was able to sit down last week and interview Mike Fitzmaurice (@mikefitz), VP of North America and Chief Evangelist for WEBCON (https://webcon.com/). Many of you know Mike from his many years within the community, and prior to that, as part of the SharePoint product team at Microsoft. But many of you are probably not aware of what Mike is up to lately, and so I took this opportunity to talk to him about his new company, and get his perspective on what is happening within the ecosystem.

You can watch this entire video below or on the CollabTalk YouTube page, listen to it via the CollabTalk Podcast, or read through the entire transcript below:

 

Transcript

Christian Buckley (CB): And we are officially underway and Mike. It’s great to talk to you.

Mike Fitzmaurice (MF): Great to be talking to you as well.

CB: So why don’t you, for the folks that don’t know you or know your background, why don’t you introduce yourself?

MF: I am Mike Fitzmaurice. I’m the vice president of North America and the chief evangelist for WEBCON, but in truth I have been around in our industry for a really really really long time.

CB: A long time, yes.

MF: One of my claims to fame is that I am, and I’ve got the white beard and graying hair to show it, I’m actually the very first SharePoint evangelist. I didn’t invent SharePoint. Jeff Teper did that, so I’m not the father of SharePoint. I’m not the mother. That was a guy named Jonathan Kaufman who’s since retired. I was however a wet nurse. I was in the delivery room. I helped clean up the kid and I helped it grow up.

CB: I’m trying to remember where your office was, it’s because I joined in 2006…

MF: Yeah.

CB: I met you a couple times…

MF: Yeah.

CB: …while we were employees. I just remember going over — trying to remember which building that was. I remember where Arpan’s office was. Who did he share an office with?

MF: For a while he shared an office with Joel (Oleson).

CB: That’s right, I knew that. Before that, but anyway back in that era so remember so I was part of MMS that was the Microsoft Managed Services which became BPOS…

MF: And Arpan came in from content management server and I was with SharePoint… oh we’re really getting into old school for the rest of you guys. We don’t even really do web publishing per se at least not public facing websites and SharePoint anymore, but for a time the world wanted Microsoft to produce a single portal product and because the word “portal” was the thing that mattered at the time and some people thought of a portal as another word for a publishing vehicle and others thought about it as an integration vehicle. We had two products that were good at each and we, in 20/20 hindsight, it was probably not the best idea to have combined them. I personally think that if Microsoft had left CMS alone it would have become Sitecore about a year before Sitecore did. Not that Sitecore isn’t good — Sitecore is a nice technology.

CB: There was a lot of interesting things going on though. So I started… so I was one of the points of contact within MMS and BPOS with the HMC (the hosted messaging and collaboration) team and they were doing some other interesting things where they were taking essentially WSS and bundling other business solutions and that small business server solution and building things where SharePoint was a component of that and they actually did they have like a turn-key…

MF: We were trying to get people to do that and that was one of our successes. Another one would be Visual Studio Team Services…

CB: Yes.

MF: That was based around Windows SharePoint services back in the day. There were others too.

CB: I always like to point out though…

MF: The idea was to make the site the solution.

CB: Right. So I was just gonna say that this as part of the backgrounder, and then we’ll move on to other topics.

MF: Oh yeah, yeah.

CB: I will say that it was it was partly you know because of your recommendation that I took that job with echoTechnology.

MF: I’m glad you did.

CB: I left Microsoft in 2009 and I was interviewing with little echoTechnology, which was migration and change management software in the SharePoint space.

MF: Eventually acquired by Axceler…

CB: But I was also interviewing with Colligo at the time and talking with a couple others, and it was Fitz who’s just like called it…I remember the phrase that you used was the “diamond in the rough.” I owe the opportunity around that…

MF: No, they were.

CB: I agree.

MF: Had you gone to Colligo, they were… they’ve been nice people too, and you know…

CB: Sure, and I still have a relationship with them, and…

MF: Yeah, and they’re a Canadian company, I’m Canadian. You know I try to look out for the home team. That said, the thing I’ve been best known for the past decade or more is business process automation, application integration, workflow management, that kind of thing. When I left even at Microsoft I did a lot of work windows workflow foundation and application integration in the world of SharePoint, but I doubled down on it when I went to… well, when I left Microsoft I went to work for Nintex for ten years. Um, left Nintex after their acquisition back in 2018 and joined up with WEBCON for the past year. It’s been absolutely great. These guys have impressed the heck out of me.

