Eliminate Traditional ETL Data Transformations with Incorta

×

Michael Weinhauer
Senior Sales Engineer
Incorta

Michael has been designing, delivering and selling analytics solutions for over 28 years. Before joining Incorta, he was at Senturus and instrumental in the development of the Senturus Analytics Connector, which lets Tableau and Power BI use Cognos as a data source. During his career, he has gained a wealth of hands-on, practical BI and big data experience in solutions architect and sales roles at Oracle, IBM and SAP.

Mazen Arar
Director, Business Development
Incorta

Mazen started at Incorta in 2018 leading the support team, then he moved to leading the Global Customer Success team before joining the Business Development team. Prior to Incorta, Mazen was at Oracle where he played the role of Sr. Director leading the global Quality/Release team for EBS for over 15 years making him an expert of Oracle technologies.

Read more

×

Welcome to today’s Senturus webinar on Getting acquainted with Incorta.

0:08
Thanks for joining us today. Just a little bit of housekeeping here upfront, you’ll see on your GoToWebinar control panel to the right of your screen, there is a section for you to submit questions. So, if you have any questions that come up during today’s webinar, just type them into the questions panel, and we’ll do some Q and A at the end of today’s session. So, you can go ahead and submit questions any time during the webinar.

0:37
Other items: to obtain a copy of today’s presentation, just visit us at Senturus.com/resources. you’ll find a downloadable PDF of today’s slide deck there on our site.

0:53
And a quick overview of today’s agenda: we will do some quick introductions, and our presenters will cover some of the common challenges to accessing timely insights, provide an overview of the Incorta’s unified platform.

1:10
We’ll walk through a demo of the Incorta platform, and then we’ll wrap up with an overview of Senturus, and as I said, we’ll do live Q&A.

1:22
Terms of today’s presenters, we have Michael Weinhauer and Mazen Arar from Incorta, our primary presenters. Many of you may recognize the voice of Senturus alum. We were very sad to see Michael go, but it’s a great thing. He is a senior sales engineer for Incorta. And, among the things he did, while he was at Senturus, he helped develop the Senturus Analytics Connector, which connects to Tableau or Power BI to Cognos. He had gained a ton of experience with all organizations, including Oracle, IBM and SAP. Mazen is the Director of Business Development at Incorta.

2:12
Mazen has been with Incorta pretty much since the beginning. He’s played a variety of roles there. He is currently Director of Business Development, and prior to Incorta Mazen work that Oracle for over 15 years.

2:29
And from here, I’m going to turn it over to Mazen. He’s going to run a couple of polls and launch into today’s presentation. So, Michael, I’ll let you take it away.

2:40
Appreciate it. And it’s good to be back. Didn’t introduce yourself. Steve is a seasoned expert himself. And thanks for having us on the webinar.

2:58
So, as we usually do in these Webinars, which, by the way, will be recorded today, and it will get posted to the Website, along with a PDF of the deck and any questions we’re unable to get to at the end of the presentation, so you’ll be able to refer to this later on.

3:17
So we like to get a handle on the audience, if you will. We have two polls today, the first one is: what are your core systems of record in terms of ERP/CRM?

3:32
Is an Oracle any flavor of that right on perm cloud?

3:37
Yes.

3:39
I would say you can even probably long PeopleSoft and JD in their nets, we sap Salesforce or something else.

3:45
And you can check multiple responses.

3:52
And we’ll give you about 60 sec to get those answers in.

3:59
Go ahead and make your votes.

4:05
Got about two thirds of the audience, responding arrow. Wait for that to get just a little bit higher.

4:14
All right, we’re at about three quarters here. I’m going to go ahead and close this out and share it back with you all.

4:20
So the majority use some flavor of Oracle. And the other half are other.

4:26
Interestingly, only 13% SAP and 13% Salesforce. It surprises me that Salesforce is that low.

4:36
And then our second poll: what are some of the biggest challenges you are experiencing in your existing analytics environment, right?

5:36
Feel free if you did say other type it in the question panel,  it would be informative.

5:43
So, OK, we’ve got about three quarters.

5:46
I’m going to close that out shared back with the others in a little more evenly distributed, so And this is not real surprising You know that there’s a high number of people that are struggling with 1, 2 and 4 There’s a bunch on other, again, I’m curious about that and then you know as a significant, but smaller percentage lacking that ability to kind of go from high level KPIs to the details. So, thank you very much.

6:12
Hopefully, that was interesting or informative for you.

6:17
So, jumping into the presentation here, what we hear from our customers a lot of times is that the business and IT are struggling to make data driven decisions.

6:34
Now, while the questions are relevant and being asked and while the data is still current  there’s a lot of latency.

6:47
But a shocking number of business decisions that are made, either from gut or from data that’s days or weeks old.

6:56
We hear that people are struggling to connect to all their data for a complete view right there.

7:03
There’s a lot of data that gets sacrificed at the altar of performance, because you have to aggregate and winnow it down through ETL just to get things to run.

