EP90: Workflow starts with the TALK

The boys work through the different stages in creating a workflow that starts with The Talk. Ok after 2 rehearsals for Episode 90 Steve and Marijn get started talking about Japanese TV shows, the location for the 100th episode the whole conversation turns towards the processes of creating a Workforce... sorry workflow!

Hey Steve, 90 episodes.
I'm so old.
Is that what it is? It is 90 episodes you fed up with it.
Already, no, not at all.
You're auditioning for Macbeth.
I don't feel we had this discussion already. We had this conversation already somewhere like 10 minutes ago. When you don't.
Again, forgot to press the record button.
We must stop having whiskey before we start.
Oh, we had not enough whiskey.
We well that might be yeah.
Before we start the deer, one or the other, then you could do a really impression of one of the witches from Macbeth.
Like beds got which?
Is Mac Basket, which is I believe my best.
Oh cool.
Got which is yes, OK.
We actually went through this earlier, didn't we?
Not the first time we read it was we were trying to work out how very little we know about Macbeth and Shakespeare and all that kind.
Not not not which is now.
Exactly, yeah.
Of stuff.
But that's not.
Why you come here? To listen to you, come here to listen to two awesome consultants.
Standing at the bar talking crap.
Exactly, and we actually had a really nice, uh, review or not some some person said some really nice words.
We did, yes, a bit of a shout out for uh, of course. The name was just shot straight out in my head for Bradley.
Yes, Bradley my man. Oh he actually said thanks for the great feedback brother. You two make it easy to write with your fantastic podcast.
So Bradley actually writes a lot of the blogs that I create, and so we're sending the podcast and even come.
Also we
Pay him.
Oh OK, so now he doesn't feel that sincere anymore. No, no?
It is. It is very good 'cause he could. Doesn't have to say it at all, but he says I've done a lot of pod to blog work pods to blog work.
That's interesting, isn't it? Yeah, we're tonnes of clients and you guys are at the top by putting a lot of technical information and strategy into funny wit and humour. So Bradley, if you listen to this.
Since go yes, go.
Well mate, thank you very much for the compliment. It went down very well indeed, so yeah.
That's good.
Yeah, I'll put that one in my bucket, yes?
Put it away my friend. Yes put it away. Well, I mean the last three months we've had.
Over averaging over 1000 listeners a month wow.
So many people, so many hours wasted.
That's cool, isn't it?
It's nothing else to do.
No, no, thank you. Dear listeners for tuning in to our amazing podcast which I always enjoy doing every time.
Even just now when we were preparing for this episode, because yes, for the last 20 episodes we've been preparing our episodes.
And you even gave me some gold Nuggets before we even started.
That's true, I try my best to give gold Nuggets all the time or part of the fun. So actually this is take three, not just take two 'cause we actually started this podcast on Tuesday.
Last week
Oh yeah, true Choo Choo Choo Choo.
So those of you that actually listen to us regularly know that I'm already late recording this or mixing this down because it should normally been released over the weekend, but today's the Monday after the weekend.
Because Myrna and I went last Tuesday to a great whisky tasting, actually we tasted the wolfbane wolfborne. Yes, that little stream that feeds a wonderful whiskey couple of guys.
We did, yeah.
We had the story, the history, and we tasted seven of their whiskies in the end.
Yeah exactly yeah yeah.
Which was rather cool.
So so yes, so we started recording. Then the food arrived so we stopped recording. Then we ran out of time to finish it.
So then we decided to start all over again, and then I now just decided not to press the record button.
So this is really going to be special.
Tonight yeah we had so many rehearsals.
So many rehearsals, but going on from 90 it will be 2010 episodes.
Then episodes left, so that's more or less five months. So that means may, June, July, August, September, so.
To number 100.
Autumn time.
Yeah, that will be October.
Ish when we get to episode 100. So that would be in time, maybe for Ignite.
If it doesn't stay virtual.
If it doesn't stay virtual, but.
If not, we will find somewhere special to record our 100th episode. My missus is already convinced that I'm going to say we're going to Japan.
Would be a good time of year to go.
Bye bye.
Uh, yes. It's always a good time.
To go, I think, but yeah.
That's true, that's very metric.
Well, it also collabed days is also in October scallop days, Belgium.
So that might also be a good moment for us, although only like 15% of our listeners actually come from Belgium.
That's true.
Uhm, so well, we'll see.
None come from Japan, so I think we should make it a marketing tour.
Of Japan and record a bunch of podcasts.
Yes, I think that is a perfect good explanation for the tax man saying we need to go to Japan on tax man dollars because we need to grow our market share. Yes, absolutely.
Go on market share.
Being a businessman is so easy and fun, isn't it?
Talking about Japan, me and the missus have been enjoying the see the first season of old enough on Netflix.
Enough old enough, yes?
So you know in Japan they're all very how. How do you say that? They're all very nice people that want to contribute to family families viktring and they all want to contribute to family and that starts at a really young age. So old enough is a a kind of a reality programme.
Where they're following two year Olds, 3 year Olds, 4 year olds who are gonna go out for the first time on their own.
And do chores for Mummy and Daddy. So that means getting a 3 year old who needs to walk 1 1/2 kilometre to the supermarket by treating's. Go to another store to get something else and then walk back all on their own. It is beautiful.
It is terrifying. It is funny.
It is it's it's crazy. It is weird. It is it's wonderful.
So the members sounded like Macbeth at the beginning of this podcast is recommending that you watch this programme where 3 year olds risk getting killed, getting their mothers.
Yes, shopping from the supermarket.
Yes it is. It is. It is just crazy.
Then Netflix have a lot.
To answer for no no so.
This has been running for years in Japan and it is.
It is super popular there and now just Netflix bought the rights to put it on on there, yeah?
Old enough, all right?
I watch at least one episode. We'll see where.
Yeah, do you want to talk something Microsoft 3/6?
Yes, that's yes. Yes yes yes.
Five or Turkish?
You don't want to talk about the rather wonderful whisky. We're going to be tasting later.
Me me.
No, let's not we'll tease people with it later.
All right so.
I came up with this weird idea by telephone the other day when.
I said to you, hey, let's talk about workforce.
Yes, and then I.
You actually.