CB: It’s an interesting space. I mean, look at the evolution of that… I mean it’s um, you you’ve got some big players that are really well-known out there and so I think that especially like here in the US I mean everybody knows the Nintex name, some folks are familiar with K2. I would say you know Europe, I mean WEBCON is as big or bigger of a brand in the EU. Talk about that, and your path over to WEBCON.

MF: Actually, this is one of the reasons I like the idea of working with them. WEBCONS has a ton of discipline. So rather than risk extending their arms too widely and spreading themselves too thin, they decided to double down and make Europe a 100% success story — and they have. We’ve just established a beachhead over here in North America, we’ve signed up some partners have made some sales, we’re doubling that if not more this year. Everything is we’re firing on all cylinders and hitting all targets but the idea this year is to become better known and better adopted over her. We don’t expect that to be an overnight kind of thing, but the truth is we have an incredibly compelling story and I haven’t demoed this to a single person who hasn’t immediately wanted to start using it.

CB: Well, it’s a… so just as far as the folks that are watching or listening in, of course, so I’ve started working with WEBCON and one of the most compelling stories is something I… actually, had the first article went live upon the WEBCON site. I’ve got another blog post it’s coming out this next week talking about… so maybe out by the time this is out, but talking about change management, and I love that perspective. Given my background in technical project and program management and the fact that that was actually my approach you know back… one of the things I liked about echoTechnology and why I brought that up was that it wasn’t about the one-time migration moving data from one system to the other but it was the concept that it is the reality is that it’s a repeated process, it’s an ongoing thing that there’s… your organization is gonna change, your business rules will change, the maturity the…of your end users and their needs will constantly change and you have to…and I say this often…

MF: You are bringing up the later term reasons to change which are ongoing. Even during the early phases… Look, you do consulting, you have done consulting I’ve done consulting. When have you encountered a client that knew exactly what they wanted in excruciating detail well before they put out a request for proposals? People tend to have a need… people have a desired outcome, but the road from idea to outcome is, shall we say, circuitous.

CB: Right, well, that’s the way that I describe that is that your lens of understanding changes as you move forward. And you might find that you might say that hey, here’s the ten things that I need, and those outcomes, and then you deliver on three of those and be like “Yeah, what I really need is…” and that shifts and that’s actually these…

MF: That’s what I asked for, but not what I wanted. Yeah, it happens all the time. But this…you see, we don’t find that annoying at WEBCON. We actually find that normal and it’s up to us as platform developers to work that out and build it into the system. So WEBCON, more than any other company out there assumes that you’re going to build something and change it constantly. And that could be changing the data schema, that could be changing the user experience in the user interface. It absolutely means changing the process and you’ll… It’s okay, none of that has to hurt. When a user or a stakeholder wants a change, the developer does not need to roll her eyes. If it turns out that assumptions turned out not to be valid, the designer does not need to sweat bullets. It’s…change is a happy thing. It’s not disruptive, it’s not just… certainly not destructive, and the nice thing about what we do is you can deploy an application and have people trying it out, then as you’ve gather feedback you can make changes to it as it’s being used. So this is a workflow process it could be halfway finished, you know ten steps you’re on step four. You make a change to what happens between step four and five, that change takes effect immediately. Change the user interface, change the forms, it takes effect on all the stuff in mid-flight. Think about what you had to do before. If you realized the stuff that was already in progress was wrong or at least incomplete, you had to either recall everything and bring it back, or you had to just accept the fact that everything that was already in flight was going to be wrong.

CB: Well, and the other side of this too, I think, and it fits in there, is that you know… It’s difficult in the IT world to go in where you understand what your business needs are, and then you have you can only go and build based on what you understand the limitations to the technology are.

MF: Of course.

CB: You know the technology is dictating how you can go and solve those business problems and you’re always gonna have to make, you know, adjustments or changes or even…

MF: Of course

CB: …you’ll have unmet needs because of those limitations.