7:14
What we also see is the need to reduce the burden of analytics on IT?

7:19
As we saw in the poll responses that there’s a big backlog and self-service, is elusive, Right? Both from a cultural and a technical standpoint.

7:31
And the reason for that is what we see is, and this will resonate with all of you, nobody will be surprised by this is that the modern, quote, unquote, data architecture, right?

7:45
Where you lose a bunch of data in through this whole process.

7:51
So when a new business question is asked, the data gets sourced from your business sources, again, the oracles, the workdays, the saps, etcetera.

8:00
And that gets pulled into a data lake at this point.

8:04
And you might get close to 100% of that using your standard ETL tools, your five trans.

8:11
But then, to get that to the next level, so it can perform you end up getting about 25% of data down into your data warehouse.

8:22
Do you have another copy that’s made, then each functional business in the area oftentimes has their own chunk in the data warehouse. You have a finance data warehouse, an HR data warehouse, and then it has to be designed with its requirements ahead of time.

8:36
So in that, you lose almost 75% of the detail we buy, including the ability to traipse at lineage or granular level security and things of that nature.

8:48
Then it gets further aggregated for business consumption typically into a star schema or dimensional model, and it may be fractured into different data marts, cubes, things of that nature.

8:58
And then you get down into things like Tableau, extract, Power BI Imports and whatnot before it’s finally consumed by the end user in various tools.

9:08
So after all that, you end up with data that lacks accuracy.

9:11
Because you have multiple data copies and transformations, often hand coded ones, and you lose a lot of timeliness.

9:19
So our vision around this was looking at the traditional way of doing this.

9:27
You have a question you call IT, get on the list, grab the data, transform the data prep, the data.

9:35
Each one of those arrows represents a lot of time.

9:39
And so in short, you looked at that and say what if there’s a way that we can do this? Or I have a question, and all the data’s already there. And I can immediately go look at it. And that’s really the problem that Incorta seeks to solve.

9:51
And with that, I’m going to hand it over to Martin who’s going to get into the value proposition on what it brings to the table.

10:00
Thank you Mike. Hello everyone, I hope everyone is doing well.

10:07
I loved the polls and the results of the poll and I’m glad that I have some answers for those. But it’s really interesting. But before we talk about Incorta what makes Incorta unique, let’s talk about what is Incorta.

10:31
Incorta is the shortest distance between data source and insight. And what I mean by that? Let me explain a little bit.

10:51
it’s an end to end solution.

10:54
It allows you to extract data from your business applications, different databases, data files. Google Sheets are used in Incorta direct data mapping. So, whether Direct Data Mapping, we have over 30 connectors, that allows you to connect to, one source or multiple data sources. In fact, this is one of our sweet spots, where you can connect to, when database that you have today, and then some maybe historical data on SAP, and NAB. Together, you can connect to them, at the same time, join them, and present them on the same dashboard. So, that’s, that’s a plus For us. Also, once you get the data, that you extract the data into incur the court, that allows you to analyze this data, at scale, we’re talking millions and billions of rows, then you can just show it in a nice dashboards. We have a visualization tool that’s part of the platform itself.

11:49
If you want to use in Incorta visualization tool but we also understand some customers. And this is from experience. They love their Tableau. They love their Power BI, they do not want to change that. You can absolutely have you tool of choice on top of Incorta. So once the data comes into incorta, the all the calculations are done and everything, and it’s ready to be presented, you can have another tool.

12:16
Also, Incorta supports ML. I’ve worked with customers before big customers, where they get all the data incorta that. And then they have their ML scripts on top of incorta. So data scientists love us because we have the data prepared for them. You know, I’ve heard before, the one of the hardest things for a data scientist is getting the data in a place where they wanted to be to run ML scripts. So once the data is an incorta that you get the scripts on, and it’s ready for you, We have customers today that are used, and it’s in production, as I said earlier. So, incorta offer is basically, it’s a one stop shop with direct data mapping and in memory technology. I forgot to mention that. So when the data comes Incorta, that we do use the memory technology for calculations and getting the data ready to be presented, So when you click on a dashboard, or you, try to drill down on a dashboard, You don’t have to wait for 15 minutes. It opens up, by the way, like I said, use in the memory technology and that front.

13:15
How is Incorta that unique?

13:21
And I’ll, yeah, if you can build that for me, please.

13:25
So, obviously, performance and data agility. it’s a big ask from the customers, and we think we nailed it in there, So rather than having to flatten the pre aggregate reshape or model your data. incorta, that is going to allow you to pull all the data for all the tables record to record schema for schema directly into the incorta platform.

13:48
And also, you keep the same security rules. That means users can only see and access what they’re allowed to see at the source itself.

13:56
And this is a big question that always comes to us, in terms of security. So, that is definitely kept intact.