Came up you said that you texted me this morning saying let's build the workforce, yes.
Let's build a workforce.
How to build a workforce before you. Before you build a workforce talking about all the stages, yes.
I I actually had that.
Talk when I was around 10 years old.
I had a really nice biology teacher telling me how you build the new workforce. So yeah. So I I thought OK.
OK nursing.
How nice was the biology teacher so?
Maybe maybe we need to.
How nice was the biology teacher?
It was actually a.
That's actually it. Will save that one so later.
She was actually a nun.
Yes, so even then I kinda had my reservations about the quality of the content, but still.
Alright anyway, yes, so my autofilling spelling software on my Android phone decided to convert workflow to workforce, which confused marine.
For quite a while.
Especially as I started listing all the things that.
We could talk about what we want to.
Lifecycle management? Yes, of course of course.
Think your life cycle intentions.
When they're old, they need to get a pension. Yes, OK anyway, but you wanted to talk about how to build a workflow.
Our workflow was the way forward.
Yeah, yes.
So I thought it would.
Be a good idea to start bringing some of the skills that we have.
We did all the change work. We talk about baseline governance and governance itself generally and of course before you start typing.
In my code or no code or making some of those selections, if you're going to make a workflow work, there's a lot of things that you need to get right first.
I thought it would make her an interesting conversation to do that, and as we're already 10 minutes in and we haven't started yet, but.
That's quite normal.
It was funny, no, it was funny.
It was yes.
And if we include the 20 minutes we've already lost because I didn't pressure the cord, then we've got a long night ahead of us.
Anyway, so.
We our original idea was to kind of come up with a bunch of questions like what do you say to the user that says I want to build a workflow and you say Are you sure it's a workflow, not a workforce or it is a workflow, so that's good all right and and so I want to start off with with the talk. OK, because you know, it's amazing that.
These people, well, actually what most people have done is they've already started working building the workflow and then finds it doesn't work. Yes because you know, they just read their YouTube.
They watch a YouTube video and read a few blogs and go. This is easy just to say that Microsoft just the way that Microsoft sell it.
Yeah, it's no code. I don't even need to.
Developer, let's just start clicking around, yeah.
Absolutely, I wonder why Microsoft had not actually been sued for that statement. You know, no code development. I mean in its own right. It's kind of.
Wrong, isn't it? You know, no code, but it is development. Yeah, that which is completely.
But it's in.
The same thing as building houses, also development so.
That's true too. Yeah, it's strange. Anyway, the talk.
So one of the first things that I do very much is kind of make sure that I've got a little hour, nothing more than that. Doesn't need to be a big thing. I'm not going to take down all the triggers and.
Outputs and inputs at this point of the the stage no, absolutely not I. If you're running agile of course, requirements are always interesting and there are to be fair to the user or to be unfair to the user.
You're not gonna go deep in the requirements, yeah?
60% of the people that come to you and say I need a workflow don't really need a workflow, they just think they need a workflow because really what they need is just to be able to see their documents in a different kind of way.
You know they need to be able to say, hey look when I fill this document out. Johnny then needs to kind of.
Add this piece onto that document.
Then so yeah, but I can do that with the library. We just change the status in the view and you know.
So it's not. It's a flow.
Of information, but.
Run a notification here and there sometimes yes.
Yeah, so that's really kind of no code. But anyway, generally if I was sitting there having that conversation, the word baseline comes into everything here, 'cause everything from this initial talk.
Could be scrapped, or it could actually be really important to the process. You never know this yet because you haven't had.
The thought process you haven't had the moment when you sit down and you try to review what you've been told by the user.
So in that first talk, with the using that first hour, I normally like to try and get a number of things. Does this work for you the same?
Way when people say, hey.
Yeah, yeah yeah, because you first need to like scope it like how much work is this going to be? Is this going to be a huge thing is?
Marine come and build the workflow.
Is it? And if I need to build it, am I going to go over my head because I'm not a power, automate expert or so?
Is this something that we need to get a team for or is this something that I can build for you in in an hour?
Let's just sit together. I'll open it up, but like I do with SharePoint and teams all the time, I'll just open up a demo site.
And let's build it together and see what is what. So how far is this going to go? Yeah.
OK, So what what I like to think about is certainly the scope. So what is it defined to deliver? You know, this workflow will ensure that.
That that this innovation idea as passed by the appropriate gates to decide whether or not the company will move it forward as a valid idea as a top of my head again.
So the first thing that I often think about is that kind of business case, so just a baseline business case of, you know, identify. OK, so it's.
Going to evolve.
Five or six people, so the document has got to move five or six through my five or six stages.
Are they all on the same site? Do I need to build a site to run the workflow because I actually want it to run all right, you know?
Oh yeah, that's a good one, yeah?
So these these are just the top top top of my head's. Does it actually need to run on a phone?
Does it actually you know what? What are the requirements? Just get an idea of a baseline business case for it.
Is it going to save money or is it going to be, you know, cost more money?
Uhm, what kind of tools might I use? Is it an Excel spreadsheet, you know, is it just? Is it a list, is it?
You know, is it actually power apps? Is it logic apps? Is it, you know, just a rough ride idea before I move forward and and of course then the the dollar sign.
Yeah, because you kind of need to be able to go 10 days, 15 days, you know yeah, because at some point.
You've just had a conversation with the end user and the end user thinks that everything is.
Easy, it's just a.
Workflow unless.
Just an approval for five people. Yeah, cool.
I saw I saw.
This YouTube video and they were explaining it in 20 minutes. That's true. Why would you need three days to do it?
Yeah, no exactly.
Talk for me is is really kind of important and don't tell my users, but then I will forget to get back to them for a.
Just to see if they really, really want it yeah or or if they couldn't figure it out watching that YouTube video by themselves, yes.
You see, you've been there.
No, but it's it's true.
Give them time to think about it and I usually give them OK. Who's is your boss into this?
Has he got buy into this as your bosses boss got buy into this? It'll be good if next time we meet we get some idea that you know they understand where's the money coming from. So the sales process, you know.
Yes, I'd like to buy these seven.
Widgets, that's cool.
Do you have 5 grand in your budget now? Yeah, yeah, we're all covered.
