Easily Connect Power BI & Tableau to Cognos Data

×

Michael Weinhauer
Director
Senturus, Inc.

In addition to BI consulting, Michael’s team created the Senturus Analytics Connector, which lets Tableau and Power BI use Cognos as a data source. Michael has been designing, delivering and selling analytics solutions for over 27 years. Prior to Senturus, he gained a wealth of hands-on, practical BI and big data experience at Oracle, IBM and SAP.

Read more

×

Welcome to the latest instalment of the Senturus Knowledge Series. Today, we are going to be presenting on the topic of how you can easily connect Power BI and Tableau to your Cognos data.

0:45
Before we get into the meat of the webinar, some housekeeping, while we have the microphones of our of our participants muted out of consideration for the presenter, we encourage you to ask questions and submit them through the GoToWebinar Question pane, as shown in the slide.

1:07
And we’ll generally wait till the end of the presentation to answer those questions.

1:12
And I encourage you to hold off on questions too until after we go through a frequently asked questions section that will answer many of your questions I’m confident.

1:24
And if we’re unable to for some reason cover all of the questions owing to time constraints or we just don’t know the answer right now, we’ll get back to you and post a written response document that we will post on Senturus.com.

1:39
Speaking of Senturus.com, everyone always asks, can I get a copy of today’s presentation.

1:44
And the answer is, absolutely.

1:48
You can head over to the aforementioned Senturus.com, a screenshot of which we have, our snazzy newly, refreshed website here.

2:00
And go to the resources tab and choose the renamed Knowledge Center or you can just click on the link that’s been posted to the chat window and go directly to it.

2:12
Today’s agenda, we’ll do a brief introduction of the speaker. We will then get into an overview of the Analytics Connector followed by a demo.

2:24
Then please stick around afterward for some pretty much always free, great additional resources, a brief overview of Senturus and then, again, the aforementioned, Q&A session.

2:38
So by introduction, my name is Michael

2:41
I am a director here at Senturus, among my many hats, I host these webinars and in some cases deliver them.

2:49
And in this case, I’m doing both as I am product manager for the Senturus Analytics Connector.

3:00
As usual, we generally like to get our finger on the pulse of the audience, so we’ll ask you a couple, watch, couple of polls here to ask you some questions.

3:10
First one, It’s a multiple choice question.

3:14
What BI platforms are you using in your organization are using Cognos Power BI, Tableau, something else?

3:24
Give me some time to get that in.

3:31
When I close it out, I’ll share the results back with you all, so you can see what the makeup of our audience is here today.

3:40
Great.

3:41
Close to 90% here.

3:43
I’m going to close out, share it back, and we can see here that it’s no surprise there, pretty much everyone here is running Cognos, then fairly evenly split little edge to Power BI there 50, 50, and then 13% are running other.

4:00
Thank you for sharing that information with us.

4:02
And then, the next question is, for those of you that are running tableau or Power BI.

4:12
How large is that user community in your organization? That’s just a single select Is less than 50 users.

4:19
Between, do you want in 250, or more than 250, or, you know, you don’t know.

4:26
So we’ll go ahead and get those answers in.

4:42
OK, I’m going to go ahead and close that out. Share this back.

4:47
So, about a third of you are on the smaller side, the third on the medium side, and about 20% bigger, Another 20% don’t know, Gats, pretty, pretty interesting, or pretty even distribution. So, Thanks for sharing that information.

5:06
I’ll give you an overview of the Analytics Connector.

5:13
So, there’s this notion of bimodal BI that was coined by a term that was coined by Gartner, back in about 2015.

5:27
And, those of you who were around then, might remember that there were the big players like Cognos that were seeking to become the standard and organizations with the idea of bringing a single version of the truth to, to an organization providing BI two.

5:46
Different types of personas from everybody, from professional report developers to enabling self-service and dashboards and scorecards, saying everything and everything in between.

5:58
And, with the advent of tools like Power BI and Tableau, they started encroaching. They started encroaching on those big players markets, right?

6:14
So, the term came around where Gartner started asking their clients are, do you have a mode? one which was generally referred to as kind of your Cognos or Business Objects or MicroStrategy?

6:27
For predictable, well understood areas, right?

6:31
So, you have, you know, finance and HR and areas that you’ve got pretty well understood data models and things like that.

6:37
Then, they asked the question, Do you have two modes to your organization?