14:07
OK, so there, there’s where, the, as I mentioned, with Incorta, that comes from the data source, Incorta. And, of course, you can add additional data files if you want, like, Excel, spreadsheets, Google Sheets and is ready for analysis.

14:24
So, here we have, is all in the platform hundred percent data ingest, and 100% data analyzed. So, today, we see in some tools, you have to add filters and parameters to, make it, to have a smaller set of data.

14:41
Because you’re always worried about performance, you cannot get, possibly, get all the data, And so, in order to avoid poor performance of seeing customers, where they used to have lots of filters, and the metadata, and that doesn’t have much, because, in the future, if you want to add more to the dashboard, If you want to see more, you’re going to have to come back and redo everything you’ve done starts from the extraction, down to the schema, and so on. So with a direct data map, data mapping, you get all your data, integral it to be analyzed for the business.

15:16
OK, so this is one of our customers, favorite features. We did a survey before with our customer base. And number one was aside from performance, which was in common with all of them self-service capability. So, if, I mean, traditionally, today, if somebody wants to see a new column of data, or just add new insights to the dashboard, or just a simple modification, right? They have to call the IT team, Or whoever is taking care of that. As for the new data, And from my experience, again, I’ve worked with some customers, where they say, amazon, it takes us a week, and sometimes more. sometimes they have to build a whole new schema. just to accommodate this request. I’ve seen it with inquiry though, because you have all your data, already. They are, they are in the incorta.

16:03
I’ll let you do is as you get the call. In fact, when I say self-service, we have super users that don’t even need to call IT. They got themselves into the dashboard and they can see their data. It’s all there, right? I can see it, Let me load it myself, Boom, it’s there. So I’ve seen it where time shrinks from a week or more to a couple of hours or sometimes instant.

16:30
So I’ll let you read, just edit your dashboard. Then you’re in another favourite feature, which is really time to value. So traditionally, to implement a BI solution, I would say it takes about 6 to 18 months. And when I present this slide and say 6 to 18 months, I get pinged by people telling me, You’re being conservative.

16:50
Sometimes implement, and a BI project takes more than that. And, obviously, with all these stages, the early steps, back to back that you see there, It doesn’t make it easy at all, right?

17:04
So, within core, though, we’ve been averaging 4 to 7 weeks with the implementation projects. In fact, I’ve been involved in a project where, from requirement gathering to govern live, it took us, literally six weeks.

17:20
So, that, was a major heads for upper management. Obviously, that’s the time, and the cost was much less for the implementation.

17:30
This slide was created by a customer, It was done on their internal presentation. We just ask for permission to grab the slide and must have a name on it. But, definitely, this is the initial planning for the customer.

17:46
They had to build it ecosystem. And the proposal they came up with was what you see, where it says, Old way, that’s like, the proposal with that iteration, everything is going to be from 4 to 10 weeks.

18:00
And, and of course, that would be very hard to follow with upper management, because it’s what, it takes time, like I said, So, within quarters, they had to incorta the proposal, and this has done already.

18:14
So they were able to eliminate a lot of those back to back steps, and shrink the time down to 3 to 5 days, and that’s a huge, like I said, taking it to the board, that was a big hit, and the customized live on incorta that today. Another points I want to bring up, which is data integrity or trust in data, right? So, incorta that maintains 100% of the integrity.

18:39
It eliminates the need for multiple data copies and transformations, which makes the data the difficult to validate.

18:47
I know we have lots of customers that like to run audits on, probably on any BI tool they have in there. And sometimes, it’s hard, when you do the transformations and multiple copies.

19:00
It’s hard to really validate what you’re saying, but here, we get all the data in, And analytics is done on 100% of the data, which means any combination, wherever you do, just have that one copy of data.

19:14
So for audits him, it’s clear, customer can easily run an additive boards on the data source. And validated. And then incorta almost 1 to 1.

19:23
So I think that’s a good feature in there.

19:28
So we also offer data apps, or what we used to call blueprints to accelerate implementations. So given the knowledge and experience at encoder, we have lots of people that came from Oracle and SAP and other places.

19:43
So, by already know, in the source system, you get the complete schemas, the reports, dashboards, and that gets down big time from weeks to months on implementation time.

19:55
If it’s a big project, obviously, But it’s a, we have customers that are fans of our Blueprints or data apps. And, let’s, see, we offer data apps for Oracle EBS, Oracle, Cloud VPN, that’s sweet Salesforce, SAP, a JD, Edwards, and many more, this has already been a game changer for many implementation projects. And for the customers themselves. So, speaking of customers, obviously, I would like to show you here a sample of customers that trust in courts or with their data.

20:28
And I have personally been working with them, a light like Apple, Facebook, Starbucks, Stanford University, Cisco, there’s also Comcast, Netflix, and, many more. This is just a small sample of our customer said today, that trust us on their data. So, with the end, I think, I hope that you guys are anxious to see a demo of the product.