MF: Yeah, that’s absolutely correct. So yeah, we are the guys who love change and accept it and make it as painless as possible. So most everyone there is either introducing some form of version control on models or maybe adding some application lifecycle management to what they’re doing. We solved those problems years ago and went far beyond that. So yeah, I think WEBCON actually fits the way people really build applications and really curate them afterwards. Curate is a fancy word but it actually really fits here because it means looking at how people use it, making adjustments, making it better on a regular basis, shepherding it, improving it, helping it grow, helping it adapt.

CB: You know, one other thought too — I did an interview with Office Apps & Services MVP Dan User out in the DC area…

MF: Sure.

CB: Uh, we were talking about… we were talking about how there’s just the reality on the ground with a lot of clients especially in the DC area and the public sector that still have your on-prem environments and are maintaining that and they have, you know… we get we could get to the detail of the real or perceived challenges of security and compliance and governance and all those kinds of things, and he argued… I don’t want to speak for him, but part of how I interpret how he answered was that he says it’s not even about those issues so much anymore. It’s just the hard facts of functionality, capability that they need where it’s best met by these hybrid or these on-premises solutions. So for organizations, again, where the technology is dictating “we must move towards the cloud”…

MF: Mm-hmm

CB: …and for a lot of organizations, part of factoring that in, the timing of moving to the cloud is you know we need to, be able to… we’re gonna go and completely rearchitect our solutions and rebuild from the ground up…

MF: Yeah, you raise an excellent point. There are a number… it’s not just government agencies. There are a number… for a lot of reasons. It’s not even necessarily large versus small organizations, because I see cloud enthusiasm everywhere, but I also see situations everywhere where it may not be the entire company. That’s more true in a government agency but it might not be the entire organization, but there are parts of the organization or certain workloads or certain use cases that for one reason or another people either don’t want to move into the cloud or they really, for lots of reasons, can’t. And to have one kind of approach for on-premises or one kind of approach for in the cloud and to make everybody do one or the other, it just it sounds like a nice idea it’s not practical. So for example, to do it to have a different set of technology at one place versus another, you introduce a migration conundrum. So to rebuild a solution that used to work on premises and build it for the cloud and change everything in the process so that it looks the same but isn’t the same, that doesn’t make for a happy customer trying to do hybrid scenarios or even migration scenarios. And asking somebody to do everything in the cloud and relegate… really not trying to pick on anyone in particular, but to say “No, your application’s going to run in the cloud but we’ll use a gateway to fetch your on-premises data” — that isn’t comforting to somebody who’s keeping that data on-premises for any of a number of auditing and regulatory and other reasons. It means stuff is moving out of their walled garden whether they want it to or not. The nice thing about WEBCON BPS is it’s the same technology platform whether you’re running it on premises or running it in the cloud in a shared services scenario, or you can even run it in the cloud in an infrastructure-as-a-service scenario like running inside of your Azure tenant or your AWS environment or something like that and therefore you get cloud functionality with cloud integration but you’re still completely in control over it. We give you all three choices and you can migrate them from one to the other to the other with impunity.

CB: Well, it made me think of… we were talking about this on this topic last month while we’re in Prague and it made me think of like software configuration management and the need to… where you might have different branches of that same business process but have four different business units and four different flavors of that business process and different infrastructural requirements and things and adjustments around that. And if you then have a business rule change, something that needs to change across all for those, you’re doing for independent changes in the in the typical model…

MF: Well, if you were using us…

CB: Right that’s my point.

MF: Yeah, we have a scenario… we have a construct called a business unit. It’s baked into the entire platform and some people use that for different countries some people use it for different branches of the organizational chart, whatever it is…is you can make your application behave differently depending on which business unit is using it at the time. You can also restrict it to certain business units or not others. But basically, you can factor in where the applications being used into the design of the application, which means again you only have to change something once and it just works everywhere. That’s another very handy thing. It turns out… you know, you bring this up, you brought up something else earlier, there are so many things that “high-code development” does right that “low-code development” really can benefit from. In some cases, you could argue that we’re reinventing the wheel, although WEBCON’s been doing it for a while, but yeah things like agile design and development and things like a DevOps approach, and rings and so on, everybody can benefit from this whether or not they do whether or not they’re doing professional code work. We try to apply those principles to a process-centric low-code, for lack of a better term, paradigm.