OK, so who's approving that for? You know, there's always in the sales process. You make sure the money is there before you move forward. In this case, it's a similar. A similar kind of thing.
So the talk for me is always kind of the first thing, at which point I then put a one pager together and write back and say look this is this is what it's going to scope. This scope was supposed to deliver and then I might put that means.
This and this and this and this, you know, uh, budget. It means that we need to involve these people.
Yes, yeah.
It means that you know.
If we need to talk to a separate system, it means that we need to interface with that system.
Exactly, those are the kinds of things you're trying to highlight, and you won't get them all because the story of the second time round will be different.
Yes, when you talk to them next time 'cause they will have forgot stuff that is important to you, but they kind of.
Yep, been there. Yes absolutely absolutely.
Yeah, it's amazing, isn't it? You can always tell the consultants that have been around this game because they.
It's not that they know the right questions to ask, they just there's a feeling isn't there when you go.
Yeah, I think I've got everything out of your head at this point in time. Now that you you know about you start to get questions. Start to ask questions and they go.
Kind of glistened.
Eyes and you go. OK, yeah I need to go away and think.
About this
Yeah, exactly.
So yeah, so that's that's the first stage for me, is always to have that talk. Get it, do a brief document together, start the scope off, understand where the money is come.
And then you may wanna give that one pager you. You might want to have that signed off by the person that's actually gonna pay for it.
Or you might want to run it with your team, or you might want to run it with your. I don't know other decision makers like.
Yeah, there's usually a.
Project model around somewhere isn't there and you know your own workload needs.
To be assessed.
If you're if you're an organisation that is kind of running agile type stuff, it's not going to go anywhere.
It's not that you're going to drop everything to to work on it, it needs to go into the backlog to be listed. And if you're in a waterfall situation, or if you're in a.
PM Oh kind of organisation.
Or you know you've got your operational stuff to do, and all those kinds of things, so it needs to be prioritised at some point in time.
Maybe there's some business analysts involved, or maybe there's a you know.
It it's it.
Really is that now? Is that thinking kind of process to kind of just get everything everything down and?
Yeah, like what? What do we want to have and what's going to be the rough estimate of scope and money? And yeah, exactly so business case first.
That's kind of stuff.
Step absolutely great now then uhm, we we wrote down next to it. A few really good things that you also need to take into consideration if you're going to go forward.
And that's pretty cool.
With this, yeah, I think I think those items.
Up on there around if it's one particular kind of workflow, but it all depends on what the workflow scope is, so if it's if it's a.
Workflow that is.
Going to move internal content around. It's a it's a business workflow. Then it has a certain amount of scope.
If it's a workflow that's going to be sent out to 40,000 customers or 4000 customers or or even 200 customers, then you need to take a different approach to that workflow. If it's just moving a document.
Brand internally between 5:00 or six department.
Since then, the kinds of questions you ask are a lot different, and the ownership is different than if you're going to build a marketing app for the marketing team to, you know, sort of sell products through or to build a community of interest or or whatever they're doing. So there will be.
Different kinds of questions to ask, so let's. Let's assume it's a big.
Yeah, and yeah exactly so.
App for now.
The first big question around that is indeed like who's going to be the owner of the app. Is it going to be it that needs to own it that needs to be delivering support later on if something goes wrong? If Microsoft decides to change Power app or to change?
A certain step that they need to handle all those changes, or is it going to be a business department? Is it going to be a local IT?
Yeah, yeah, so that's that's definitely a good question.
Police have one person as the owner I think, or a role in the company. So if in my mind if it's that marketing app where they're building a community of interest for the business, then it's the marketing director that owns us.
And and even if he's not paying for it, he needs to understand that he will be the person that has to.
To decide it's direction and approved changes and updates and and funding that you know there's. There's always a long term story to any kind of workflow or tool that you build.
Yeah, so so the owner is really important. So you you're right, if it's going to be a corporate app then it generally own it or somebody.
Within it, the software director or software for software manager certainly does that, because there are all kinds of different risks associated with that kind of.
Kind of tool, yeah, and they need to be managed in in a lot different way, yeah?
Yeah, absolutely yeah.
But if our business needs change, what if the market changes? What if Microsoft changes the product? And as you said, what about new features that people want to get added? So it's that one person that one.
Role that needs to decide on these things.
Somebody certainly needs to own it. It can't be owned by a team or it's, well, I suppose it could be, but at the end of the day, they'll be a team manager, so ultimate responsibility lies there, so the risk associated with these kinds of things is.
You know if if I have an app out there with 400 of my customers and the app fails, then there's a whole sort of hey this company crap they can't even run their own apps.
They can't even do their own thing if the app never gets updated, there's a perception that your customers, so those kinds of things need to be done. And of course there's consultants. You need to make sure that.
The client understands this, you know. Yeah, you know I. The amount of Times Now I say the amount of times at least half a dozen times I've had.
Yeah yeah I want to be able to get this information out to our customers. So the easiest way is to I can either keep emailing it or I can put an app together and they can log on and do it.
OK, well the email is kind of under your control. It's simple, it works every time, so that's fine. But if it's an app, it's a great idea.
It allows us to have a lot of you know it's a nice baseline to build from in terms of building up, but you also need a communications plan. You need to manage that, and you mentioned earlier about.
Compliance on these kinds of stuff. There's a big question.
About if your.
Bank if you're in healthcare. If you're in talking with external parties, if you're communicating with customers or or whatever.
Don't Sarah.
OK, so the dog just switched positions and it doesn't look very. He looks very comfortable but it doesn't look like a comfortable position.
Looks like had his head chopped off, doesn't she? She looks like she'd be beheaded as a guillotine.
No part of him.
So, but actually, no, it's what you said is is brilliant, like if do you want your customers or suppliers to go to your app in order to know something. I was walking through the supermarket.
Clear and I was walking next to the bus stop and on the bus stop there was this sign saying if you want to know when your bus is coming or if you want to know if there's any delays or Rd works. Log into our app and I was thinking what a piece of crap.
Why wouldn't it be cool if there would just be an app like an augmented reality app where I would just be walking on the street and I would just see all kinds of information from all these services instead of logging onto the bus application app, logging onto the restaurant app, logging onto the whatever app to just?
Have it all come to me and just decide if I want to do anything with it or not. It's called Google, it's true, it's true.