20:59
So, all of you that are on the line, watching this, if you’re like me and, you know, you heard in the introduction that I’ve been doing this for, I’ll just say, 25 plus years to not date myself beyond that already, huge number.

21:12
I’ve worked with just about every analytics tool out there from Oracle of business objects to Cognos to micro structures, Power BI.

21:20
And so, I’m very familiar with the moderate way on the old-school ETL pipelines.

21:26
And when I heard of Incorta, I was probably like, you know, I get it right.

21:31
So, I’m not going to go revivalists preacher here, but I’m asking you to take a step back and there’s a bit of an evangelical.

21:50
And I’m in the Incorta environment. Now, what I’m going to do is run you through a quick demo of the product that shows how we can ingest a large amount of data from an ERP like system, and combine it with some other data around the weather to do a combined analysis of how the weather impacts sales.

22:12
And so it’s designed to show the scale, the ability to combine different data sources, and the ease with which I can pull in and get from data to insights in a very short period of time. With a nominal amount of effort.

22:28
So, the first place I start is, everything is done here, in the browser, and Incorta can reside.

22:34
We’re a cloud first company, but you can have this on perm, if you want, and you go to the page list, select tools where you are, your various dashboards and whatnot, and those are to come up. I’ve got my KPIs.

22:52
And then I’ve got my various charts and graphs, and I’ve got some pivot tables, really taken its sweet time, I think it’s, uh, goto meeting is sucking up my bandwidth. So I’ve got a map and I’ve got, you know a bunch of information here. This is kind of, you know pulling the finished pie or turkey or whatever out of the oven, Right.

23:12
And you’ll notice that in this dashboard, I’ve got about eight point two billion dollars in revenue.

23:17
I’ve got nine million orders, 266 distinct skews, over 20 million products.

23:24
All right, and you’ll notice that when I go all the way down to the details here, I’ve got nine million, 65 thousand individual transactions.

23:33
And you’ll notice that I’m able to query this information from top to bottom very easily.

23:39
If I click on this, it’s automatically, you will see, when I build this, I don’t have to build these filters in, it does it automatically. Automatically filter down by state, New York, shows me the revolutionary, my top little guys.

23:53
All the way down to the 951,000 specific records that are germane to cities.

23:59
In New York, I can navigate extremely easily across a large set of data that has coverage, about 32 tables, and N by P or P case. And about nine million records in a flat file.

24:19
So, now, let’s kind of go back and I’m going to walk you through how we got there, right.

24:25
And so, you kind of got these six different tabs up here, and you start from the data pane.

24:31
And as Martin mentioned, we have about 250 connectors. We have a partnership with C data, and we also have our connectors, which gives us a ton of accessibility.

24:49
Whether it’s a database, a data lake data files, query services, you name it, you can, if there’s a connection to be had to your data source, you can probably get to it with the Incorta. I just simply pick one of those. And what I’m going to use is, I actually have a mySQL database here that has my online store. I’m just going to test that connection.

25:10
Then my local data file here is my weather file. And it’s a big old CSV file, about 10 gigabytes, and about, I want to say, nine million records of data.

25:22
So once I’ve created that connection, the next step is to create a schema.

25:29
And so, everyone on this call, I’m sure you all know what schema’s, I’ve got a bunch of schemas in here, and I’ll just show you this one. For, starters, it’s just like a schema and I’m in a standard relational database. Although, you’d have a little more flexibility here, this might be multi source, right? This includes some CSPs, I show you this CSP, because this is actually from an Oracle system, and as a billion records in it.

25:52
We’re able to compress that down to 37 gigabytes on disk, and we were, we loaded that in six minutes and 52 sec.

26:00
So that’s the first thing, is, we can really load data very quickly here.

26:04
And since I don’t want to spend six minutes and 52 sec of your time here today, I’m just going to call this. I’m going to create one of my own here.

26:14
And I’m going to pick the online store data source that I just tested out there. That’s a little smaller, but still 25, 25 million records.

26:23
And I’m going to use our schema wizard, which is going to expose all of the tables in that. Again, this is kind of an OLTP, it’s highly normalized, set up, and from here.

26:35
So Incorta has query the metadata. And it pulls back all the columns, and the data types, and it gets all the metadata information around that, And tries to tell you kind of what the roles are, of those of those fields, and whatnot. And then you can go ahead and change all this, right? You can pick and choose the ones that you want the tables. You want, the field you want, rename them, Change, the types, do all kinds of stuff. even customize the SQL.

27:02
But in the interest of, the time here, I’m just going to grab all of these, and the schema Wizards’ is actually going to interpolate all those joins.

27:13
If I choose the, the create joins got to disable my Slack Air, apologies for that, the popup’s there.

27:25
And so what in court is doing now is it’s actually going in, it’s pulling in all that metadata, and it’s going to create this schema for me, And then I’m going to go ahead.

27:36
And the next thing to do is to load that schema. So you can see here, I’ve got 32 tables and 33 joins.