CB: That’s very cool. Well…I talk a lot about change management because…in my background in the project management space…I mean so much of my life was spent kind of juggling those different, the nuances between the flavors of the system, which is essentially the same solution in four or five different locations, of course having to go in and juggle between those, so yeah you know that that’s one of the messages that just I think really resonated with me when we started talking about that. You know, and the…that other side of having your requirements being led by the known limitations of the technology. I mean, how does that compare…I mean maybe you could compare WEBCON versus, you know, Microsoft’s current stack, and the solutions: Power Automate, the things that they’re pulling out there. What are some of the leading differences?

MF: They’re meant for two different things. Actually, I was one of the people that was happy about the name change because calling it Power Automate really accurately describes what the tool is designed to do. It is an automation tool. If you look at a Power Automate design, it starts out with perform this task, now calculate this, now do this, now touch that, now connect to this, now do that. It’s kind of like a program flowchart. It’s like a set of instructions you would give an assistant and they’re checking them off and performing them as you go along, and handing some the output from one thing as to into input in another step. That’s automation. It’s actually an integration. That’s a very valid use case. But a business process is something like I have an outcome I’d like to have happen. I have a certain number of assets that are going to be involved in this. I have people and roles that will be involved. I need…I have those assets, I have those people, I have these requirements, and I need to go through these various steps, sometimes moving back and forth, sometimes repeating something, sometimes not. The logic is the star of the show, and the payload is the star of the show. It’s very stateful, it’s not stateless. It has a life cycle and that life cycle could be very long-running it could be short but being very long. It’s like working on a case if you’re a physician or a detective or something like that, and it can take you in lots of different directions. Business process software certainly uses animation to  accomplish certain tasks in certain steps, but it’s operating in an entirely higher level. You could do business process automation with a combination of Power Automate and Power Apps and a bunch of…but you would wind up writing a lot of extra code yourself, or non-code assets, and the real business process management would be taking place inside your head, not in a platform design for the purpose. It might exist in documents, and you create a lot of automation steps to implement pieces of it. WEBCON and lots of…and look we’re not the only people out there in business process, and lots of companies are doing this. But the thing that differentiates us is we’re operating at a different level and we’re tackling a different problem. We use automation as part of it. It actually means that you could combine WEBCON and Power Automate to accomplish a whole bunch of things. Some people have even done articles and webinars on that topic.

CB: Yes, speaking of…kind of…what are your active topics? What are you presenting on? Like what are you talking about when you come for a Microsoft 365 Friday here in Utah in two weeks?

MF: Actually, the thing I just was telling you about that then…identifying different kinds of automation scenarios, and choosing which kinds of tools or which type of tools…I’m not making specific tool recommendations, but to say: If you’re doing application integration, look at these kinds of tools. If you’re doing data harvesting, look at these kinds of tools. If you’re doing content collaboration, look at the following set of tools, and take this kind of approaches and watch out for these…these bear traps and so on. I talk about that and I also talked about the difference between automation and business process management. So I’ll be talking about that…I also often talk about designing applications so that they can be more easily changed. Both this particular topic is designed not to have a single demo or a single screen shot. This is advice you can use no matter which tools you use, but basically there are ways you can design your approach to solving a workflow problem and that makes it really easy to change it later, and has a greater chance of success. And I’ve delivered that talk in different forms over the years, no matter who I’m working for. Competitors have been in the room and found themselves nodding the entire time. This is something everyone likes. I’ve talked about the stage and state machine type workflow design. This is a little bit geeky, but it turns out to have incredibly powerful benefits for people. Turns out that WEBCON based their entire platform on that. I didn’t know that at the time I started being an advocate for state machine thinking, that was nice. I talk about citizen development a lot, and in fact, I’m getting ready to debut a session in a few months about rethinking what we wanted citizen development to accomplish versus what is really happening out there.

CB: Yeah, that actually…

MF: I’m not celebrating it. I’m picking it apart, and seeing where it works and where it doesn’t.

CB: Well, so I was gonna ask you along those lines of whether you see that…is your focus, and WEBCON’s focus more on the citizen developer slash power user type, or is it on the typical…just business user?