It's another another idea that Marine finally hit on 30 years too late.
But it is a billion dollar idea.
It's a pretty bad.
Idea that is very, very true, but.
But that kind of stuff certainly needs to be considered. Yet the audience, the support, the support.
Audience form pliancy
Actually, of course.
You know you need to make sure that when somebody reports in a when I look at those apps on Google and I go look at some of the responses.
You know, I emailed them off and.
Nobody come.
Back to me, yeah, so yeah, that's that's neat. So yeah, there's one kind of story for one kind of app and that's why that initial talk is just about getting the baseline in.
Yes yeah, and and trust me when they say yeah we want it for 40,000 people. You've got to stop yourself getting down the rabbit hole of with this customer about what this app is going to do at this point in time that first.
The conversation is not. Let's talk about the other kind of app.
No, yeah.
Which is, you know, an internal process, a kind of real workflow. This is where you have to try and read my.
Handwriting, yes.
So I complain about yours, but yours is very clear. It's sexy. You know that you've got sexy handwriting.
Tell that to my teacher in in second grade who said my handwriting was terrible, but anyway.
My handwriting was so bad that the art teacher had gave me special lessons at lunchtime on how to write.
Brought neat handwriting when I was about 12 years old. OK, Wignall Miss Wignall Miss Wignall the art teacher, OK?
OK OK OK so so yeah.
Right kind of internal process so and and the thing about this is that there are a few big kind of apps and big kind of workflows and processes, but there's lots of smaller internal beneficial processors so you know some of the questions I ask.
When it gets to this kind of thing is is lifetime cost.
It's so important to understand whether this thing is going to be around for a year or 10 years because there's a big difference on how you build it and how you put the foundations in for it.
Yeah, exactly.
And you know the support of it and all that.
Kind of stuff, but also your ROI. It's also differently, yeah?
Yep, so I mean, if you've got, uh, some a lifetime cost like a.
Hey we we want to build a workflow that will do annual leave for the organisation as something that's going to be there forever. Yeah, you know it's so let's invest.
A good app here that will allow us to and do this and do that because we know this is not going to get changed for forever.
And then there are all kinds of other implications on those kinds of longer term things like legacy and whether it's you know what you're going to build it on and will that.
Platform be there forever. Yeah you know otherwise you if you build it in power platform. For example, you know when there's a few upgrades over the next few years you could end up actually having quite a lot of costs associated for the lifetime of that app.
So yeah, something that needs to be there. The other thing I always ask is OK.
So we're going.
To build this app, what do you do now? How do you do this today?
And therefore why do you want to change?
It yeah, So what? What's going to be the yeah, return the value.
The value yeah versus how you do it now and and how you want to do it in the future, and that might be a cost saving. That might be time saving better.
Might also be around compliancy and and and other kinds of reasons, yeah.
He could easily, but you know, the question is kind of.
You know what are we going to remove then? So we build this and we invest some time and money.
So what are we taking out? Because every time I have a new app I have new backup, new support, new tickets that I need to do.
There's so many things that you need to do. It's a bit like rolling at Microsoft 365. Yes, you need a ticket to report your Microsoft Teams.
Problem and yes, you need a ticket to report your one drive problem and oh, you gotta share point. You need to share point ticket to to report that so you know I roll out a new service.
Yes cheque cheque cheque done oh just a minute I need to make sure the helpdesk is geared up and I need to make sure the tickets are available to report it.
All that kind of stuff, and it's exactly the same with the kind of workflow and and so yes. And then thinking about the night of support to do.
You also said something something else earlier, like.
If we're going to build a leaf app for the whole organisation, do we actually want to build that app or do we want to buy something that's already available on them?
Market well at this.
Point in time. We're still collecting data, aren't we? We're not really making decisions. Well that yes, no, I think I'm right.
So our initial talk that gets us into a kind of process that's hey, this is a big gap in it.
Therefore, we need to have a big app conversation or it's an internal app, so we need to have an internal conversation.
We need to have the big boy conversation. Yes, the big pants conversation, yeah?
That's true, but but it is very valid, isn't it? That you know we're still talking, we're still collecting data at this point in time.
We're just trying to work it out and one of them, as you say, is that as you go through this process, the decision at some point is whether you build it or whether you buy it and that there are pluses and minuses to both.
Yeah I, I'm working on a project at the moment and trying to change people into Microsoft 365.
Oh yeah.
Therefore you need to.
Do Office 365 on the.
Messed up and then all of a sudden a bunch of the add ONS don't work on the Office 365 anymore.
So then I have to create Office 2016 for some people that need to use this app because the supplier stopped supporting it eight years ago and all that kind of stuff. So you don't really want to buy something that's only got a limited amount of time.
If you are going to buy it then you need to buy it from a kind of big professional company that will look after it.
If you're going to build it, you need to accept that you.
Know yeah, but for example, I've got a smaller customer and 60 people and they said OK we need some kind of help desk application.
So are we now just going to create some kind of shared mailbox for the three people that do it work, or are we going to go with a paid application? Or maybe a free tier of a yeah? Something like that so.
Also, this now I mean, depending on what the money want to spend, I mean for three people I think service now would be a perfect purchase.
You know it will never go wrong. It's always supported, always accessible work web. It's rock solid for three people.
Oh, you mean it's going to cost them €2000 per person per?
Month ah.
I might not be right.
All right, so let's just recap where we're at. So if we we've somebody wants to build a workflow or they worked out what it is we know whether it's going to fit into one category or another category based upon things like number of users, internal or external users.
You know the kind of support who's going to own it? Who's not going to own it? That kind of stuff.
So we get to the point now where we're thinking all right, we need to move this forward. It's all been signed off.
You've got kind of approval. You're now not dealing with one person anymore. Of course, there's a small team putting this together for any of them. It doesn't really matter, so you know, we have to think about our ad car stuff.
Yeah, yes, as you do need to think about the change process associated with both of them and they were bigger or small.
We are.
We're depending on what needs to be done, which is cool so.
I guess we're.
Probably at the point where we we think about what platform we're going to build this on.
Unless we're buying it, of course, but assuming that we're building.
This then we need to kind of think about that kind of stuff.