27:45
Now, the difference is, here’s a view of it and I can see all the relationships. It’s detected all of those.

27:52
The important difference here is, what you would normally do now is this would go through a big ETL process, right? You would have to de normalize a lot of this.

28:02
You had exploded in size, because you flatten out things like all your product tables and your geos, and all that stuff. So that you can actually query it, right?

28:12
Because even modern databases, you’d go beyond 2 or 3 joins and that query is going to churn.

28:20
That’s not the case with Incorta, we’re actually going to load this. I’m going to make a straight up copy of every table in that source.

28:30
And what we’re also doing is a thing called Direct Data Mapping that Mazen alluded to before.

28:36
So, it’s low burden on the source system, because we’re just doing kind of a no select, specific column names from the table, We’re not doing any of the joins, on that source system, or just pulling that data straight over.

28:49
And then, by the way, subsequently, we would probably do incremental loads.

28:54
So we were only picking up the delta’s and minimizing that impact.

28:58
But the magic here is that incorta is a adjusting this very quickly. So you will see this should.

29:04
It should load up in about a minute or so.

29:08
But it’s also going to store this in a very compact format again, because it doesn’t have to de normalize it in parquet files.

29:15
And it’s also doing direct data mapping.

29:18
So the direct data mapping, the most expensive part of doing all these queries, is figuring out those query paths for my browser just went away, OK, Sorry, yes, Let me pull this back up here.

29:38
And I’m just going to go back in, the beauty of this is it’s in the cloud. So it should still be turned on our way there when I get there.

29:51
We’ll pick up right where we left off. So I’m going to go into my schema.

29:56
Back to my MKW Online.

29:58
And you can see here that that has finished loading now, and it took 41 sec to load 32 tables and 25 million records of data, right? So, it’s pulling this down really fast. And this is not a big box. This is, I want to say it’s about, it’s like 8 to 12 cores. And I want to say about 16 to 32 gigabytes of ram. So it’s not a giant box.

30:21
And it’s loaded all this stuff up. And I didn’t have to do anything with it. And it’s all joined up.

30:27
And I can actually go in now, and if I wanted to, I could go ahead and explore the data from there, and start building insights.

30:35
But what I really want to do here is, I need to join up my, my weather data, right? So, I’m going to go ahead. I’m going to go back out to my schema and to save time, we’ve already loaded that file up, and just to show you, I’m not lying to you. It loads in about a minute, 22nd, 74 million rows, and it’s only about 1.4 gigs.

30:55
Alright, and so what I’m going to do here, there’s I’m going to simply create a join where the Whether it’s Whoops. Backward.

31:13
Where the weather table is the parachutes.

31:19
And I’m going to join it to my OLTP schema.

32:02
Let’s go back out to the schema to my OLTP schema.

32:09
And I’m going to go to my sales order header table.

32:13
And I’m going to create the join here.

32:20
OK, and I’m going to grab my sales order header.

32:25
And I am going to pick my zip code.

32:30
I’m going to join that, too, the, to that weather file. I need to join that.

32:43
I’m the zip code and I need to join it on the date because this, that file has whether by zip code over date. So you can imagine how that gets to 74 million.

32:56
OK, so once I’ve done that, that’s all I need to do.

33:00
it up like 18 times. And I say done.

33:04
And so that’s it I don’t have to do any kind of other ETL here and you can see here’s my schema, and here it’s joined up to the weather, flat file, And you can see the joint criteria here, and I can just look at that.

33:17
And from here, I could actually go ahead and just start exploring the data, So, really net of all my talking, if I spent, if I, I just pulled this stuff in and did it, I could go and build insights from this in the space of about five minutes, right? But this is kind of a mess, because you’ve got all these tables that are not friendly named, and there’s a lot of them, so this might be a lot for your business users.

33:44
Maybe this is something you use as a power user, or something to sort of, to sort of validate the data, or when they really want to get into it. But 100% of my data is here.

33:55
Every single record from the top level, all the way down to the most granular level of detail, is in here across both of those. So then, what I do is I create what’s called a business schema.

34:06
Now, the business schema, you can think of it as a view, or just a friendly name for a view of your data, right?

34:17
The traditional kind of, I would call it, a star schema, right?

34:23
So you want to say online store.

34:27
And then what I would go through here is I would just pull in the columns that I want to bring in from that physical schema or schemas, in this case, my online store, and my weather file, OK.

34:44
Let’s just a second here, alright, then, so what I’m going to do this is I’m going to go ahead and I’m going to add a new view.

34:52
And I can folder that.

34:53
So if I wanted to, I could go ahead and I can go ahead and?

35:02
And create folders for that, for a safe products and time, and stuff like that.

35:05
But I’m just going to call this sales, whether, because, in the, again, in the interest of time, and then I’m going to go ahead and pull in my online store schema, and.

35:23
Or whether.

35:26
Then, from here, I can just go ahead and I can throw in, right, here’s my where.