MF: It’s sort of somewhere in between. WEBCON’s actually kind of fond of the notion of “citizen-assisted development.” So, in traditional development you sort of start requirements by interviewing your stakeholders, your citizens, and then you take all of that and go work on it somewhere else. And then you bring it back and hope everything is nice and happy, right? In citizen development it’s…well, we have an application backlog, and we have developers that don’t understand citizens, so let’s just let the citizens do all the work for themselves and hope that it works out nicely. Turns out there are a number of caveats with that, so we have a disconnect between what people want and what developers build. That’s one problem to solve. We also have an application backlog. Not enough of these guys to meet the demand that these guys…

CB: So that is…a lot of the push for self-service was around that exact thing…

MF: Right.

CB: …it’s let’s give you a lot of these different options. It is…let’s, you know, let’s focus on creating those things which serve multiple constituencies, multiple groups of users, put that out there, but then at some level then it’s for the end users to go and say, “Well I’m 90% there with the solution that my IT team built, and I’m gonna tweak it and adjust to get what I need for my specific requirements.”

MF: And sometimes that works and sometimes it doesn’t. So on the far end of the spectrum, you could just say let these guys do all the work for themselves. And there are some companies out there that advocate that approach, and you could argue the Power Platform people…well, their marketing people paint that picture rather nicely. I would argue the people actually delivering the product don’t make those claims at the exact same time, so that’s marketing kind of getting ahead of itself, because in reality there are a couple of scenarios…. when it doesn’t work, citizen development creates chaos, and a mess…

CB: That’s the “Shadow IT” …that’s, I mean, that’s the whole thing…if I can’t figure out how to do it and somebody puts a nice…tells a nice story about this other third-party product that seems to do better than what’s available there, and I can go purchase that and suddenly have these unsupported…

MF: Stick a pin in that shadow ID idea, cuz it’s really important. I’ll come back to it in a second. But when citizen development works, it works by turning citizens into developers. Some of the biggest success stories for Power Apps…they’ve got like their team of ten heroes or something like that, they tended to have their jobs change and now they’re in the business of building Power… They’re developers now. They’re low-code developers, but they’re developers. They stopped being citizens, and they might still remember what it’s like to be a citizen, and that’s a good thing, but one of two things happens: either you rise to the occasion, you become a non-traditional developer, or you stay a citizen and create a mess. We saw that happening time after time after time, and wanted to do something about it. By the way, the other reason why this is so attractive to have the citizens do all the work for themselves is because they don’t have to communicate this idea with the developer over here and have so much stuff get lost in translation. The problem is you need to do that anyway, because if I am over here building an application for myself, and I’m not the only one using it, at some point I’m going to go on vacation, or I’m going to leave, or I’ll get promoted or transferred, or I just won’t have time at the moment – and someone else is going to need to work on this. And if I didn’t explain it well, or document that well, or make it change worthy, no one over here is gonna have a good time. So that work you had to do, you still had to do it. Now what could happen, now let’s unpin that thing about shadow IT…

CB: Mm-hmm

MF: I want to talk about “Satellite IT” or “Satellite development” where you do have people in the trenches, but you accept the fact that they’re not regular citizens. They’re…they have development and IT and some dev skills. Let’s call them business analysts, and some people do that…but some people would call them operations people. Like a sales organization almost invariably, when it hits a certain size, dedicates one person the sales ops, and they manage Salesforce or Microsoft Dynamics CRM, or something like that and they do a lot of low-code development with it. Sometime actual code. That’s actually a wonderful model. In fact, I would argue that after a while citizens get tired of being asked to develop things for themselves, and they punt things to that operations person, or that analyst anyway. And so what becomes important there that they’re in the trenches, there’s satellite IT, they have a friendly relationship with central IT, so it’s not like they don’t get it and it’s not like they don’t have skills. But they also get these guys better. It used to be an analysts job was to translate user requirements into things geeks could understand. I would say that letting the analyst do the building as opposed to the citizens has a much greater chance of success, and so having citizens “assist” in the development process. But we have WEBCON call “Citizen-Assisted Development” ..that is something worth taking a look at. And so making it easily explainable as to what the analyst or operations person or satellite developer is doing, making it easy to collect feedback and have people participate in the process without actually having to do that work. Every time we’ve seen someone do that, it’s worked swimmingly, without exception, and in truth everybody I’ve seen attempt citizen development in the classic sense, where you…everybody has the option to do it for themselves, it winds up evolving into this anyway.