Uhm, that's true. Is pointing something out very subtly? Because this.
Don't yeah yeah yeah my.
We're we're now going straight into solution mode, but actually what we first need to do.
Is and that's actually where I go wrong often, and where you always correct me, so I'm really happy that I can do that. For once is actually do a deep dive on the.
Ray wins, yeah.
You who I do a deep dive on on the requirements so really get those requirements straight. And if we're doing it in agile as you always say, stories, features.
Yeah yeah, make sure that you get those.
Yeah, yeah, but that's true and it's going to be the same for for either of them.
So if you're going to do it something small, if you're going to do something big, it doesn't matter. There's just more of them to deal with, and most of it will be based upon risk. So if it's an internal application.
OK, we might lose a bit of you know, respect if the thing collapses after two days, but at least it's not going to necessarily.
You know, take cost the company, customers, et cetera. Et cetera. Yep, but yes, you're right, we need to think about requirements. We need to think about user store.
There is and we need to get to the point where we can.
Could work out what features are going to be part of this workflow or application. And then there's all the kind of ancillary stuff around it. Processes like test plans and you know quality control documentation.
Documentation processors for go no go. How often does it need to be updated? Who's actually going to own it and decide RIT taking this on as a corporate app? We already talked about that there will be.
A set of processors for that.
Exactly, yeah.
And at this point we.
Decide to move on now.
Yes, I see there's some interesting points that.
I kind of think about here, but I kind of think about them myself and not really share them because depending.
How nice of you.
No. Well it it's kind of really a a personal kind of book cheque. You know, I want to kind of make sure that I'm not going to lose face on this at the end of the day if I'm running this project, even the team that's running it needs to know that we can do this. So what's going to show you that this organisation?
Is capable.
So it's worth at this point looking at what apps they already support and how well they support them. Do they keep going down?
Do they stay up? Do they get changed regularly? Have a look at the change process. Is it a process that they can regularly updates and they do them satisfactory and those processors run?
If they don't.
Quite honestly, they may not, you know Azure then then maybe it's you. Think about outsourcing. Maybe you say look OK.
Yeah, yeah.
Marketing department need this app, but it needs to be outside and it needs to be up all the time. But if I look at the way that.
Our processes need to work on and that how difficult it is to get things through change control, and dah Dah dah dah dah.
Then as a consultant, I would consider those options and go. You know guys, it might be better to just outsource the app.
I know there are compliance E and audience and all that kind of stuff deals so you end up bringing legal in and things.
But the end of the day you're trying to create an app that is stable, delivering the scope easily.
Updatable, you know, et cetera.
And you might have some really good contracts with some nearshore or offshore offshore development company.
That's really all about building these kind of applications.
Yep, so you need to kind of look at the best way to move on to those next steps.
There's no doubt about that, but yeah, so I quietly kind of just do a little assessment of the the risks associated with the decisions that you're about to make. And at the end of the day, you're going to have to create a.
Document and that document is going to include all the information you collected from the first chat. The information that you've kind of the processors, the the refinement and requirements, and all that kind of stuff. So yeah.
So that you can tick off some boxes and say, OK, we've we're done. Yeah, we're done at some point.
So what have we missed?
Are we now ready to rock and?
Roll and ready to build.
Uh, well we do, as you said, need to decide where are we going to build?
This correct what what's the best environment for it?
Yeah, exactly so. On the one hand, we've got power automate, which is your end user productivity automation, and we've got logic gaps.
Have you actually ever met an end user that can actually build a workflow in power automate?
Ah yes, I've seen a few, yeah.
Working for it. I know finance then no, no. Not at all, no. I've I've had a few engineering guys and I had some marketing people as well. No OK, no that's cool. Yeah so power apps, power automate role is one or power apps.
Is that choice? We've got logic apps.
And we've got logic apps, which is your power automate running on Azure?
Uh, should be your more.
A system, uhm?
And what I had, uh, or what I heard from a few organisations. What I told you earlier before we started recording as well plus.
Wow, Judge, you never take that other.
Knife out my shoulders.
No no no no no.
No, no it was way before we start recording.
Actually, when we were drawing up everything on the board and how some organisations approach this, and this might also help some of our listeners when they are trying to build some kind of.
Building our notes.
Governance for this is that a lot of companies are using power out.
Made for end user work, so they say OK their end users. You can do your know code, local personal productivity stuff and you can do that on power automate so those would be workflows owned by you.
Maybe not supported by it may be supported by T. That's what you can decide for yourself for your organisation, but that's power automate. That will be the stuff that your end users.
Build if it's going to be run by it. If it is going to be the owner we're going to build it in logic gaps.
We will use some service accounts principle accounts for that, but that way we know it's running logic gaps, we can do all the maintenance. We can do all the.
Reporting on it and stuff like that. So that's how a bunch of organisations do it. That's how they make the distinction logic apps versus power automate.
Yeah, I don't really agree.
With that I know you don't.
You're not. You're not going to build a.
Document based workflow in logic apps.
Why not?
'cause it's kind of an overkill when it literally is sledgehammer and walnut, you know.
I don't know to be honest. I have never built something in logic apps, so I don't know what it can or can't do.
And that alone is the the justification for not building simple workflows in logic apps in that you need specialist people to be able to build it so.
Now that's true, no?
You know that that internal process we talked about earlier in our app. I think you build the platform that that suits all of the requirements and your governance has to be flexible enough to make sure that you guess the best value in the best investment logic apps will cost more money to build.
Probably probably, but logic apps will be your more heavy, mature workflows. While power automate will be your notify me or do a.
Different licencing approach.
Simple approval so.
Maybe, but the point is that if you have a governance says these apps were going to be developed in this way and these apps are going to develop in that way, then you need to have a very, very clear idea of the kind of workflow that you're going to use in the organisation. And you're also limiting. Therefore the.
Capability of what they can do because nobody will want to invest in building a logic app.
It it's.
How can I explain this? I I think there's some logic in what you're saying. First of all, I think you're right.
I think that no.
No, no, I I I think that.
You can put a.
Clear distinctive line that says, look anything that is going to run against a personal account on a SharePoint site. OK with users using power automate.
You need to manage it and support it yourself.
To an extent but.
