Chevron, PWC, Power Apps and Azure AI with Mark Slosberg

Transcript

WEBVTT

00:00:01.826 --> 00:00:03.431
Welcome to the Power Platform Show.

00:00:03.431 --> 00:00:04.945
Thanks for joining me today.

00:00:04.945 --> 00:00:11.449
I hope today's guest inspires and educates you on the possibilities of the Microsoft Power Platform.

00:00:11.449 --> 00:00:13.868
Now let's get on with the show.

00:00:13.868 --> 00:00:26.990
Today's guest is from Bo in Washington, united States.

00:00:26.990 --> 00:00:30.190
He's a principal owner of Hanging Valley Enterprises.

00:00:30.190 --> 00:00:37.619
He's spent 42 years successfully convincing, selling, planning and delivering profitable information technology business outcomes.

00:00:37.619 --> 00:00:49.143
Right now, his core technology is a strategic business and solution architecture and successful use of enterprise class Microsoft Cloud technologies.

00:00:49.143 --> 00:00:52.350
You can find links to his bio, social media, etc.

00:00:52.350 --> 00:00:54.241
In the show notes for this episode.

00:00:54.241 --> 00:00:55.384
Welcome to the show, mark.

00:00:55.865 --> 00:00:57.371
Thank you, pleasure to be here.

00:00:58.119 --> 00:00:59.204
Good to have you here today.

00:00:59.204 --> 00:01:05.049
I always like to start with getting to know you so the audience can feel like they have your sense of place.

00:01:05.049 --> 00:01:08.165
So tell me a bit about food, family and fun.

00:01:08.165 --> 00:01:11.713
What you do outside your technology skill set.

00:01:12.140 --> 00:01:12.521
You bet?

00:01:12.521 --> 00:01:19.260
Well, I'm working on my 42nd year of marriage and we always joke when it comes to food.

00:01:19.260 --> 00:01:26.373
We just finished a remodel where this will be our fourth kitchen that we have designed and implemented.

00:01:26.373 --> 00:01:47.001
This one is a similar to the last one we did, but we bought a little farmhouse outside of Seattle and spent about six months redoing it, and the real, the real focus on that was we have found ourselves to be two type A cooks, so the trick was to build a kitchen that we could both work in without killing each other.

00:01:47.001 --> 00:01:54.084
But in terms of food, I always joke that I'm on a seafood diet, meaning I see food and I eat it.

00:01:54.284 --> 00:01:55.287
Nice, nice.

00:01:55.287 --> 00:01:57.662
Do you have any specialty dishes, though, that you like to cook?

00:01:58.504 --> 00:01:59.966
Well, we just had pizza night.

00:01:59.966 --> 00:02:11.967
That's always a good thing with the kids and we've got grandkids and so everybody gets around and they all pick a pizza and they do their own pizzas and we just sort of supervise at that point.

00:02:11.967 --> 00:02:20.550
But I'm the smoking king and I'm the short order breakfast thing and my wife builds the dishes that have 27,000 instructions.

00:02:21.451 --> 00:02:22.554
Nice, nice.

00:02:22.554 --> 00:02:30.788
So when you say the smoking, does that mean that you have like one of those ceramic eggs or a smoker and things like that?

00:02:31.348 --> 00:02:35.181
Well, I have two things that I use generally.

00:02:35.181 --> 00:02:41.921
One is I think it's called smoking it, which is a little box smoker, and I use that primarily for the meats.

00:02:41.921 --> 00:02:50.495
We just did a smoked turkey breast the other day, but I'll do brisket, then I'll do pastrami and all that.

00:02:50.495 --> 00:03:03.254
But I really enjoy doing fish and so I have a killer black cod or a sable recipe and I do a lot of salmon, lox and smoked salmon, hot and cold smoked.

00:03:03.254 --> 00:03:10.610
And there I actually have a cold smoker that I just put on an old Weber and it generates just a bunch of smoke but no heat.

00:03:11.132 --> 00:03:13.679
Nice, nice, very cool.

00:03:13.679 --> 00:03:19.409
Tell me, tell me about what you do when you're not thinking about technology.

00:03:19.409 --> 00:03:22.401
So, we try to do a lot of hiking.

00:03:22.441 --> 00:03:23.663
We do a lot of gardening.

00:03:23.663 --> 00:03:31.883
We moved out of Seattle about two years ago after almost 50 years I had been there almost 50 years and so we have this big.

00:03:31.883 --> 00:03:37.560
It's not so big, it's 0.84 of an acre, but it's just a lot of land.

00:03:37.560 --> 00:03:50.610
But we've got gardening and we do a lot of both ornamental and food, and then I also enjoy, and what we're trying to do with this one is this property backs up on a little slough.

00:03:50.610 --> 00:03:59.989
We're out in farm country, there's cows everywhere, and so we're looking to put in natives around the slough and sort of reclaim that.

00:03:59.989 --> 00:04:07.770
So the goal, though, is to build a landscape that has the least amount of maintenance as possible.

00:04:07.770 --> 00:04:11.967
Nice, and we're not there now, so we have a long way to go with this new piece.

00:04:13.001 --> 00:04:15.569
It sounds like you and I are on similar journeys.