6:42
Second one is more exploratory, and experimental, and innovative, used to solve new problems, and, that typically was regarded as the tableau’s and the Power BI and the click tax, et cetera, of the world.

6:54
Well, the interesting thing was this is the 2016 and 2017 responses.

6:59
And you can see the exact question that was asked here.

7:03
The interesting thing about that was that, in 2016, more than a third, almost 40% said that they did, in fact, have two parts to its IT organization.

7:17
So, in other words, they were using Cognos and Tableau, Cognos and Power BI Cognos and maybe, Cognos Tableau and Power BI. Who knows?

7:25
In big organizations, you tend to have a lot of different BI tools.

7:30
A standard is more the exception, but importantly, that it grew significantly by about 5% to, you know, you’re approaching half here in 2017, which, you know, I wouldn’t have guessed now, they didn’t, I’ve never seen any data for 2018 and beyond, But, our hunch is that, is that this is likely grown.

7:52
I would imagine significantly by this point. But that, in and of itself, doesn’t matter, right? So, so great. So, companies are running a couple of different BI tools.

7:59
Big deal, well, the more important thing is that the companies that are using it successfully would appear to be performing better.

8:07
So, it’s fairly well established and documented that organizations that leverage analytics effectively perform better.

8:15
You can look at organizations that make better decisions with their and leverage their data, outperform their competitors.

8:22
And this has been demonstrated.

8:25
And this would seem to indicate that, doing that effectively with the right tool for the job, if you will.

8:33
We, even, if that’s multiple tools, helps these companies perform even better.

8:38
OK, so, you’re thinking, Oh, great, Mike, but what does that have to do with your Connector?

8:44
The thing is that you’ve all invested and set up 97% Cognos, so, you Cognos folks will recognize this if you’re an end user or, someone who, who hasn’t done any of the modeling, you might not but, this is a snapshot of a Cognos Framework Manager model and that’s that tool that really was the foundation of Cognos and it allowed you to model data and present it in an analysis ready data analysis ready format.

9:13
Turn complex data sources into user friendly measures and dimensions that can, then, be consumed across the organization and secured all that good stuff.

9:23
So, it was sort of a highly curated, governed set of data, and these, often times, a lot of work to create, a lot of human capital, a lot of actual dollar investment, too, to get these things, to get these things set up, deploy them, so, there’s a heavy investment.

9:41
And, they generally work really, well.

9:45
The challenge with those, though, is that with the adoption of tools like Tableau and Power BI, they have no access. They had no access to those data assets.

9:58
So, what did they do? Well, then, they would use Cognos as an ETL tool.

10:03
So, we definitely, I’ve seen tons of clients who, if you look at their Cognos audits, what are their biggest reports? I mean, their top five, their top 10 reports.

10:15
Big old CSV or Excel dumps big tables of data, that guess what, they’re dumping them out of Cognos and they’re pulling them into Tableau or Power BI or one of these other tools.

10:29
Which presents a whole set of challenges, right?

10:33
The, generally, that’s one of those, that’s nightmare fuel for folks who, for sox, compliance, GDPR, et cetera, PII, data breaches, having big flat files with God knows what in them that are sort of floating around your organization, especially with covert and remote work, and people working on a coffee shops and stuff like that.

10:54
The other challenge is that the, the, the other option was, is they’d be like, Well, we’re just going to go recreate this model in and Tableau or Power BI.

11:04
And that’s oftentimes easier said than done, right.

11:08
If you have transactional systems, they’ve got tens, thousands, tens of thousands of tables, of which these users really don’t know the relationships and don’t understand them, So doing that wrong is not only possible, it’s I’d argue, likely, not to mentioned.

11:26
Tools like Tableau only recently introduced an abstraction layer versus sort of flattening out into one table.

11:33
So there were scenarios that you had you’d be very challenged, just sort of even model in Tableau and Power BI had some challenges too. It’s a little more robust in that department. But regardless, it’s, it’s at a minimum redundant.

11:4
So there’s a huge waste of time where your very valuable analysts are remodeling data.

11:53
That’s very time consuming.

11:56
You can compromise the data integrity because of the things I mentioned here.

12:00
And you have metadata layers, then residing in different environments that you have to maintain.

12:08
And again, that can be incorrect, the data security isn’t a plat on them, oftentimes, with framework manager models or data modules, you’ve applied security to them.