Not many people can build their own power app. I said to you, I said to you earlier. Do you know anybody that's build it and you went a few, whereas there are a lot of processes that need to be done at that local end that do need to be supported by it. That will make sense to build them in power automate.
Quite honestly, yeah, so that question that we need to kind of pull together is from that initial talk. That baseline is where we need to provide the evidences that support our baseline decisions, and if they don't, then we we have to change that baseline decision. So we will expand it into whatever the requirements and needs are.
From the business, not technical, so I think I agree with you in essence that this would great dividing line.
You know logic apps is great for corporate applications, but not for six people. It needs to be a corporate application for large numbers and all.
IT need to then take on that process and support it and do the upgrades and manage the process and the business.
Need to understand that their IT budget is just going to go up because it's being supported and monitored by it, so there will be implications. I'm I'm actually reading for about the 4th time.
Now, because it's just such a great book.
Hitchhiker's guide no.
Oh, and of course it's the bloody title has just gone straight out my head as I talk about it.
So it's a business book or OK.
It is it. Yeah I there's only one business book that I read on a regular basis and I I can find it here on my reading app. The Phoenix project.
I don't know whether you've ever read it or not, but it is well worth worth a read. It basically talks about.
Tell me all about it.
How an IT operations IT department does DevOps and it starts off with this guy taking over the role of the director 'cause he got fired?
Is that one of those romance business books? There are no no romance, but Rome. How you call it?
Does that really know scribit?
Do role model these I don't.
Know what you could.
No, no actually written in a in in a in as a story exactly.
It's story.
Yeah, but it it's a very good and of course like all of the best ones, it's 25 years old.
Or something 20 years old.
But it it just reminds me of the game we play, you know. And there's this kind of weirdo guy dressed in a T shirt that the the the.
The hero of our story.
Mistakes for the guy delivering the Donuts but but he basically says that any of these processes and what we're talking about here is just the same as if you're building a car or it's the process for the organisation.
You get stuff coming in and you get stuff going out, and it's the bit in the middle that's important and it's the same thing.
So when we have that talk, we talk about that what's coming in.
And then we kind of got our scope and we're trying to work out how to get to the end of it.
But it's a great book if you haven't read it for this project. It will make you think whoever you are, whatever you do about your IT services and it's just funny.
I mean, you know they talk about businesses, say yes, we're going to launch it two weeks time because because they said so they haven't even got the VMS up and running. They've not done.
Any testing they've not gone, and you know it's going to fail and you're starting to feel for the guy that he gets landed on you.
Know he there's a guy who goes yeah, but I've got these 75 projects and we only have 60 people in the organisation and the the management just go. I don't want to hear it. Just get on and fix it.
Yeah, yeah and so.
That sounds pretty accurate actually.
Yeah it is.
And of course.
The story starts to pan out. Eventually you get the whole dev OPS story and he learns about different kinds of work.
Highly recommend it, it's good fun. I don't read business books. They're boring as crap, but this one. It's one of the few that's worth a.
Funny OK.
But I was thinking the same thing, so when so when you're actually looking at where these things are going to go, you want it to succeed, you know.
And if we take the old ad car stuff, you know the awareness and the desire to succeed is important and the knowledge is even more important because you've got to.
Make sure that.
When you have a tool you choose.
Whatever environment you choose to build it on actually needs to be sorted out. It's that time already.
We got a stranger in the house.
Have we? Yes, OK, never goes running around there. OK, cool, cool.
So yes, I think it's. I think it's an interesting process.
I quite enjoyed these steps. I enjoy them when I do them for real. That's the real fun for me.
I mean, it's you know yes when we build this SharePoint site we want to be able to do this and automate that and automate this and and make that work.
Of course, yeah, because just the document repository by itself is 9 times out of 10, not your end goal, no?
Now and of course, we talked about content types about 10 episodes ago. You know that's all part of the foundation that allows you to build those kinds of things.
And you talked about governance for power apps and logic apps and all those kinds of things. You're right, you need that kind of governance and those decisions in place so that you can kind of at least fit this in there and somehow and make sure that it's going to be managed. But but but I absolutely guarantee that.
Even if everything we've talked about.
Today is put into place and managed and coordinated and everything else. You will still end up with people building their own workflows that you don't know about, that you will end up having to fix at some stage during some migration or upgrade. You know your upgrade power apps and you put in some new features.
Microsoft depreciate something you know, and then all of a sudden you get a phone call to say.
I I don't get these printed documents anymore or I don't receive approvals for that funding anymore.
And of course, you've got no idea what's happened, so it's an interesting process.
Yeah, yes, that's true.
You can of course track all of these things.
You can actually see with power apps. There's a lot of good reporting tools, so your governance potentially would turn out and say something like maybe.
On a monthly basis, we will assess the workflows that are working. Any workflow that goes over this kind of run number, say 25 times a month.
Then that workflow will have to be registered with the IT department if you want to maintain some support.
If it does so many runs, or if it has so many failures every month, then we'll need to take a.
Look at it.
Yep, so maybe there are some tools out there around this process that you can make some use of that allows you at least to monitor them and to to be able to say to the business that even though we allow you to do.
Power apps.
'cause you know you can't really turn it off for anybody.
Getting you no.
It just doesn't turn off. Yeah, it's there.
There are ways of controlling it in your MS Teams, and you know taking away from there by not adding the app and stuff like this, but Microsoft want you to use your the the amount of usages and things you buy.
And also I think for the the next few years I think that no code low code is going to be.
Uh, pretty important topic.
I, I think we're I mean, I think I'm definitely too old to be of a of a generation where I would even think about doing that.
But you know, if you look at students now and even you as like grey out, but you know you talk about your 20 somethings, they're they're smart enough and.
Even me.
Clever enough to be able to make the most out.
Of these things and make them work.
Yeah, but also everybody is now used to working in teams for example, and they might want to see what would be the next step.
Not all.
I agree with you, but do you do?
That before lists.
Or do you do that before? Or is that a different podcast? It is yes.
It's a different podcast. That's a different episode, yes.
But it it is true, I think that.
Yeah, I like it. It's a nice process. It's not too complicated. It's five or six levels starting off with that initial baseline talk, making sure that when you get to the point where there's no governance in place to help you, you go and write the governance before you move on because you will be there again.