00:04:15.569 --> 00:04:42.180
So I've got about an acre and a half of land and it's been farmed for 150 odd years and there's only one original tree on the property from when I purchased it and now there's about 3,000, 4,000 trees that I've planted and I'm on a mission to get the property back into more of a forestry but a food forestry.

00:04:42.180 --> 00:05:09.430
So I'm using a concept called syntropic forestry, which is kind of, you know, around permaculture, and I want to produce as much food, particularly fruit and as well as vegetables, that are all on the land, right, all on my property, and then I'm totally surrounded by cows, as you say, like I've got farming all around me, except for one thing I'm high, so I look out over the Pacific Ocean, which is nice, yeah.

00:05:10.031 --> 00:05:11.716
Well, it's interesting.

00:05:11.716 --> 00:05:20.262
One of the reasons we moved to where we are is we're two miles north of our son and daughter-in-law's 57-acre organic vegetable farm.

00:05:20.262 --> 00:05:26.300
That took about 10 years for us to get it organized and in, and I'm glad it's theirs and not mine.

00:05:26.300 --> 00:05:33.281
But they've got about 45 acres under tillage right now and it's pretty exciting it's been.

00:05:33.300 --> 00:05:35.004
really it's been very impressive.

00:05:35.004 --> 00:05:45.663
As a dad, it makes you proud to see them do this, but it's a lot of work, so anyway, so we're there to close the farm, stand and take care of the kids and do stuff like that.

00:05:45.663 --> 00:05:47.406
So there's a lot of that.

00:05:47.406 --> 00:05:49.610
That's outside of whatever I'm working.

00:05:49.971 --> 00:05:50.372
I love it.

00:05:50.372 --> 00:05:50.980
I love it.

00:05:50.980 --> 00:05:52.161
Listen.

00:05:52.161 --> 00:05:58.473
In this show I want to really drill into the wisdom of your tech career.

00:05:58.473 --> 00:06:00.923
How did you get involved in Microsoft technology?

00:06:02.125 --> 00:06:07.958
So it's funny because my career is 45 or so years.

00:06:07.958 --> 00:06:10.564
I got really started in 1981.

00:06:10.564 --> 00:06:12.629
And I'm actually a forester by training.

00:06:12.629 --> 00:06:14.937
That is my college training.

00:06:14.937 --> 00:06:16.800
But I was really into bio math.

00:06:17.141 --> 00:06:41.002
But at 20 or 21, I found myself on Saturdays working at the computer land of Bellevue, washington, which was just down the road from the original Microsoft campus the original Microsoft campus and Microsoft was in fact one of our earliest customers at the time and they were buying Apple IIs to put the Z80 card on.

00:06:41.002 --> 00:06:47.622
They were selling the Z80 card so they could put microsoft basic on to apple twos, and so we were very involved.

00:06:47.622 --> 00:06:52.237
But I was kind I evolved into an anti-microsoft for many years.

00:06:52.237 --> 00:06:56.531
I probably would have been a lot richer if I had just gone with the flow.

00:06:56.531 --> 00:06:56.831
Yeah.

00:06:56.831 --> 00:07:03.168
But I actually ended up driving towards the earliest low-code tools like dbase2.

00:07:03.168 --> 00:07:17.110
And I did my original coding in there and then from there moved on to CPM and then Unix, and so I had a long career in Unix all the way up to about 2002.

00:07:17.660 --> 00:07:36.560
And I formed my own firm and we sold SEO Xenix and we sold NCR Towers and we did custom application development and we were probably the first Oracle integrator in the Pacific Northwest and that was back in 1985 through 88.

00:07:36.560 --> 00:07:38.904
And that's when the joke was.

00:07:38.904 --> 00:07:44.668
When you were asked what platform does Oracle run best on, the answer was the slide projector.

00:07:44.668 --> 00:07:48.629
But it was one of the original cross-platform tool sets.

00:07:48.629 --> 00:07:50.845
That's brilliant and it was.

00:07:50.845 --> 00:07:52.322
It was brilliant and I did.

00:07:52.322 --> 00:07:55.271
We started selling larger and larger servers.

00:07:55.271 --> 00:08:07.490
We did very cost-effective Oracle servers back in those days and at some point in the late mid, early to mid 90s picked up Sun as a Sun microsystem.

00:08:07.490 --> 00:08:14.146
So I was a diehard Unix Solaris guy and I bought into McNeely's.

00:08:14.146 --> 00:08:18.350
You know he loved to trash Microsoft back in those days and I loved it.

00:08:18.449 --> 00:08:18.951
It was great.

00:08:18.951 --> 00:08:25.949
But I also knew that you know, it was hard to argue with Microsoft in different areas.

00:08:25.949 --> 00:08:29.341
They just had achieved such a dominance.

00:08:29.341 --> 00:08:45.330
So when I sold my firm back in 99 and 2000, I sold it as a Sun reseller and as an Oracle database development team and we were very into Java at that point I had.

00:08:45.330 --> 00:08:49.663
The stories of me getting into Java are just as much fun as my Microsoft stories.

00:08:49.663 --> 00:09:06.499
But what happened then was towards the early part of 2000, I started getting after I shut the firm down, we sold it and then I worked for that firm for a while and then gradually we sold off different pieces of it.

00:09:07.410 --> 00:09:24.201
So in early 2000s, 2003, 2004, I joined a Canadian consulting firm as a partner and they had a Microsoft practice and they were involved in Dynamics and this was, you know, the earliest Dynamics versions, on-prem.