12:19
one of framework manager’s big strengths is the very granular, complex security scenarios that you can model in it, and when you go remodel that or you pull a flat file, that security is kind of it’s it’s gone.

12:36
So you have to either remodel it or it gets skipped, or you re-apply it, and it takes time and an effort, and can be done wrong.

12:45
And then you can always, you always end up with data silos.

12:48
So if you remodel the data and one of the, the strength of these tools is also, it’s weaknesses, Alright?

12:59
The fact that any Tableau user, any Power BI user, can go in and connect to a data source or a flat file, pull it in and model it, kind of  however they want.

13:12
And, Bob, from accounting Susy from HR, Kathy, from supply chain, model at different ways.

13:21
You end up with very different answers that aren’t necessarily accurate and make decisions off of those. So, you’re really kind of circumventing that single version of the truth. Which is the reason that most of you folks that are still using Cognos are invested in it.

13:37
So, we saw that as a real gap and an opportunity.

13:42
And so, we created the Senturus Analytics Connector to give those visual analytics users direct access to those great Cognos data assets and enabling you to just leverage those nicely modeled datasets in your visualizations without having to do a whole bunch of modeling.

14:05
So, this is a visual of what it looks like in Power BI.

14:08
This is, uh, screenshot of the browser interface, right?

14:13
So, if I’m creating a report Cognos, I’ve got my nice measures and then my dimensions all modeled and Cognos against a data source, which by the way, is about 75 tables and is a very complex OLTP source, so remodeling that.

14:28
It is truly non-trivial in one of these environments.

14:31
So instead of doing that, and getting it wrong rate calculations, and stuff like that, I can just pull in those measures and dimensions, the ones that I want, then go use them directly in my Power BI Visualization, in this case.

14:48
Or pull it into Tableau.

14:50
So you can see, we mapped to the, we mapped the database to the package, or the data module, the namespaces to the schemas, and the query subjects, and query items. Just become tables and fields.

15:05
And you can also access reports.

15:07
So if you have reports that have information that’s not in your data model, in your framework manager model or data modules, it’s maybe a skinny down steadier data. Maybe there’s calculations in there you want to use, or logic you haven’t pushed upstream into that model. You can access those.

15:26
You can even pull in prompts, and use those in your visualizations, and so you don’t have to sacrifice the integrity and the security of your data to use the tool that, you know, you’ve come to love and want to use.

15:43
So these are just screenshots of, this is a Tableau dashboard. And this is a click tech dashboard, and this is a Power BI dashboard. And I’ll build some of these for you in just a minute.

15:58
Then really importantly is, you can then publish these up to Tableau server or power BI.com deploy them across your organization.

16:14
So you only have to pull that in once and set it up. And then you can publish it.

16:20
And people can use that content, or they can create their own visualizations off that same model data and get consistent results across tools across the organization.

16:31
It’s also fundamentally enable some really different architectures as well, because you can extend those, right?

16:40
You can start with this kind of sanction dataset that you have, that’s maybe here, sales information around revenue and profits and whatnot.

16:48
And I can augment it with another data source that pulls in, maybe quotas or something.

16:53
And I can align that with that data.

16:55
So you can do all the great things you could do around, extending those data modules, start started extending those data models in the visualization tools, But you’re starting with that nice section set of data.

17:08
So, instead of all the risks around pulling flat files and whatnot, can, instead have faster and more accurate reporting, because your analysts aren’t spending much time remodeling or pulling flat files.

17:19
They’re doing actual analysis.

17:22
You reduce the risk and the cost around change management because changes are handled again in the framework manager models, and you pull them in, which eliminates costly and time consuming changes to individual metadata models in Power BI or Tableau.

17:39
You have, you see, increased trust in the data, because you have audible, auditable, govern data, and Cognos.

17:48
Again, that single version of the truth, So your users trust it, and that’s one of the key tenants of adoption of BI, tools, is they have to trust the data, and it has to perform right.

18:02
Everything else is secondary.

18:04
You vastly improve security.

18:07
You can’t really overstate this, again, in this environment you.

18:10
It reduce the risks associated with the use of unsecured data sources, flat files and Excel spreadsheets that are floating around, and eliminates the redundant application of security because it, we literally mimic a user in Cognos and leverage the security that supplied to that user from the top-down Ray. Can they access Cognos? Can they see the package or the data module?