CB: Well…it’s, you know..I think five six seven years ago we really started talking about kind of what’s the future role of the IT pro…

MF: Yes.

CB: …and I think one of the messages that I promoted, in fact, the first Ignite Conference that was in Chicago, so I had a session where I actually went and did research on, you know, it’s I would…I pushed into the partner and the MVP community to get feedback of what they thought of what would happen with that role, and one of the loudest messages that came out of that was that, well it’s these people that are really perfectly placed to move into that more business analyst-type function with their technical skillset to play this exact role.

MF: Great.

CB: Yeah, and I..so I completely agree with you. If your…if your job had been, you know, managing the servers and managing these giant business platforms and applications, things like SharePoint, and it’s not just like “Hey, I’m out of a job now that we use Office 365” you know, we know that that’s never been true, that the roles evolve, but this is a great example of what can happen – what that role can evolve into.

MF: Yeah. Plus, it actually fits the way organizations actually behave, rather than a vision some independent software vendor had for the way organizations “could” behave.

CB: Right. Well, that’s…I kind of put that in that category of…again letting, you know, the technology-driven decisions versus…

MF: Sure.

CB: …your culture of the organization and the capabilities of the organization…I mean, it may be that for some organizations where, depending on what those frontline workers are doing and their tasks, that  they just…they don’t have the time, or the skills to….capability to even step up into that you know expanded role and play that role, and you might hire somebody specifically for that  function, or pull somebody from IT down in the organization, but it’s….but I agree with that, it’s when you…in my personal experience is that when you make everything self-serve, it automatically, and what happens over time, is that people naturally gravitate into those roles, and end up playing that…you know, that that, again that…Assisted IT role…

MF: Yeah, as an advocate of citizen development, and as an idea, in general, I mean, I’ve been refining my opinion about this over the last year or two, but one of the things that many people including me recommended was: it’s all about the tools. Nailing the tools correctly is so important. Give citizen developers tools they can use where they don’t hurt themselves, where they’ve deliberately got limits, and the things they will build don’t need to be carefully scrutinized and managed. That in theory works well, and that’s a technology-driven approach. In truth, again looking at the way people actually behave, no, they won’t if they decide that they really want to push…if they want to use it, they find themselves spending all their time trying to hack their way beyond the limitations of the tool, and creating more chaos.

CB: Yeah, and this is a…this is a common problem within the DevOps space, and you know…

MF: Of course.

CB: It’s funny that…I brought up SCM, we talked about, you know, DevOps, it’s like the same problems…in fact, a lot of the issues that we’ve experienced within our community in the SharePoint world…you know, I was in the…so, back in the late 90s and early 2000s and I wrote books for Rational Software, IBM Rational and…

MF: Right.

CB: …and it was largely in that DevOps space. Of course, we didn’t call it DevOps. It wasn’t a category back then, but it’s all the same problems, same issues. We just have different labels now, in this community, but we’re living the same things over again. Hey, you know… that’s actually… so I had a thought earlier too… You know the sales cycle with enterprise collaboration, with any platform, that…with technology…

MF: Sure.

CB: There’s the cycle of the centralized versus decentralized.

MF: Of course.

CB: This is kind of a separate thought, it’s like…but I’ve been thinking that the we seem to be in this decentralized… this pushed, kind of like the Big Bang Theory of IT, and you know, will it all come crashing back in and we’ll be back to that path of where there’ll be some provider out there, Microsoft or some other OEM, that will try to pull everything together and be one platform solution that’ll serve all needs, and then see it explode out there again. Is that inevitable, or have we learned…

MF: The attempt is inevitable; the success is not. And I’m not even sure someone should be cheering for success for all of that. As much as there are definite patterns and similarities between the way eight different organizations handle customer relationship management, there’s enough differences between them that, yeah, we need to…there’s going to be work done of a decentralized nature. What I see is a constant balancing act between trying to distribute all the work out into the trenches and trying to consolidate some things for scale and for reliability. You can argue that the…look, this has happened before, it will happen again. We had everything happen in mainframes, and then we put everything in personal computers, and then we created networks, and it became client-server, and so on, which was a balance between the two. What you see now is trying to re-centralize everything up in common cloud services, and yeah, it works to some extent. But the problem is if people need a lot of different things and no one application, whether it’s software as a service, or anything, can do it. So yeah, I’d loved it…I like the idea of resources being consolidated or centralized in various cloud data centers, but the work is still going to be heavily distributed, and a lot of edge stuff is going to be distributed. If everything worked from a centralized system, we’d all be living our lives out of SAP.