00:09:24.201 --> 00:09:29.049
And you know I went to training and I did all kinds of things and I thought this is really clever.

00:09:29.049 --> 00:09:41.601
And of course Salesforce was coming out at the time with a SaaS version and then Dynamics then at that point was on-prem with a SAS version and then Dynamics then at that point was on-prem and I got very interested in particularly because I'd always been doing these low-code types of things.

00:09:41.601 --> 00:09:48.162
My deviation with Java was all right, I'm going to drop back to a 3GL and I spent a lot of time doing that.

00:09:48.671 --> 00:09:57.182
But I was always more interested in what we called fourth generation languages but are now called low-code it's, you know, very hot, hot, trendy term.

00:09:57.182 --> 00:10:03.950
So I kind of followed it along and I went through a number of large consulting firms.

00:10:03.950 --> 00:10:26.091
I went from the Canadian firm to a firm called BearingPoint which had spun out of KPMG, and that was an interesting experience Because when I joined BearingPoint we were 17,500 employees and within two years I was one of the last 300 as they sunk into bankruptcy because we learned that consultants are not very good at running public companies.

00:10:26.091 --> 00:10:34.462
But while I was doing that, I had jumped into a big project at Chevron and Chevron at the time.

00:10:34.462 --> 00:10:41.899
So this is 2006 through 2009, roughly, and Chevron was a huge Microsoft consumer.

00:10:41.918 --> 00:11:23.009
They had probably one of the largest volume license agreements at the time and we were doing I was actually working with management consultants doing a turnaround for, if you can imagine, there was a division of Chevron that wasn't making quite enough money and you know it was a very interesting story about that and they had hired this group of turnaround specialists to do a turnaround and these guys had been working with sort of lightweight Microsoft technologies Excel they were Excel jockeys and a little access and things like that and they had developed the methodology for turning companies around for driving profitability.

00:11:23.009 --> 00:11:34.128
That was their specialty and they snagged this job at Chevron before BearingPoint was in trouble or actually BearingPoint was already in trouble but nobody knew about it at the time.

00:11:34.128 --> 00:11:35.533
That's kind of the way that worked.

00:11:35.533 --> 00:11:53.833
But what was interesting was I got parachuted in after the guy who did the initial design for the software left to go to Microsoft, as it turns out and he's still there, and so he left me with this design and what they had was they had these small scale tools that suddenly had to be made larger.

00:11:53.833 --> 00:12:02.938
So where they had done things for five or 10 users, suddenly we had 1,200 users and it was determined to go with SharePoint.

00:12:03.269 --> 00:12:13.216
So we went, we did, went with SharePoint and I for me it was really the first experience with enterprise class Microsoft tools and there were a lot of issues.

00:12:13.216 --> 00:12:30.081
I mean, you know, we all know that oftentimes Microsoft releases stuff that's not quite ready for prime time and makes promises to its customers that are slightly in excess or in advance of what's actually being delivered.

00:12:30.081 --> 00:12:42.821
But it was a hugely successful project because in 18 months we delivered this tool set that was entirely SQL Server, infopath and SharePoint.

00:12:42.821 --> 00:12:52.783
That was the only tools that we used and it drove profitability from about $3 million a year for this business unit to $300 million a year.

00:12:52.783 --> 00:13:07.778
And, interestingly enough, that was sustainable for 10 years because I, after I finished that project, I walked away from Chevron, came back almost 10 years to the day and they were still using this application.

00:13:07.778 --> 00:13:18.000
It was pretty amazing and it's funny because when we implemented it, chevron had very strict enterprise architecture standards and they were not.

00:13:18.701 --> 00:13:23.166
This was 2009, they weren't or 2008,.

00:13:23.166 --> 00:13:26.120
They had yet to certify SQL Server 2008.

00:13:26.120 --> 00:13:37.000
They were still certified on the previous version and I had to go to bat and say, look, I'm not going to release this software on the old version of SQL Server because it's going to be obsolete.

00:13:37.000 --> 00:13:39.376
You guys are going to have to upgrade it and do all this kind of stuff.

00:13:39.376 --> 00:13:49.080
So I forced it and a few of us were forcing their enterprise architecture team to accept sort of a trial usage of the database.

00:13:49.080 --> 00:14:02.323
And it's funny because when I came back 10 years later, the only reason they were thinking about changing it is because SQL Server 2008 was finally being obsoleted by Microsoft and they needed to make a change.

00:14:02.323 --> 00:14:07.655
So it was like, ok, that's what's forcing us, but that gives you a little background on you know.

00:14:07.696 --> 00:14:21.749
So at that point I had Microsoft stamped all over my face and, through a number of acquisitions, I ended up being acquired, ultimately by Presswaterhouse Coopers PwC, and that was in 2010.

00:14:21.749 --> 00:14:31.404
I joined them in January of 2010 and spent almost seven years there as a director and managing director.

00:14:31.404 --> 00:14:37.897
I was a little old to become a partner when you join the firm in there, so it was a bit of a challenge.

00:14:37.897 --> 00:14:47.778
But what I started as a profitability analyst you know, specialist, how do you use technology to drive profits, et cetera, et cetera.