18:34
Any row level, column level security that you applied in the data module or the framework manager model is automatically applied as by that user that’s logged in?

18:44
Cognos doesn’t really know the difference.

18:47
And you optimize your performance because you avoid the pitfalls of poorly model data, so, if that user goes out and instead joins the column the wrong way.

18:57
And it’s not indexed, or partition the right way, You can see poor performance. So you avoid that by using the model that you know.

19:05
You’ve already vetted with your DBAs, and it you know, that it works.

19:11
So what’s new?

19:13
In version 5.1, our latest release, we’ve enabled SSO using Windows authentication using SQL server connection.

19:24
So, historically, we’ve used an ODBC connection and that’s still an option.

19:30
You can still use that sort of legacy connection, but now we’ve, with Version five, we implemented SQL Server using their native TTS protocol.

19:40
And we did that for a couple of reasons.

19:43
one of them was so that we could enable things like Single Sign On, and also to eliminate some of the need for client installs, or custom connectors in the case of Power BI, which really streamlines the administration.

20:00
Then, with regard to data modules, we enhance those by adding support for standalone calculations, and for navigation paths, which are basically hierarchies in data modules.

20:13
So with that, I want to give you a demonstration of what this looks like.

20:19
So I’m jumping over to my, my remote desktop machine that has Tableau desktop on it.

20:28
And I’m actually going to connect. So those of you who’ve seen these before you, unless I usually would connect other databases ODBC.

20:34
So, instead on this one, I’m going to, I use the SQL Server connection, so, again, no special drivers required. All I need to do is click on SQL Server.

20:45
I’m pointing this to my connector, serve, my Senturus Analytics connector server that is running on this internal BI one machine, and I’m going to login using my credentials here, and this is my Cognos credential, so, this isn’t actually a Microsoft SQL Server, right?

21:04
It’s just using it’s treating Cognos as a database using Microsoft SQL Server protocol, so, and, the that’s the great outdoor sales query package.

21:14
It then connects to Cognos logs me in and shows me all of the different tables that are in the various namespaces in the Cognos Go sales package.

21:27
So, I’m just going to choose some of the measures in here. So these are my sales measures.

21:34
And then, I’m going to grab branch order method.

21:44
And you can see how these exist products and several these exist in different namespaces there. So you can see we’ve got no take them. So that’s the inventory one product forecast.

21:54
And I’m kind of focusing on this, the sales query, once I’m pulling all of those from the same one, you can mix and match those as long as it works within the model.

22:14
OK, so it’s just a simple double click exercise, where I pull those in.

22:19
And now you can see, I have all of those elements pulled into Tableau, OK. So, I’ve got my branch. I order method, my products, my time.

22:29
Then, what Tableau does is it puts anything that’s a number into under Measures.

22:38
And, so, I’m just going to do this really quickly, and pull these in. I’m going to, I’m going to just click all of these.

22:45
These aren’t measures. These are dimensions. I just drag them up.

22:50
And then, I’m pretty much done with, you know, most of the, the modeling I would need to do in Tableau.

22:59
Now, we do create these dummy link columns, and that’s because Tableau, all these tools require that you create relationships between any tables that you bring in. And it just takes these Cognos query subjects or tables, right?

23:14
In the presentation layer, and in a model, design for best practices. And then, the join criteria are in there, right?

23:23
So, say, my sales table, doesn’t have a time key or a product here, order method, key or a branch key, They don’t exist.

23:30
So, we create those trip, that AAA link, as a dummy link, to kind of  pick out Tableau or Power BI. And then we ignore it at runtime and we just pass it back to Cognos.

23:41
Cognos ignores it, and we get the results set back.

23:45
And keep in mind so.

23:46
remodeling this again, this is about 75 tables, OLTP data sources should be non trivial to remodel, especially for your average user. And, I only have to do this once.

23:59
Alright, so, I’m going to pull this stuff in.

24:00
And, once I publish this to Tableau server, then I’m done with it.

24:04
I never really have to touch it again unless I go and change the model.

24:09
Now, the rest of this is really a Tableau exercise and, I’m going to connect live to Cognos. I’m issuing a query to Cognos. Every time I do this, right? I’m going to doubleclick country.

24:19
It’s pulled everything into this little map here.

24:23
And then I’m going to add.

24:28
I seem to have a connection issue here.