CB: And what a life that would be.

MF: Well, SAP is really good at what they claim to be good at, or…you know…

CB: But I like the idea, though…

MF: That is an excellent piece of software. But I would want to use it for everything.

CB: But if…remember again, living through this experience where we have increasingly…all the data is residing essentially in the same location is…if you…the more and more that you try to consolidate…look there’s arguments both directions, but the more you try to consolidate it makes me think of single point of failure. One system goes down, it shuts down everything. But the reality is that our data doesn’t all sit in one place, it’s…

MF: No.

CB: …and however you define your data the application is…

MF: You need to service the needs of the real world, and in some cases I need data to be persisted even if it’s in the cloud in Canada, in Germany, in places that have data sovereignty laws I need to comply with different sets of laws and rules and generally accepted practices and so on. We’re not going to be able to do everything from one place, and to their credit Microsoft and Amazon and Google are not saying that – they have data centers all over the world that are, yeah, to some extent those data centers adapt to local conditions, to some extent, but I don’t see your return to everybody doing everything in their local data center, but I also don’t see everything happening in a common set of services up in the cloud. Among other things, different people have legitimately different ways of accomplishing goals. You might decide you want to do everything with WEBCON BPS, you might decide you want to do everything with K2, we’ve mentioned them before. We might want to do everything with the Power Platform and, along with a lot of extra work on our part, five different organizations will come to five different conclusions. We need to have room for all of them, and you’re not going to have room for all of them in one place from one vendor.

CB: Right. Well, something that you mentioned…you were talking about the sessions that you do, and maybe this is kind of a, you know, parting thoughts here, but where you’re giving advice to organizations and to end-users that are, um, like, setting up…establishing the right practices. I think immediately of a…like a community of experts within an organization to share best practices on how we should even be approaching developing our…and documenting and building out our business process solutions, whatever that is. It could be across different skills. So, what are some of those best practices? People love lists, Mike, as you know.

MF: Sure.

CB: What are the five things I need to be doing to prepare for successful…ongoing success?

MF: Okay, uh…sure, I can give you the easiest one: The number one is, if it’s a business process or even an automation solution, the process comes first. Most people start with the data. They model it independently of what they’re gonna do with it, and then they have to figure out to write some reactive logic for it. That’s crazy. When they build a form first without knowing what the form is supposed to be used for, that’s crazy. Start with your process. Your process will then tell you what data you need, and then the process and the data will tell you what you’re supposed to do with your forms. So do it in the right order. That’s a piece of advice from me. Second big thing: most of the time when I see people attempt to build workflow solutions, whether it’s simple automation or whether it’s a full-blown business process, there is the branching logic, in other words, do this then do that then depending on the following thing go this way go that way and so on. And then there are the individual steps being performed, like go get this information out of your ERP system, and then update someone’s calendar over here, and then check this a piece of equipment and record what the current levels are. These kinds of things. Most people I see try to deal with that will focus on those steps first rather than the overall logic. In other words, okay, how am I going to do some “Internet of Things” thing to read that meter automatically so people don’t have to do it. Or why don’t I….or how can I make a direct…which method calls do I need to make to which API in SAP to get at our HR module, or, like, should I be using the Graph API or the Exchange web services API in order to talk to somebody’s calendar? Don’t start with that. You should start with the overall branching logic, because if you make sure that the process is followed every time, with some sort of automated system, even if the steps are manual, no one’s gonna forget to do a step anymore. And you’re going to be able to figure out if you’re doing something inefficiently or wrong. In other words, if you focus on the process first and the automation is second, after you get the process right, you’re going to have a much better time, you’ll build a better solution. You can start using it right away, even before it’s completely finished. And you will just get better results. So instead of tackling at the automation level, tackle it at the process level, fill in the automation after the fact.