And and, uh, yeah, categorise it, work out what the best way forward is, make sure you've got an owner.
Understand the lifecycle of the workflow and the the app and then get on and.
*** building and the overmanned it over to the project.
Manager Yep.
So where are we on time?
We were at a perfect time to actually take a breath and drink a whiskey.
Let's do that.
Let's do that.
Are you talking or am I talking and you're pouring?
Uhm, why don't you introduce what we're drinking and I will pour.
Do that.
We are drinking the whiskey from my Barber. Yes, let's do that one alright. It's a very red whiskey because it's finished in a bird Bordeaux red wine cap.
It is from our very very special distillery, D instance. We've been here before.
This is the distillery that actually is on the side of the river, and it uses the river water obviously to create the whiskey, but it also uses the water to generate the electricity so it's self support.
Thing and I yeah you just keep readjusting it my friend.
But so the distance whiskey we like, we've we've got the virgin oak is the virgin oak. We tasted a few weeks ago, so that is their standard task from the virgin oak.
But this one is a rather special one limited edition.
We also taste the 12 year old.
Yes we did. Yep, none of them are.
Bad and we always.
Come to the conclusion that this is a drinking man.
's whiskey, yes.
It's kind of sits there, it gets drunk it.
To you know.
Yeah, yeah, absolutely.
But this is unusual for them. I think this is unusual for them, but they have some very weird whiskies. I mean, they're centenary. You can only get from the distilleries made from a whisky from four different.
10 year old periods. So bit of 70s bit of eight is bitter. 90s bit of 2010. Yeah yeah, sort of a rather rather cool Vega.
Really OK.
So here we go. So this is the Dean Stones, 9 year old Bordeaux Red Cask matured so.
So it so it's not just red wine, it is Bordeaux cask. It's a Bordeaux red wine. Yeah, so it's actually French.
Nice OK yes, so the yeah absolutely the colour is the first thing you immediately noticed.
You've not stuck it under your nose yet. I've just done that. You're going to go.
It's just this bit. I only had a quick.
To be honest, I really like red wine casks. I I have a few at home, not red wine Cask, but whiskies that were matured there with Minecraft and I have to say I really.
Finished in Red Wings.
Like the the the complexity that they bring, yeah.
I like the nose of them, they're brilliant. We did an Irish one once that had been done in a French wine cask that was superb. Yeah, absolutely couldn't get over the bottle of that. I've tried to find one and it went.
Yes, yeah.
I can also remember that Teeling that we have the Peter Teeling that was finished in the southern cast that was then white wine. But still yeah my God.
Yeah, more sweeter. Well, this is red and you can certainly find it in the colour. It's got an elegance and I had to describe that elegant amber.
Hmm yes.
It's we're just working out what pets are going to go in or out or out and in.
None of them. They're thinking there might be some food here.
Don't worry, OK.
Yeah cool cool.
All right, yeah knows. Let me tell you what the official line says, or at least from wish cucheron scholtens Cook.
Cooking fruits, warm apples, red fruits, spicy wood and cracked pepper on the nose.
OK, with our sepehri spicy nose then.
The pepper is good. 'cause it does take your breath away.
I don't have a lot of spice on there.
Well, it's it's not that Christmassy spice but it, but it has got a bit harshness on the nose. I think that's what they're probably referring to.
No no no.
And there is a bit of that stewed fruit.
It's not overly sweet, though. It is quite nice.
No, no, it's not sweet, but it's there's something stewed in there, something with a little of honey.
It's more, yeah, not. Caramelise is not the right word but something that.
I think it's crooked apples. It's stewed apples.
Yeah, might be something like that, yeah?
But it's a nice nose. It's the kind of knows that would stop me drinking.
It for for a while.
And just just enjoying it, yeah?
There's a smell in.
Umm harshness is there I'm I'm gonna put some water in this one. I'm gonna drink half and put a water in 'cause I think yeah actually you know we normally said it would take the edge away. Might actually bring out a better nose.
It might be. It might bring out that spies in the nose. Yeah, we've sought, we've seen that. We smelled that a few times already, yeah?
You're going in.
Oh, I see your eyes lit.
Up no, but not in a.
Good way oh Oh well.
First taste is harsh. It's not spicy.
It's it's got a nice.
Finish nice chilli kind of finish but.
But that first taste is a bit harsh. I think the second taste in a minute might better. It describes it as warm and full, dry and fruiting.
A fair amount of red currants.
I get the warm and the fruity part, but I get tonnes of spice.
I I find.
On that first, yeah.
Yeah, it's very. It is very dry though.
Like a good Bordeaux.
I I was just thinking the same yeah yeah.
OK, it's it's not bad.
I definitely like.
It it's at instance.
Yes, but it is. It's it does have a lot of of that spice in there with a little a little bit of salty spice as well.
Second taste is better.
Once the shocks over and done with.
Finish is quite nice. It's a medium finish.
Still, I have a lot of that salty.
Salty, definitely think it's probably the saltiest whiskey of.
The spice, yeah, and that's what probably takes out that the the moisture in your mouth.
That leaves you with that dry taste is I'm. I'm guessing it's a lot of that salty flavour.
Is a really good question. Would you buy a bottle?
I'm not sure I.
Would I'm not sure.
It actually reminds me of my my new bottle of Bona Haven that I have, which is also very salty.
Thank you, which is also very salty and and that's actually the part I don't like and I believe that's also a red wine cask. It is also a red wine cask.
Oh wow, so there might be something to.
That new *****
Mm-hmm OK, OK, Well we've dropped a bit of water in this it it does improve the nose.
It allows those fruits to come out so the red fruits, the kind of spicy fruit is there now. It's not lost in that harshness.
Oh yeah.
I don't get any spies on the.
Nose what do you? What do you see? By spice I mean spice for me is is anything from pepper to chilli to Christmas, yeah.
That I don't get any. I don't get any chilli pepper either.
No. Well I I got that to start over.
You say?
Best nice I can. The red fruits are really heavy once you drop the watering.
True, yes, that current black currants and and and berries.
Definitely there, yeah.
So what does that do on the taste?
Actually makes me think I might buy a.
Bottle oh wow OK good.
Do we do we know if this is a it's? It's not costs, right? But it's not 40% either so.