24:35
Sorry, guys, bear with me a second. Every once in a while I lose this. It comes back pretty quick.

24:49
OK, sorry about that. So, it’s done a live query to Cognos and it’s created my map for me.

24:56
And then, I’m just going to But Revenue up here on size, and I’ll put profit up here on color.

25:07
And I’m just going to give this guy a name.

25:15
I’m going to call it Cognos Geo Sales. Dashboard.

25:31
I’ll add some numbers to it and I’ll turn up the size on that a little bit just so it jumps out a little bit more. But just show you how we can take advantage of the. A lot of tableau’s great sort of built-in functionality, right? So this is a map.

25:44
Then you, can, you show me, so I can say order method type.

25:49
Then I’ll do revenue plan revenue and I can go over to show me and choose the bullet chart.

25:57
All right, that’s going to create that.

25:59
Then, I can also take advantage of time series analysis. Alright? So I can go over here and I can just take date for up on columns. And Tableau does what it does, right? It takes that date field, knows.

26:12
It, knows that it’s a date, rolls it up into the hierarchy for me, so I don’t have to really, you know, do much around that.

26:20
Then I can take quantity.

26:40
Great.

26:50
There we go.

26:53
So I’m just going to create a line chart with that, and then put product line over on color. And then I can drill on this, right, I’m making contiguous, continue us.

27:03
And then I’ve got a nice timeshares analysis of my quantities over time.

27:07
Then, I kind of assemble all these into a dashboard up top.

27:14
I’ll drop the bullet chart down here, and I’ll put the line chart down here.

27:20
Alright, so then, the rest of this is really formatting, right? So I can play around with this and do whatever I want to do with it. Then, I would just save this.

27:38
And two, then deploy it.

27:42
I would just click Publish Workbook, connect to our Tableau server.

28:05
I should’ve done that before.

28:07
And that’s just a publishing exercise now, The thing I wanted to show you in here, is, when I publish this data source, is, I can say, I want to publish this separately.

28:17
Then, authentication, you can either embed, impersonate or prompt you user.

28:23
So, if you want to enforce that row level security, I can do that and have Tableau prompt me. Now, I’m not going to publish this right now because it takes a little time and I kind of  want to  get moving. So, I’m going to jump over here.

28:39
to Tableau.

28:42
And this is the visualization in Tableau server that I’ve already published.

28:47
Alright, so it’s got my map and it’s got my bullet chart.

28:49
It’s got my, my line chart.

28:52
I’ve had to login to, I’m logged in as Steve Re Pitman here, and I can interact with this dashboard.

29:01
Now, if I want to either create content, or, or edit this, I have that data source available to me.

29:08
So, I’m just going to show you the edit functionality, And it’s going to prompt me.

29:29
Theoretically.

29:35
So, it was prompt me for my Cognos credentials, here.

29:39
I didn’t embed them purposely, I’m not sure if goto meetings having it having an impact on this.

29:46
I’ll give it another second, and then, I’m going to move on.

29:50
But, the idea being, here, that, OK, here we go. So, analytics. So, it’s, this is, again, my Cognos user, right? So, it’s asking me to login.

30:04
And, again, I don’t, usually, I could check that, just save it, but I’m not going to because I wanted to show how it can enforce that.

30:10
And so, if you’ve got row level security that’s embedded in that model, that’s where there’s a lot of value in that.

30:16
And, it’s going to authenticate, and then, I can go ahead and create content and here’s that nice metadata model that I pulled in that’s coming directly from Cognos.

30:27
So, once I’ve done that once in Tableau Desktop, I publish it to server and now, the rest of the organization can we leverage this and, you know, the numbers a line up between Cognos and Tableau?

30:44
OK, so, I am going to free up some resources here by closing tableau and I’m going to show you how this works in Power BI.

30:55
So, Power BI for sitting there patiently, sorry for keeping you waiting.

31:00
I’ve got Power BI Desktop running here.

31:03
And per usually, you click the Get Data button, and so the legacy way you would do this, either you use DNS or use the custom connector that’s still available.

31:15
But here, I can use SQL Server database.

31:18
And then you don’t have any clients you have to use or anything like that.

31:21
I’m going to say, again, zero point two.

31:24
Why Cognos Mice interests Analytics Connector.

31:31
Server then, point to my database, which is, again, that Go sales. That great outdoor sales query model.