00:14:47.778 --> 00:14:51.139
But I always had Microsoft stamped on my forehead.

00:14:51.870 --> 00:15:13.524
And when I joined the Seattle office of PwC, I came to find out that they had a tens of million dollar worth of direct consulting business with Microsoft, particularly in compliance and tax, and not traditional reseller business, but just really classic consulting.

00:15:13.524 --> 00:15:22.572
Interestingly enough, what we found out was that Microsoft would frequently say to the leadership of PwC.

00:15:22.572 --> 00:15:29.044
You know we give you a lot of money every year for consulting, traditional consulting.

00:15:29.044 --> 00:15:36.057
It would sure be nice if you had a Microsoft practice, and at the time they literally had no practice.

00:15:36.136 --> 00:15:40.688
Had a Microsoft practice and at the time they literally had no practice.

00:15:40.688 --> 00:15:48.649
There were thousands, tens of thousands of Excel jockeys and people with SQL Server experience and things like that spread throughout the firm, but they did no traditional Microsoft reseller partner consulting.

00:15:48.649 --> 00:15:56.482
So I spent pretty much my seven years starting and growing that practice, getting them to finally.

00:15:56.482 --> 00:16:02.217
You know, at the time Oracle Financials was the darling inside of POBC or SAP.

00:16:02.217 --> 00:16:06.532
Dynamics was not a tool at all that anybody was interested in.

00:16:06.532 --> 00:16:24.341
It was treated as a low value tool, and so during the seven years that I was there, we affected that change and so as I left, they had just started a Dynamics FinApps practice and CRM practice.

00:16:24.341 --> 00:16:28.355
We were doing more and more with sort of evolving from SharePoint.

00:16:28.514 --> 00:16:35.076
And that's actually how I got into the Power Platform is that this would have been 2016,.

00:16:35.076 --> 00:16:46.754
Roughly, and I had spent a lot of time helping convert all of the Excel jockeys at the firm into Power BI, basically getting them up to speed.

00:16:46.754 --> 00:16:48.583
And we started, of course, with Power Pivot.

00:16:48.583 --> 00:17:13.464
People probably never even heard of that anymore, but that was the first step in going there and I actually did some interesting consulting inside of Microsoft during their product planning exercises when they were looking for I have to be careful how I say this but the discussion was can we find and build a Tableau killer back in those days?

00:17:13.464 --> 00:17:16.839
And of course, power BI has emerged as that player.

00:17:16.839 --> 00:17:20.240
So Power BI emerges very familiar with it.

00:17:20.240 --> 00:17:32.696
We had done a lot of SharePoint work and I had done way more with InfoPath than I ever cared to admit, and we all knew that there were missing pieces in the stack.

00:17:33.490 --> 00:17:43.059
And just as I was leaving in my last four to six months at PwC, they were sort of getting ready to release PowerApps.

00:17:43.059 --> 00:17:44.844
They just bought ADX Studio.

00:17:44.844 --> 00:17:48.297
So we had meetings with the guy who had sold ADX Studio.

00:17:48.297 --> 00:18:05.173
He went from being an important CEO to a product manager inside the firm and we had all these meetings and so I got familiar with the fact that they were trying to get PwC to pick up these tools and become an active reseller or, you know, developer of it.

00:18:05.173 --> 00:18:14.895
And um, and we, pwc and I parted ways in uh November of um, uh, whatever, that was uh 16.

00:18:14.895 --> 00:18:23.825
Cause it was the day before my 60th birthday, because of course, partners have to retire at a place like PwC at 16.

00:18:23.825 --> 00:18:29.218
Now, while I technically wasn't a full partner, I was what they call a non-equity partner.

00:18:29.598 --> 00:18:32.092
They're pretty much tired of you by that time.

00:18:32.092 --> 00:18:37.873
So I walked out the door, sat around for a couple of months just sort of saying oh, this is nice.

00:18:37.873 --> 00:19:04.058
And got back involved, and because I was down the street from Microsoft and I had met a bunch of these people, I talked to them and started looking at Power Apps right in early 2017, when, frankly, the platform was really soft and squishy and half the stuff didn't work the way it was supposed to and there was no documentation and nobody knew what it was doing.

00:19:04.058 --> 00:19:06.209
But I was intrigued.

00:19:06.209 --> 00:19:11.920
I was intrigued by the thing and so I built some sample apps and enjoyed that a lot.

00:19:12.730 --> 00:19:17.359
And then I went off to return to Chevron, did a very big.

00:19:17.359 --> 00:19:21.717
When I returned there, I was doing a big Azure backend AI data thing.

00:19:21.717 --> 00:19:42.417
But while I was there, I started to see Power Apps starting to show up on the desktop at Chevron and again, thinking back to how far backwards they usually are in their enterprise architecture approvals, I was surprised and I said you know, there must be something to this from an enterprise point of view that's going to make it.

00:19:42.865 --> 00:19:59.509
And then I jumped back in and the next thing I knew I had nailed down a job with a big commercial seed distributor to do a Power Apps application that is still running right now and I'm in the process of packaging for release in the AppSource.

00:20:00.089 --> 00:20:00.893
Nice, Nice.

00:20:00.893 --> 00:20:02.376
So give us a.

00:20:02.376 --> 00:20:11.218
It's interesting your whole InfoPath experience, your SharePoint experience, because I think InfoPath's biggest shortcoming was that it didn't work on a mobile device.