CB: I think it’s… it there’s a that’s a great point… I think that it’s something that when I would go in and build out PMO’s and work with project managers and business analysts and help them kind of build out centers of excellence, that same idea…in fact, Dan Home and I at ESPC in Copenhagen, the first time, so a few years back, we did a keynote where we talked about that…we said look, the technology doesn’t matter. That we were talking about SharePoint. SharePoint doesn’t matter. It’s the business that matters. Where you go in and first define what the business needs are, then…then you look at and say okay, this is what the business requires, where’s the data for those processes, and then the technology to be able to access the right data and meet those needs. Another benefit that comes out of it…doing it that way, is then you also…then as you go through that refining process of building that out and determine, hey, there’s inefficiencies in the process you might not otherwise see. If you…if you’re…you jump to the technology first, or the limitation the data says…

MF: Let’s come back to that in a second. The thing I wanted to say is if you automate the logic, even if you don’t have the steps automated yet. If you automate the logic you effectively have a living checklist. The aviation industry works off of checklists with manual steps. Surgical theaters now, based on a lot of laws that have been passed over the past decade or two, now make sure that before any surgery is performed you go through an elaborate checklist. Surgeons, by the way, hate this, but it turns out to save lives. It’s really cut down on a lot of, shall we say, surgical mishaps. Checklists are awesome. Checklists actually give you half of your business benefits right there before you automate a single step, so yeah, that would be piece number two. And the third piece…I don’t want to do this too…if we do all five it’s gonna take a while. Yeah, I’ll end with three.

CB: Sure.

MF: The third thing that you should keep in mind, and this one you don’t have to do this, but you should do it for selfish reasons, and this is what you were just mentioning that I asked a pause on. Let’s come back to it right now, which is: if you don’t examine what you’re doing now and measure it right now, after you model it, automate it, make it better, implement it, you need to be able to prove that you had an impact. And there’s no better way to prove it than to measure the old way versus the new way. Otherwise, if you did this at the beginning of the year, and you come up for a review at the end of the year, people are gonna wonder what have you done for me lately. And by the way, if you created a solution that made a problem go away, people tend to forget about the problem after the fact unless you’re constantly doing something new and novel. People don’t remember unless you have metrics to remind them of the difference you made.

CB: I believe that Janet Jackson was…was talking about IT when she said “What have you done for me lately?”

MF: Exactly.

CB: I’m pretty sure that was it.

MF: Dude…I read something last week about somebody thinking that Y2K was a hoax. Twenty years later, and then so…but again, people are talking about…and there’s also Y2020 now, because a lot of people just solved the Y2K problem by delaying it 20 years, so there have been some software systems breaking lately and that’s probably what got the topic raised again.

CB: All I’ll say is that if Y2K was a hoax, there were a lot of COBOL programmers that made a fortune off of that hoax…

MF: The hoax. And it’s not. Y2K was an example of people discovering a problem…probably later than they should have, but realizing that a problem existed and a widespread effort was taken to fix it. And in all but a couple of cases that were then fixed quickly, it worked right. But because there was no disaster, people thought it was a hoax. You don’t want that to happen to you. You want to be able to prove that the work you did made a huge difference, so measure it before and measure it after. Your application is still going to work, but you deserve credit for it.

CB: That is a great place to end on, and I know we can we can talk for an hour just about measurement and measuring success, defining that, and popular strategies around that, as well, and as someone who has…in a couple places that I worked I worked…intentionally worked myself out of a job solving the problems and…and then coming to the end of that and realizing there wasn’t a place for me going forward…I came, I did what I was supposed to go and do, and it was just time to then move, which is perfectly all right. I’m not an ongoing operations kind of guy, so it was okay. Well Mike, great discussion. I think those are some helpful hints for an organization that is starting up thinking about what is the best way to set ourselves up for long-term success, and in identifying and solving these kinds of business, you know, problems, that are meeting our business needs. If people want to find out more about you, get in touch with you, what are the best ways to reach you?

MF: MikeFitz@WEBCON.com if you want to email me directly. I’m on Twitter @MikeFitz, and LinkedIn.com/MikeFitzmaurice spelled out.

CB: Excellent! Well, thanks a lot for your time Mike.

MF: My pleasure. Thank you.

Christian Buckley

Christian is a Microsoft Regional Director and Office Apps & Services MVP, and the Founder & CEO of CollabTalk LLC, an independent research and technical marketing services firm based in Lehi, Utah, through which he provides fractional-CMO for partners in the Microsoft ecosystem.