31:38
And I’m going to, it’s going to go login.

31:39
It’s saved my credentials because I already connect it to this once before, So that’s saved in Power BI Desktop. So, rest assured, it would, it will prompt you.

31:52
Then, Power BI models it a little bit differently actually organizes it a little bit better here.

31:57
I’m going to pull in the same things.

31:59
I’m going to go order method products and then I’m going to pull in, time can’t wait for the preview.

32:07
It’s going to issue a bunch of queries to Cognos that are going to get metadata information about those tables.

32:16
And it’s going to pull it into its metadata model.

32:22
Again, this is a one-time exercise.

32:25
So once I do that, I go to the model page, and you can see I have all of my query subjects that now show up as tables.

32:36
And I am just going to join these again.

32:41
There’s there aren’t relationships between these that are displayed in the presentation layer of this best practices model. So I have to use these AAA links that we create as dummy column’s. Power BI, unlike Tableau, doesn’t automatically link those columns, So I have to do this myself, but, and not a big deal.

33:00
Drag and drop, check assume referential integrity, anytime you’re creating something in Power BI.

33:07
If you know that those there is referential integrity between the tables, we recommend you check that. Because otherwise, Power BI will, by default, issue an outer join, which is a lot slower and more expensive.

33:18
It’ll create that relationship. I’m just going to do that for these other tables here real quickly.

33:40
Last time.

33:47
I’m done.

33:49
Now, I go back to the presentation.

33:52
Pain, we’re actually create my Power BI content.

33:56
And just like with Tableau, this becomes really a Power BI design exercise, so, I’m going to build more or less the same visualizations I’ve built over in Tableau, Power BI, so I’m I create this map.

34:12
And drag this over here, again, this is issuing Live Queries based upon that user that I’m logged into. So it’s going to do a revenue by country, again, recognizes the Geo stuff. It’s going to create a map for me.

34:23
Then I’m going to click down here.

34:26
I am going to put a matrix visualization on here and you can create hierarchies in Tableau and in power BI.

34:37
Harvey has made this a little more challenging in recent releases, but you just create a hierarchy, and it adds that element to it, and then I know that, Product Type.

34:48
I want to add that to my hierarchy, and, boom, I have a hierarchy. Alright, and then I can just add that to my visualization.

34:57
And I can say, Ad Revenue and then ad plant Revenue, and now I have a nice Dribble Hierarchy and Power BI that, oh, by the way, is interactive. So you see, you know, if I click on, if I hover over North America, I can see the revenue for just camping equipment.

35:16
If I click off of that, the revenue numbers changed because it’s filtered. So the cross filtering all works.

35:24
What I’m going to do here is add another visualization.

35:29
Sorry, I’m trying to shrink this.

35:33
Again, just to show you, you know, kind of a line chart and some of the time series stuff.

35:37
So with the line chart I’m going to throw in Year, cross my axis, and I’ll do my order method.

35:47
Type on Legend, And I’ll grab Quantity. Put it in the values.

35:56
OK, so then I’ve got a little timeline there.

35:58
Again, the cross filtering will work, and I can put in things like a slicer.

36:07
I do.

36:12
Let’s click over here first.

36:14
So I’m going to have a new visualization type there.

36:16
And I’m just going to filter this by year, so create a slicer now. That will filter all these visualizations.

36:23
Time Slicer.

36:27
Then, again, to publish this, I’m just going to save this thing.

36:44
I just click Publish.

36:50
Choose a workspace and I click Select. Again, this is a little time consuming. Interest of time.

36:55
I’m going to Martha Stewart, you guys are jumping over to the finished product.

37:00
So, again, this is now in on power BI.com. I’m in my browser.

37:05
And, I have the same visualization that I just created over on the desktop that I can now share it throughout my organization and interact with, right.

37:17
So, that’s all the same stuff.

37:19
Let’s go back and it’s, it’s querying Cognos and filtering this by camping equipment.

37:25
Then, again, that data source goes along with it.

37:29
So, now, if I want to actually edit this or create new content, there is the web desktop and this data source, so I can use this and create new visualizations.

37:41
I don’t really have to mess around with that ever.

37:45
So, that is kind of a demonstration of how you know, start from scratch, build a basic visualization in Tableau and power BI, then push that up and deploy it out to your organization.