00:20:13.826 --> 00:20:28.153
Well, what I will tell you is this is a little story Someone took the app that we built in InfoPath and showed it to the original InfoPath developer and he about had a heart attack.

00:20:28.153 --> 00:20:33.066
He said I had never imagined and this forget mobile and stuff.

00:20:33.066 --> 00:20:34.347
We're back a long time ago.

00:20:34.347 --> 00:20:35.971
This is, you know, 2009.

00:20:35.971 --> 00:20:47.378
He says I had never imagined anyone in their right mind to do what, what you guys did with this app, and it was not the right thing to do.

00:20:47.378 --> 00:20:54.719
But you know, sometimes when you have a budget and you have a timeline, you don't get to choose exactly the perfect architecture.

00:20:55.546 --> 00:21:11.038
It's interesting too that these tools are created and then put out into the wild, into the universe, into the business and organizations, and people pick them up and they just you know, they build things with them and I love that.

00:21:11.038 --> 00:21:16.634
You know that story there, that what he intended he couldn't believe.

00:21:16.634 --> 00:21:22.253
You know what people had forward and taken and turned into something else, and I think we see this all the time with the Power Platform.

00:21:22.253 --> 00:21:31.752
Tell us a bit about the seed application as in what's it designed to do, who's it for and what are your thoughts around it.

00:21:32.404 --> 00:21:34.432
Yeah, so it's been fascinating.

00:21:34.432 --> 00:21:38.876
So, as I mentioned, my son has this farm two miles away.

00:21:38.876 --> 00:21:49.032
He was working for probably the largest seed distributor in the country top on the global side and it was an interesting firm.

00:21:49.032 --> 00:22:12.417
He worked for the seed distribution unit, which is about $130 million in revenue so a big company, not teeny, but they were a subsidiary of a bigger billion-dollar firm that does all kinds of other agriculture-related products fertilizers and chemicals and a myriad of products and in fact they have.

00:22:12.417 --> 00:22:18.057
Of that, billion dollars of revenue is spread across 32 separate subsidiaries.

00:22:18.196 --> 00:22:25.176
Wow, okay, those 32 separate subsidiaries were all running on one single instance.

00:22:25.176 --> 00:22:33.544
And when I say were, they are still and they will be for the foreseeable future, till long after I'm gone is what I predict?

00:22:33.544 --> 00:22:46.935
Wow, on a single, very large instance of JD Edwards that's 25 years old maybe 28 by this time and it's so old that it predates the purchase by Oracle.

00:22:46.935 --> 00:22:55.351
And so back in those days when you bought JD Edwards, you could pick which database was going to be on the backend, and of course it was SQL Server.

00:22:55.351 --> 00:23:01.846
Now, if they had bought it, you know, 10 years later or something after the thing, it would have been an Oracle database, but anyway.

00:23:01.846 --> 00:23:20.394
So as a result of this, each subsidiary was not allowed to make any customizations whatsoever to the app and, to their credit, they were very disciplined about it and that's the only reason it's running today, 25 or 28 years later, is that they could take all the updates.

00:23:20.394 --> 00:23:23.955
Whenever JD Edwards sent an update, they could take it and apply it.

00:23:24.464 --> 00:23:30.032
Now the problem was that that meant that this distributor did not get any customization whatsoever.

00:23:30.032 --> 00:23:31.156
Was that?

00:23:31.156 --> 00:23:33.785
That meant that this distributor did not get any customization whatsoever?

00:23:33.785 --> 00:23:51.545
And it was a vintage 1985 Java web service application which I knew well because I built a lot of those in the old days Ugly boring web-based interface, not very much fun and very few people actually in the organization could actually use it.

00:23:51.605 --> 00:24:08.375
They had some very highly trained but really just you know pretty much data monkeys, unfortunately, that would use the system and the salespeople were out selling all over the country in the wilds, wherever they were.

00:24:08.375 --> 00:24:20.230
You know they were visiting farms and ranches and things wherever they happened to be, both in the US and in Mexico, and if they had a question about do we have this seed to sell?

00:24:20.230 --> 00:24:22.634
How much of this seed do we have?

00:24:22.634 --> 00:24:24.159
How much is it going to be.

00:24:24.159 --> 00:24:26.047
How quickly can I get it?

00:24:26.047 --> 00:24:27.028
Can I get more?

00:24:27.250 --> 00:24:39.117
All of those types of questions required phone calls, emails, texts, every imaginable communication with a sales coordinator, of which they had a team of 25.

00:24:39.117 --> 00:25:01.395
There was probably 25 salespeople teamed up with 15 or 20 sales coordinators and it was all manual, every bit of it, and texts were getting lost and emails were getting lost and calls were being missed and it was just painful and it was interesting because they were trying to grow at the time and they brought in a couple of salespeople from another company.

00:25:01.395 --> 00:25:13.016
They kind of picked them up, as they say, and they all said you know, mr President, I'll spare the names and all you need to buy a CRM system.

00:25:13.016 --> 00:25:15.753
And they're all telling him you need a CRM system.

00:25:15.753 --> 00:25:22.198
And that was because, from the old company they were at, they were using something that they called a CRM system.