35:32
Here’s my country region name, Right? So here’s all my, my geo information.

35:38
And then maybe I want to throw in my customer ID, and I want to throw in, obviously, I want my product information for my product ID, And I want my Product category name.

35:58
And I’m going to go all the way through all of these things and just narrow this down into just the information. I want to expose to my users, right?

36:08
Maybe now I want my max temperature from my weather file and my zip code. And so you can see, I can go and change this, make this friendly. I can go ahead, and I can add calculations, as well.

36:20
So if there’s something that’s not in my data, I can add these. I can do that at the physical schema level.

36:27
But I can also do it here with a very handy formula builder that allows me to access all the fields and a whole set of formulas, they give me, write aggregate, Boolean functions, conditional statements, conversion, and pull various variables and stuff like that. So there’s a very robust calculation engine here that you can leverage.

36:46
So once I’ve created that, and I’m actually going to pull in Martha Stewart on you again here, I’m just going to jump out of this and not create it.

36:54
And I’m going to use one that’s already been finalized, so you can see what this looks like.

36:59
Right, so in here, I’ve got a calculation that has my calculation that calculates my profit on the fly. It’s got data from my weather, and then it has information from the online store.

37:11
OK, and so now this, when I go explore data from that, this is a lot more manageable.

37:18
Again, we’re probably even better if we broke it down into the, all the various components, But now, I can already start going ahead and building this visualization. So the tool is very straightforward, again, browser based. I just pull in the data set that I want to use.

37:34
There’s a wide array of visualization types from your standard sort of tables too bar, line charts, and maps, and funnels, all the way down to things like KPIs and engages versus extensible through the use of an SDK.

37:49
So, I’m going to go ahead and I’m going to pull in, I’m just going to build a little bit of this, right.

37:54
So I’m going to do the revenue, and I’m going to do my order quantity.

37:58
And then I’m going to pull in my product ID, and my Sales Order number.

38:08
So, in this case, Revenue, I wanted to sum up, and maybe I want to change that to rounded and then order quantity. I’m going to do the same thing. I still want to sum that and I want that to be rounded.

38:21
And then, my product ID that’s my distinct skews, So I’m actually going to change the label of that.

38:30
And I’m going to do a count distinct on that, And then my sales order numbers.

38:36
I’m going to change that to account, and that’s going to be rounded, and so you can start to see how this takes shape already, right? And I’m just going to call this, I’m not actually going to label this one because it’s just KPIs. And I’m going to save this thing.

38:48
All is my Senturus, demo MR put that in content.

39:03
OK, and it takes me into my insight now, I’m going to add a couple more insights here just to show you kind of how this process works, right?

39:10
So, the next one, I’m going to go ahead and build a bubble map.

39:15
And as you might expect, I’m going to put in my state information. There’s a nice little search bar it gives me no sample data and shows me the column lineage and stuff like that. I’m going to throw that in here along with my revenue.

39:42
And then I’m going to change the map type here to a State map.

39:50
And just like that, I have my map, right? And so there’s a whole bunch of formatting I can do here, right? I can do conditional formatting on things and, and do stuff like that. I’m just going to change this Insight title and call it Revenue by Geo.

40:06
Save that and then save that out here, you can add tool tips to this and all kinds of stuff, right?

40:12
And there, I have that insight, so I can go ahead and keep adding insights to this, right, I’m going to add in this case.

40:18
So I want to do a I’ll do a combo dual axis here Just to show you what this looks like I’m going to do my average temperature and My order quantity.

40:39
Right.

40:48
And what I can do here is change that orient or quantity, and I get those. So I get my revenue at my order quantity as it pertains to temperature.

40:59
Right? And then I can change the colors of that, and then add this to this visualization.

41:04
I want to do one more here, just to show you what this looks like, and I’m going to, I’m going to do a cross tab.

41:11
And the reason I want to show this one is because it’s, um, it’s pretty computationally intensive, right?

41:17
And this is something, again, we’re operating on, you know, 25 million records, and then 10 million records, and the other 1, 9 million sales orders and whatnot.

41:25
So I’m going to put my month name in here, I’m going to put my product category.

41:35
And subcategory in the rows, and I’m going to grab, again, my friends revenue.

41:42
You don’t want to.

41:48
The little lag here.

41:52
And so you can see here that very quickly, that almost instantaneously populates it with all my information across all my product categories and subcategories. And I can add that to my visualization.

42:10
And immediately, this visualization is dynamic from my high level KPIs from my map. Again, and selecting New York drills down and narrows down all these records.

42:22
There’s nothing I have to do to bring all that information together, right. So I just continue adding visualizations to this, and then it’s there for everyone to consume.

42:34
Now, the last thing I’ll show you is that I can actually share the access to this or scheduled delivery. So if I want to share this with colleagues, they can create their own bookmark versions of it, or whatever. I can deliver it to somebody, or I can schedule delivery of that. I can even embed these dashboards in other applications like Salesforce for example.