38:02
Again, we do encourage you to ask questions, But I have some frequently asked questions here that might save you some time and typing, and I’ll run through those, and then we’ll get, we’ll get through the Q and A So, we frequently get asked about live versus direct query and extract versus import.

38:20
So, Tableaux supports, live, and extract, and you can do either.

38:26
And, with Power BI, you have direct query, or, you can create imports, and you can do both of those.

38:33
And I will throw out the caveat that, you know, if you’re connecting to your data warehouse, that’s 12 Terabytes, you know, be mindful of that.

38:42
And, you want to be smart about what you pull in, because there are, you know, there are limitations and you’ll have your Cognos admins calling you or it will fail because they have GOV’s on it.

38:53
But, you can definitely do both of those in terms of the supported Cognos platforms and versions, we support Cognos 10, 1 and 11 and higher, so, basically, you know, anything in the 11 range   11, 11.7, which is the mmm stable maintenance release for 1001. And, we support 1002.

39:15
We support both on-prem and cloud deployments. The only caveat there is you have to be able to access the dispatcher.

39:20
That’s how we communicate between the Connector and Cognos.

39:26
Supportive data sources so, this is what Cognos data sources does the Connector support, and we support relational databases.

39:34
Power Cubes, dimensionally modeled relational sources, dynamic cubes.

39:39
And, in the form of Framework Manager packages, data modules and or reports. We often get the question, Do you support data sets?

39:49
Not directly, but you can pull a data set into a data module, and we will support it that way.

40:00
Terms how long the Analytics Connector has been around, and where are we? It’s been around for five years now, in development for over six.

40:09
And we are currently on version 5.1.

40:12
As I mentioned before, the software requirements, it has a very small footprint, couple hundred megabytes of disk space.

40:19
Any sort of modern CPU with, but eight gigs of ram and 20 gigs disk will work. It doesn’t really impact the CPU that much. It’s more. It runs a JVM.

40:31
That consumes some ram, but it’s not a big resource hog.

40:40
Supported versions and platforms, we support Power BI Desktop June 2018 or later, usually at auto updates.

40:47
So I have a hard time imagining anyone isn’t on something much more recent than that.

40:52
Actually earlier versions might even work. This is just for custom connectors.

40:58
And then we support Tableau Desktop Server an online version 10 or later again, almost everyone is probably light-years ahead of that but they all pretty much work because it uses these standard protocols.

41:10
We run on the Connector will run on any Windows operating system, Windows 10, Windows 11.

41:17
And any of the Windows servers that are really kind of out there, 2012, 2016, 2019. You can also run it on Linux.

41:26
And we have instructions for how you do that and you can run it on Tableau, desktop, Mac OS X, So desired.

41:33
And so, where does the Connector live if it exists wherever connectivity to Cognos is required?

41:38
So, that would be Power BI Desktop or Tableau Desktop Tableau server and or the Power BI on premise data gateway.

41:48
We do, it is architected as a client server and I’ll show you some architecture diagrams in just a few minutes, and that’s how we recommend you deploy it.

41:58
It’s really think about it like a database driver, right?

42:00
You need the driver to connect to SQL Server or Oracle or your Salesforce or whatever it is You’re connecting to think of, we treat Cognos as a database, so just like you would need an Oracle driver or a SQL server driver on Tableau Desktop or Power BI Desktop.

42:14
Any of those servers, as well, and on the gateway.

42:18
If you’re communicating back to the firewall, we’re no different, but we have architected it as a client server.

42:24
So, you can have an Analytics Connector server that has the licensing, and the configuration, and all that stuff, on that server box.

42:34
Then, you just connect using that SQL Server connection. Which, which requires you to do nothing further, and you just point to it.

42:41
It doesn’t require that you install anything on the Cognos servers, right?

42:45
It’s just communicating over the this, the dispatcher, URI, dispatcher, you are eyes!

42:53
And, as you’re going for proof of concept, oftentimes, we have people that they’ll install it on their Cognos Dev server to try it out, especially if they have encrypted dispatcher’s.

43:05
It save some effort on the security configuration, on the connector side, But, nothing is technically required to install on the cognitive side.

43:14
When you’re thinking, OK, this sounds great, but how is it going to impact my Cognos environment, having like, Tableau or Power BI users out there? It’s a it’s a great question.

43:25
And you may have observed, or if you use these tools, you realize that people who use those tools are more interactive right there.