00:25:22.724 --> 00:25:28.636
Now, nobody knew what the actual functionality was, but it was being treated as a CRM system because that's what they said it was.

00:25:28.636 --> 00:25:32.105
But it was being treated as a CRM system because that's what they said it was.

00:25:32.105 --> 00:25:41.470
So one day I get a call from the CEO and he knew from my son that I did this kind of work.

00:25:41.470 --> 00:25:44.175
He says, mark, everybody's telling me I need a CRM system and I don't know what to do.

00:25:44.175 --> 00:25:45.017
Can you help me?

00:25:45.017 --> 00:25:50.915
And so I said yeah, and so I took a small entry, sort of discovery project.

00:25:50.996 --> 00:25:53.108
I said I'll come in, I'll figure out what's going on.

00:25:53.108 --> 00:26:04.198
I'll figure out what these people are asking for and see what the requirements are and examine your backend to see what's going on, because some people had said, oh well, you could switch to a new accounting system.

00:26:04.198 --> 00:26:06.048
Well, that wasn't going to happen, but that's you know.

00:26:06.048 --> 00:26:08.315
Somebody would say, well, we'll just go buy a new accounting system.

00:26:08.315 --> 00:26:21.298
And there was a lot of unrealistic conversation, which is pretty much typically what happens and it became very clear very early in the discussion that they did not need a CRM system.

00:26:21.298 --> 00:26:23.252
They didn't even know what a CRM system did.

00:26:23.252 --> 00:26:33.892
The notion of an opportunity and turning into a, you know, a prospect turning into a customer and an opportunity turning into a sale that was not what they were tracking.

00:26:33.892 --> 00:26:36.113
These were established customers.

00:26:36.113 --> 00:26:40.096
They had different kinds of customers in different parts of the country.

00:26:40.096 --> 00:26:46.854
None of them would have worked in a standard CRM thing and all the sales guys had their own way of doing things.

00:26:46.854 --> 00:26:48.825
So the last thing they wanted was something standardized.

00:26:48.825 --> 00:27:01.375
But what they did need is they needed access, information, access to the central servers, and they needed it at their fingertips, whether that was Now.

00:27:01.576 --> 00:27:06.631
Initially they weren't thinking mobile devices, although that very rapidly moved into that form.

00:27:06.631 --> 00:27:24.914
They were initially thinking laptops that they have out in the wilds, and so I did a little bit of prototyping to see whether I could use a Power App to get to, and, as it turned out, there was two aspects to this project.

00:27:24.914 --> 00:27:28.611
The first aspect was getting the information out to the salespeople.

00:27:28.611 --> 00:27:38.885
The other aspect was that they had a controller who was a very good controller who didn't know anything about database systems whatsoever.

00:27:38.885 --> 00:27:40.328
This just had no idea.

00:27:40.328 --> 00:27:49.211
And he had gotten all hooked up with Power BI and built this really, really sophisticated, very cool dashboard that everyone was bought into.

00:27:49.211 --> 00:27:50.453
They said this is incredible.

00:27:51.135 --> 00:28:11.419
The only problem was that they were populating the data entirely manually, with stripping out data strips and sticking them in and doing this, and the truth was the dashboard could barely be updated and be ready six weeks after the close of a period.

00:28:11.419 --> 00:28:12.126
It was just.

00:28:12.126 --> 00:28:14.334
That was just the reality of it.

00:28:14.334 --> 00:28:38.413
Now, on top of that, this firm the larger billion dollar firm they'd spent several million probably five to 10, I don't know exactly on Cognos and they had a huge Cognos implementation and the problem there was all the Cognos reports were wrong because nobody had reconciled the data and done data validation and partly because each sub was using.

00:28:38.413 --> 00:28:47.496
Now, they couldn't customize the software, but they could customize how they use the fields, so that meant any standard reporting couldn't be used across the companies.

00:28:47.496 --> 00:28:58.480
So they handed me this dashboard and said, oh and, while you're at it, can you hook this up so that it generates a thing?

00:28:58.480 --> 00:29:05.286
Well, as it turned out, while they'd spent a humongous amount, of money on this Cognos and weren't getting much value.

00:29:05.306 --> 00:29:11.878
What they were getting and what they did have, was an amazing data replication engine in SQL Server.

00:29:11.878 --> 00:29:18.577
Everything from the transactionals, the OLTP, was being replicated in a data warehouse every five minutes.

00:29:18.577 --> 00:29:36.660
So we actually had real-time inventory data or as real-time as was needed Five minutes was fine and so I had total access to this, and they gave me one SQL programmer who sort of knew where the data was at a very, very raw level.

00:29:37.425 --> 00:29:40.836
And what we did is we did a two-pronged assignment.

00:29:40.836 --> 00:29:48.313
The first thing we did was we standardized the views from all of that data, and that was very, very painful.

00:29:48.313 --> 00:29:49.047
It was not.

00:29:49.047 --> 00:30:04.239
There was a certain amount of data cleansing, there was a lot of data validation and mostly it was creating views that were correct, that reflected the information as it was supposed to be reflected for the seed company.

00:30:05.204 --> 00:30:16.915
And then what I did was I built some screens in Power Apps that would pull this stuff up in real time so that a salesperson could go in and say, hey, do I have this, do I have that?

00:30:16.915 --> 00:30:17.738
When can I have it?