43:59
And faster time to insights with rapid incremental, refreshes of the alternation of complex, and fragile, ETL or ELT processes.

44:09
So, we have some more links. If you’re interested in learning more about Incorta, please head over to some of these links. Again, these will be in that PDF that we provide.

44:18
We also encourage you to sign up for a free cloud instance. The link is in there. You can test drive it for free. It includes sample data and you can try out those data apps that we talked about.

44:27
We encourage you to go out to the community and sign up for an account, there.

44:30
It’s a very rich and growing community, and we offer free education.

44:33
There’s some great, free, online, self paced courses, as well as instructor led courses. So go and take advantage of that.

44:44
And with that, I’m going to hand it over to Steve for the about Senturus section.

44:52
Alright! Thank you, Mike. And that was perfect timing for wrapping up your demo, because your audio was starting to fade in and out.

45:07
Just quick note here about Senturus can help If you’re interested in Incorta, no matter what your data sources, or your existing infrastructure, we can help you facilitate seamless integration of Incorta into your existing environments. We can help you determine a phased implementation strategy. We can assist with the data source migration.

45:33
We can help you connect Power BI and Tableau to Cognos all of which we have a great deal of experience with. And we can help you just generally with creation of dashboards and reports.

45:53
Bit about additional resources that are available. If you visit Senturus.com, we have a wealth of information. We’ve been in this business for a very long time, and have slide decks from past webinars, tips and tricks: how to blog posts.

46:11
So, I encourage you to visit Senturus.com and take a look at our free Knowledge Center.

46:21
Overview of some upcoming events, so we do continue to have a series of ongoing webinars.

46:28
So, if you’re interested in joining us again, in the weeks ahead, we have introduction to the Azure Data Platform coming up on August 18th Intro to data visualization with Python coming up at the beginning of September.

46:42
And at the tail end of September, we’ll be covering what’s new in Cognos 11.2.3 for those of you who are Cognos shops.

46:53
Little quick background on Senturus, we concentrate on BI modernizations and migrations across the entire BI stock.

47:01
We provide a full spectrum of services including training and consulting and implementation work. We’re experts in Power BI, Cognos and Tableau, and we particularly shine in assisting with hybrid environments.

47:22
We’ve been in business for over 20 years inching up towards 1500 clients at 3000 plus projects. We’ve helped clients, both large and small, and we’d be happy to help you out as well.

47:38
We are currently hiring, so if you happen to be looking for a new opportunity, we’re currently looking for an analytics and BI trainer, senior Microsoft BI consultant, and a management consultant. So, if you’re interested in any of those opportunities, please reach out to us that the jobs, that Senturus.com, You can also find more information on our careers page on the Senturus website.

48:06
And as promised, we’re going to do some live Q&A here. There are a few questions I see in the question panel. One of the questions was answered, and this might be a problem only for me because I’m on a Mac. But I can’t actually see the full answer, so I can actually see all the questions coming up.

48:46
Senturus can help you figure out where Incorta of plugs into your environment.

48:56
It’s, it’s really a game changer for a lot of that operational agile reporting, Closing the books, finance data hub, things of that nature.

49:22
So, someone asked if in court is similar to at scale and we get a lot of that.

49:30
It’s certainly similar in one way but it’s really more of a semantic layer solution and a I think it does some query acceleration.

49:43
But Incorta is really a comprehensive as you saw data platform from ingestion to that schema creation to the business layer, the semantic layer, and the visualization and distribution. So we really have all of that in one platform.

50:02
That kind of segues nicely into.

50:04
And if again, yeah, if you do want to talk about that in greater detail, like how it compares to other platforms, I’d invite you to reach out to Senturus.

50:16
Right, and don’t really go look at the Gartner report. We’re in the two Gardner reports, one for accelerators and the as well as business analytics.

50:27
So that’s what’s really interesting about Incorta.

50:30
You won’t find other companies that are in both of those.

50:37
The short answer is no, they are semantic layer. We are a data platform.

50:52
And, I know there are issues with scaling. so, definitely, we are there scaling is a bigger thing for us, but, yeah, thank you for bringing up that gotten the reports anyway.

51:05
And then, someone asked about Power BI, and I don’t have it on here right now, in the interest of time.

51:12
Like Mazen said at the beginning, there’s a lot everybody loves their Tableau and Power BI is in their collection Excel. And we have connectivity to all of that.

51:26
We use SQL, I layer, it’s a postgres connection that allows you to access all that information the schemas and the business schemas, so we don’t pry those tools out of your hands. We simply empower them with better data access.

51:47
If anybody wants to see a Power BI demo, definitely reach out to our friends at Senturus and we can, together arrange for a Power BI demo.

51:59
June, thanks for asking this. ERP are not exactly known for the good name of their tables and fields.

52:09
So, how do you handle generic code combinations? Segment one, etcetera?