I I don't know. I have a feeling I have read that it's at cash strength might be right. Yeah 58.7 that that explains why we're feeling that.
This is.
Oh yeah, lovely yes. If you put a drop of water in there, it actually kills a lot of that chilli saltiness. Yeah, that's down.
Ha ha ha.
Well, it's.
There was the saltiness, but leaves the chilli there because that's just warming me.
I think that this is a very different whiskey with a couple of drops of water.
In it really is, it is.
So you could basically buy it, buy this bottle and it would go a long way 'cause she died, diluting it all the time.
How I can? Yeah, you know. Something I we've described whiskies as fireside, whiskeys before, but inevitably they're kind of petered and and stuff there. I there's virtually no peat in this at all, but it is definitely a nice warming winter.
Kind of, you know, autumn that's the right period of time, autumn whiskey.
And autumn whiskey. I think I would.
Very cool.
I don't think so, I don't.
Think I would go for.
A full bottle, but it's the idea. Put in the watering very nice. Anyway, there we go.
So D'instance 2008 it's nine years old before it's bottled. Very fruity, all round, a bit salty on the taste and on the nose. Can be a bit harsh.
But add a bit of water and it transforms it into something very different, but I think we'll put this one down as undecided undecideds just a little bit like our workflows, yes.
Yes, yes.
Do you like that?
Yeah, that that's a nice segue, yeah?
Yeah, nice segue.
We took a different subject. We we kind of just pretended to be consultant.
Yes, or change.
We never said it depends once yet.
Do you know?
That no, actually we didn't know anyway, but.
This is that.
Basically, we see four basic steps here so that initial talk that chat to get a baseline of what the workflow is. Yeah, yeah, it's interesting what the user wants.
What the user wants.
Have we talked about the user?
At all, yeah.
No OK, 'cause the user never knows what they.
Want yeah true.
That's actually is something we didn't talk about because I think part of our skill set is to get out of the.
User what they think they want to be able to turn it into some kind of reality and help them understand what it can actually deliver.
By asking the right questions, yeah, yeah.
Yeah, exactly, now that that is key, and then we kind of. Although there's probably other levels, but we kind of took one end which is a corporate kind of external application with big external customer audiences.
Is one kind of workflow and application and then the other side is that internal app and they both have.
Different kind of requirements and questions that we kind of need to obtain, and lifecycle is key to all of them.
So the the workflow in the application lifecycle is is key. I always like it when you can put an age on it so if you can say yeah well we think it'll last for 18 months and then we'll need to move on and do something else.
Because that way, at least you know you've not got a consistent update and support it, and make sure it works with you. Know 'cause every.
And you can also you can.
Also put a value on it.
If you say we've got it for 18 months, so that probably means 300 runs.
If we're going to build it for €3000, that means €30 per run. Yeah, are you happy with that cost?
That's true, I like that. I like that very much.
Uhm, but but I think in terms of of knowing where that lifecycle is, 'cause what you have to remember is every workflow you build has a cost in terms of your operations and support and all that.
And then I know that we think yeah, have so many tickets and there's a cost associated with it. But at some point your.
Operations organisation can only support a certain number of applications, you know because of the amount of upgrades and changes and that kind of stuff.
So you know, from an IT director's perspective, you need to be sort of very careful that you don't end up taking on 75 power apps written by users and.
Yeah, that will eventually come your way.
True, but then again, if you say that your power platform is going to be your.
For main platform that you want to build stuff on it, you get a proper team to not only build stuff on it, but also to manage it.
I'm giving it down.
That might also drive your costs down and drive your risks down and your compliance. Yup, and.
It will do.
And and that will work well for any organisation. That is, well, that is process driven and and it has to be.
Yes, yes.
Process driven, you can't otherwise.
End up with, you know, 1000 little workflows all doing something that only gets run twice a year, et cetera, et cetera.
But if you got a.
Good Heineken were a great process organisation so they were able to work out how to get from sand to tins and put beer in. It's a lemon bars so a lot of the workflows and stuff were very much around that process.
When things dropped into a finance process or moved into a research and development process or move back to a sales process, it's very cool, but you kind of got to have a a big picture on all that and work out where it fits in place. So yes, you're right for processors done, it's a minefield really, isn't it?
This I mean, if you think about it, it's so complex the you know the the variables and the options and the choices for developing any kind of workflows you've kind of got to put some level of control into it somehow.
So you know only allowing people, for example, to have the power automate app on MST.
Oh yeah, yeah.
Teams, if they take on certain responsibilities and acceptance so.
It's the same thing with SharePoint. Yeah, do you want to give people the capability of creating their own share point sites or libraries or adding content types?
Or adding metadata or whatever.
And of course we can back off to our conversations on the last podcast about personas and which personas.
Needs to create workflows which personas need to or even have the.
Depth of knowledge in the organisation to be able to even.
Think about it. Do you wanna provide?
Training to those.
Or do you want to give them some?
Kind of driver.
's licence after they pass the test that they know that you know that they can actually build a proper workflow.
So does that mean?
We we got this a little wrong.
No, that means we've got something to talk about on our next episode.
But but now I was just thinking that you know, maybe we have two clear processes here. That kind of app process which has a fixed kind.
Of you know, processing governance but also that kind of no code thing. So maybe maybe we need to think about it as a separate process.
Yeah I think so, yeah.
All right cool all right, that was actually more interesting than I thought it was.
In the end, even.
Though I did say workforce instead of workflow.
Yeah, well.
And you have to drive so you finished your whiskey. But hey, I still have a little drop here, cheers. See how the water actually makes a change on that, but that was quite fun.
Good number 90 over and done with.
All right cool. Thank you, dear listeners. And as always Steve, you will say the goodbyes.
I will like I said that's where you got the last word.
Yeah, you you go.
Right, say goodbye to everybody then.
Bye bye.
Not millennials, we are the true.
Steve, maybe more than my name.
Maturing the business like whiskey and about.

Creators and Guests

Steve Dalby
Steve Dalby
Podcaster "Office365Distilled" Driving Collaboration Business Goals, Speaking about Governance, Whiskey taster and imbiber all round father and good guy.
EP90: Workflow starts with the TALK
Broadcast by