00:30:17.738 --> 00:30:18.598
When's it coming in?

00:30:18.598 --> 00:30:20.172
What does the purchase orders look like?

00:30:20.172 --> 00:30:22.351
Which warehouse is it sitting in?

00:30:22.351 --> 00:30:23.273
All those things.

00:30:23.805 --> 00:30:28.352
So we had views that put that up and I just displayed them in Power Apps.

00:30:28.352 --> 00:30:33.519
And that was kind of interesting because I had to work through some design patterns.

00:30:33.519 --> 00:30:43.770
You start by putting the connectivity directly into Power Apps and that was clearly not performing adequately and the Power Apps were not really designed for that kind of volume of data.

00:30:43.770 --> 00:30:46.534
But we quickly created a design pattern.

00:30:46.534 --> 00:31:01.605
I kind of copied one from one of the guys that was out there pitching things and put in a Power Automate flow in between and just use that as the design pattern and just replicated it repeatedly for all the things with the stored procedures.

00:31:01.605 --> 00:31:02.607
So we did all that.

00:31:02.607 --> 00:31:05.915
And then I contracted with a Power BI specialist.

00:31:05.915 --> 00:31:12.067
So I did all of the data cleansing the views and all that, and then I did all the Power Apps development.

00:31:12.106 --> 00:31:19.912
And then I got a guy who was a really strong Power BI guy and I said here is this dashboard that they really like that.

00:31:19.912 --> 00:31:22.798
They don't know how to get populated.

00:31:22.798 --> 00:31:33.320
And we then together, because we were sitting in this data core, we built all the data flows so that that refreshed once a day or twice a day as necessary.

00:31:33.320 --> 00:31:36.112
And the CEO still uses that.

00:31:36.112 --> 00:31:39.314
Every morning he comes in and he gets what did I sell yesterday?

00:31:39.314 --> 00:31:41.051
What didn't I sell yesterday?

00:31:41.051 --> 00:31:47.508
All of those questions can be answered how profitable, who's selling, who's not, which of the salespeople are doing it.

00:31:47.907 --> 00:32:05.758
We implemented some hierarchical data controls so that the sales guy saw his data, the sales manager saw all the data for his salespeople and then the management layer were able to look across all the sales regions and do that.

00:32:05.758 --> 00:32:18.518
So we did these in tandem and then, while I was working on the Power Apps, we really started to talk about once we had the basic queries into the JD Edwards instance.

00:32:18.518 --> 00:32:48.266
Then it became how can we make this whole process efficient, more efficient, and how can we control what gets done and make it more standardized, because the salespeople are all going to sell the way they want to sell, but the backend processes need to be standardized and we went very, very far in terms of pulling all these together and we ended up creating a full-on pending order entry system, integrated with the query, so that salespeople could actually build.

00:32:48.688 --> 00:32:50.954
They would typically be called core quotes.

00:32:50.954 --> 00:32:53.671
It would be a quote that's not a sales order.

00:32:53.671 --> 00:32:57.346
They didn't like that term, so we called it a pending order.

00:32:57.346 --> 00:33:02.662
It had all of the same thing like a shopping cart, same kind of concepts in there.

00:33:02.662 --> 00:33:17.392
So we built shopping carts and all those things and then the real key was to hook it back to JD Edwards and that was a really fun part of the project, because everybody talks about uploading orders through an API but very few people can do it.

00:33:17.392 --> 00:33:23.765
We dabbled a little bit with some of the early Power Automate tools, but they weren't solid enough.

00:33:23.765 --> 00:33:33.093
It turns out that Oracle and JD Edwards had a pretty nice API builder orchestrator tool on their side, so they're internal IT guys.

00:33:33.093 --> 00:33:49.233
We worked together to create a contract and, with using Power BI, automate and Flows, I basically call the API, I drop the JSON record into JD Edwards, they pick it up and they go ahead and process it.

00:33:49.233 --> 00:33:54.758
I turned this thing over to them about a year and a half two years ago and it's still working.

00:33:54.964 --> 00:33:57.491
Wow, this is amazing, so good.

00:33:57.491 --> 00:34:03.645
I was thinking about the title for the show, and of course, it's going to be Power Apps over JD Edwards, because it's not something that you commonly hear.

00:34:03.645 --> 00:34:08.871
When you mentioned JD Edwards, I was like I had to go back to my mind the last time I heard of.

00:34:08.911 --> 00:34:09.652
JD Edwards.

00:34:09.652 --> 00:34:17.300
So, as an aside, there are still 9,000 worldwide implementations of JD Edwards, I believe it.

00:34:17.300 --> 00:34:23.637
Yeah, and I'm sure Oracle was planning on killing them all and pushing them into Oracle Financials, but that didn't happen.

00:34:23.657 --> 00:34:35.054
Yeah, the other thing I took away from that is the value of keeping a clean core in your systems, because they kept that clean core in JD Edwards and didn't do the modification there.

00:34:35.054 --> 00:34:39.253
It's enabled them to sweat that asset for a heck of a long time.

00:34:39.253 --> 00:34:41.059
Obviously we're well over time.

00:34:41.059 --> 00:34:57.275
I just will still wrap with one final question for you, and that is your observation now of the Power Platform what you can do with Power Apps, Power Automate, Power Pages, which, of course, was originally that ADX Studio product you talked about in its current iteration and form.