52:15
And this is where the blueprints really help with these well-known systems, especially.

52:21
So, we have blueprints for Oracle, EBS for Oracle Cloud, for JD Edwards, or SAP And, you know, so, they’re there, table, table names are really incomprehensible and generally in German.

52:38
And so, we have all that sort of stuff and that’s really where, that semantic layer is super important, so, we pull in all the data with all those funky column names. And you can leave that as it is. But then, we pull in, we create business views that allow you to look at things from a business perspective. IEGLARAP. Things of that nature, right?

52:59
So hopefully that answers your question, definitely check out.

53:01
If you’re in our cloud trial, you can even pull in a blueprint and see what it looks like.

53:09
And we’re always expanding that, and we’re happy to talk to you a little talk to you about that.

53:13
And, again, that’s something I think that so tourists can certainly, I’ll help you with as well and how that would fit into your environment, potentially.

53:24
With the big name, ERP our connectors take care of that. We definitely recognize the naming, and so on. So hopefully, the connectors are taking care of it, but I’ve also worked with big customers that actually have their own in house built database, which is a little different.

53:45
We were able to go around it by, we have a small, quick mapping project. Once that is done, the rest is history. So hopefully connectors will take care of that issue.

53:55
Yeah. That’s a good point.

53:57
So that the blueprints are not black boxes. Do you can really see what they’re, what’s in them, and they’re fully extensible.

54:03
So if there’s a module or something that we haven’t covered, custom fields that you have, what we will recognize those, you can pull those in all the way down to the, the SQL level. And then, like I said, there’s 240 plus connectors.

54:16
For example, I just created some connectivity to a rest API and pulled in some data from a system called WEX, Didn’t have a specific connector for it, and I was able to do that in a pretty fair amount, short amount of time.

54:34
But June, you also didn’t mention on your question here, Do you do it at a quarter?

54:39
Do you do that Semantic Layer and Incorta or a separate semantic layer like framework manager in Cognos or Power BI datasets?

54:47
The answer is you could, but, generally, no, you would want to, because you want a semantic layer in Power BI and you pull an import or something, and then you’re creating another copy of that data. And there are scalability issues because, like I showed you that EBS had a billion records in it.

55:10
We don’t even break a sweat at that level when you start querying that sort of stuff. We’re trying to pull that and having Power BI do that query, instead of having the letting the incorta engine do it.

55:21
You could run into challenges. And I know for a fact, Cognos will struggle with them mightily.

55:27
So generally you would want to do that in Incorta.

55:34
Thing I wanted to note also while we’re speaking about Incorta and data sources and so on, and different applications, when feedback we get from customers when Incorta that runs, the actual application or the database does not slow down. It does not affect the performance of the data source wherever that is, because everything is done an incorta itself. So, that is a symptom probably worth mentioning as well.

55:58
The combination of the rapid ingest, the direct data mapping into those nicely compressed parquet files and then the leveraging of in memory analytics makes Incorta super fast against large volumes of data and that’s very unique.

56:18
Then the last question I see here is, Can you do SAED two with Incorta, so a type two slowly changing dimension?

56:24
That’s actually a great question, because we do the full loads, which are required initially, and then we do incremental loads to reduce the load on that source system.

56:39
But we also allow you to do what we call snapshotted against the system, either dense or sparse. And the dense snapshots are kind of taking everything, at a point in time, writing daily.

56:52
And the sparse ones are actually handling some of that, just the incremental changes and using those snapshots with like a modified date, etcetera, etcetera. Current flag, things like that you can, incorta, you can implement a slowly changing dimension attached to slowly changing dimension and caught up yet.

57:10
So there’s a lot of flexibility there that kind of falls in the same realm, as it’s similar to what we talk about, where you did some of that machine learning and things like that.

57:18
We use our Spark engine, and you can use Spark Scala are to do machine learning to do complex transformations if you need to do all that sort of stuff within incorta.

57:55
Yep, so the materialized views give you this infinite flexibility.

57:59
So if you want to know about more about that again, that’s something we can follow up with you.

58:19
Oh, we have one quick question that Rahul asked for your data and it does, it pulls the data over, but just pulls one straight copy over, and then you can do incremental loads on top of that. So we do copied the data one time.

58:36
And then for everyone else, go to Incorta.com and get a demo instance to check it out, but definitely work with Senturus, because they can guide you through the process.

58:52
So again, thanks to Senturus for having us on here, and Steve, I’ll let you wrap it up.

58:57
All right, well, thank you, Mike, and Mazen, appreciate both of you being here today to share some of the details.

59:10
Thank you, everybody, for attending. Feel free to reach out to us, either on the web, by phone, or by e-mail. We’re happy to talk to you. Scott Felten also dropped his numbers into the chat window. You can reach out to Scott anytime. If you’d like to set up a demo that Incorta talked about Senturus can help you with an Incorta implementation.  We look forward to seeing you, on a future Senturus webinar.

Read more