00:34:57.275 --> 00:34:59.170
What's your thoughts around?

00:34:59.170 --> 00:35:19.219
One, the robustness of it, the enterprise nature of it, the usefulness of it to extract data from anywhere across the organization not really who cares what the backend systems are to build assets that are highly mobile, useful for the use cases of a modern world.

00:35:19.864 --> 00:35:26.949
So it's interesting Now that I've done this for really a long time now I'm like I can't even believe how long I've been doing this.

00:35:26.949 --> 00:35:33.230
You know, early on I cursed every minute that I had to work on it because it didn't.

00:35:33.230 --> 00:35:34.432
Nobody knew how to do it.

00:35:34.432 --> 00:35:37.827
There was no documentation on syntax he had there was.

00:35:37.827 --> 00:35:42.298
The syntax for Power Automate was different than the syntax for Power Apps.

00:35:42.298 --> 00:35:49.831
And then I had to do all the SQL syntax on the backend and it was a nightmare to do and yet it couldn't be done any other way.

00:35:49.831 --> 00:35:50.347
You know what I'm saying.

00:35:50.347 --> 00:35:51.621
In other words, it was a nightmare to do and yet it couldn't be done any other way.

00:35:51.621 --> 00:35:51.858
You know what I'm saying.

00:35:51.858 --> 00:35:59.594
In other words, it was a unique, it's a unique set of power all put together, and then you combine that with the data flows and Power BI.

00:36:00.344 --> 00:36:12.257
The problem is the early promise of it being something that is appropriate for smaller or middle-sized companies to use is, I think, still elusive out there.

00:36:12.257 --> 00:36:15.431
You just have to know a lot, and I'm kind of a maniac.

00:36:15.431 --> 00:36:23.056
I've been trying to teach a few people how to do things and stuff, but part of what I do is because I've done this for 42 years.

00:36:23.056 --> 00:36:32.668
I have stuff in my head that I can call on, but to think about how you were going to learn this from scratch, I think that's still a challenge.

00:36:32.668 --> 00:36:33.510
So that's part of it.

00:36:33.510 --> 00:36:42.077
I think, weirdly, copilot has been just a huge thing because, you know, it allows me not to have to know everything.

00:36:42.077 --> 00:36:46.737
I can just ask when I know what I want to do, just tell me how to do it kind of thing.

00:36:46.757 --> 00:36:48.289
So that part's amazing.

00:36:48.289 --> 00:36:57.733
I am still struggling, I'm beating, the app is rock solid and I could build rock solid apps and I'm very comfortable and I would swear that to everyone.

00:36:57.733 --> 00:37:11.012
I'm building a system for my son's farm right now for vegetable production control, and I'm finally forced myself to go and learn model-driven apps and have them, you know, use them in a certain type of application versus Canvas and all that.

00:37:11.012 --> 00:37:34.634
I am trying to publish right now into AppSource and I'm finding that it's very similar to where it was, you know, three, four, five years ago, where stuff doesn't quite work the way it's supposed to and the documentation's sort of weird and there's a video that sort of does it, but then when it fails, there's no, you know there's, and that's where the challenges are.

00:37:34.634 --> 00:37:35.869
It's in that part.

00:37:36.565 --> 00:37:46.097
AppSource is changing all the time and it's recently gone through a bunch of changes that make it, dare I say, more confusing at the moment than enlightening.

00:37:46.097 --> 00:37:48.771
So I totally get what you mean.

00:37:48.771 --> 00:37:50.532
Thank you, mark, so much for coming on the show.

00:37:50.532 --> 00:37:51.447
I loved your story.

00:37:51.447 --> 00:37:57.789
There's other areas I wanted to delve in as you went through, because you know I have a background in some micro systems.

00:37:57.789 --> 00:38:08.126
I wanted to reminisce a bit about that and the Solaris, and you know I bought a bunch of Solaris machines back in the day for the org I was working for.

00:38:08.126 --> 00:38:10.050
So but we're out of time.

00:38:10.050 --> 00:38:11.614
Thank you so much for coming on the show.

00:38:12.054 --> 00:38:12.295
All right.

00:38:12.295 --> 00:38:13.038
Well, it's a pleasure.

00:38:13.038 --> 00:38:14.871
I look forward to talking to you again sometime.

00:38:16.606 --> 00:38:17.731
Hey, thanks for listening.

00:38:17.731 --> 00:38:22.817
I'm your host business application MVP Mark Smith, otherwise known as the NZ365 guy.

00:38:22.817 --> 00:38:26.715
If there's a guest you'd like to see on the show, please message me on LinkedIn.

00:38:26.715 --> 00:38:38.951
If you want to be a supporter of the show, please check out buymeacoffeecom.

00:38:38.951 --> 00:38:40.173
Forward slash NZ365guy.

00:38:40.173 --> 00:38:43.300
Stay safe out there and shoot for the stars.

Chevron, PWC, Power Apps and Azure AI with Mark Slosberg

Listen On

Recent Episodes

Chevron, PWC, Power Apps and Azure AI with Mark Slosberg

Join over 1500 Power Platform & Dynamics 365 pros for a quick, fun, and insightful 4-minute read each week!

Subscribe to Tech Trailblazer newsletter

Listen On

Recent Episodes