43:32
They’re playing around with things there, pivoting things, changing visualizations, and adding data, deleting data, removing fields, creating hierarchies, and stuff like that.

43:40
So, they’re more similar to what you’d call it an interactive report user on Cognos.

43:46
So, that can increase the load of the cognitive servers.

43:49
However, you may have also observed, or in your personal experience or in this demo, that the queries tend to be summarized, right?

43:58
You’re building visualizations, where you’re trying to say, what is my revenue by state or by country?

44:04
And I want to see, you know, which one is, or what’s my profitability, which one’s? the lowest. And drill into that.

44:10
I want to see, Where am I?

44:13
My planned revenue hasn’t hit my actual revenue and drill into that.

44:19
So, the, the high level queries that you’re issuing are summarized and then, you know, hopefully, you’re designing, it’s just that you drill in.

44:27
And, those queries are limited so, that reduces the load on Cognos.

44:32
So, the net of that is it’s very use case specific.

44:37
Best practices are important. Does it pull data in right away? You pull a field that it wants to pull it in.

44:46
You might not want to do that.

44:47
You might want to put some filters on first and build the visualization and turn off that auto loading and Tableau. For example, you know, apply filters and things like that.

44:56
So, it’s an important step to sort of ascertain what the net load is on the Cognos system and design your visualizations appropriately.

45:08
Licensing requirements.

45:09
If you’re doing live connections, then the users of your visualization tools need to be a Cognos Analytics user or equivalent, right, because it’s basically they’re logging into Cognos and impersonating a user that would be using Report Studio or creating dashboards or doing exploration. So, they need to see the metadata.

45:33
They need to be able to execute reports and whatnot.

45:37
If you’re doing extracts or imports, then the user creating those extracts or imports needs to be a Cognos Analytics user.

45:45
But then, once you create those, those are proprietary data sources to Tableau and or Power BI, either attached from Cognos.

45:53
So, the users accessing that don’t need a Cognos license, they do need a centrist Analytics Connector license. However, in terms of security, we automatically enforce the security defined within the Cognos environment, and that is a key.

46:09
The value proposition is that you are impersonating that user and it’s automatically enforce, you don’t have to do anything. So, can I get into Cognos? Can I see that package or that data module? Do I?

46:21
Are there rows or columns? I shouldn’t see.

46:24
That’s automatically handled.

46:25
So, the value of the Connector goes up considerably, the complexity of the underlying data source, the complexity of the modeling that’s been done, and the extent to which you’ve, you’ve built-in security into those models.

46:38
If you have a very simple data source that has no security put on the model.

46:45
The value of the Connector is, it isn’t as strong as it is with a complex model with security, embedded in it, that’s just saying it’s not valuable.

46:55
It’s just extremely valuable with the former there.

47:00
We don’t have any special modeling requirements, when it comes to, I should say, here, modeling requirements, just in general, because your data modules or your Framework Manager models, there’s no special requirements.

47:11
But, as I spoke to, and you saw earlier, the tools all require that joins exist between query subjects, you know, IE, the tables that you pull into Tableau, or Power BI.

47:22
And we create those dummy joins by default.

47:24
Now, that said, if you want to join it to something else, or if you have those relationships in your models, and you want to use the, quote, unquote, real columns, then, by all means, go ahead and use them. And we’ll respect those, it’s just if we see the dummy joints, it’s setting and we ignore it.

47:43
And that’s the default.

47:45
And usually, kind of what you want to do because, again, the point being you want to use the model as it exists in Cognos so that the numbers are the same and everything lines up between your visualization tools.

48:01
How do these tools access Cognos?

48:03
We can use just a direct SQL Server connection.

48:09
You also do have the option of ODBC.

48:12
So, with Power BI, you can create a GSN, or we they use Power BI, has this notion of custom connectors.

48:18
So, we have a, a custom Connector that you can use. Or, again, you can create GSN like, you would normally do with an ODBC source and use that support and maintenance.

48:30
We provide the support maintenance and upgrades and all that stuff that’s all included with your annual subscription.

48:38
So, a couple of architecture slides here, just so you can see how this works for Tableau from a standalone perspective.

48:45
So, say you have the full stack, the client, and server installed at each point.

48:52
So, Tableau Desktop communicates with our connector, which communicates to Cognos, returns a dataset back to Tableau.

49:00
You then publish to Tableau server.

49:02
Again, the full connector lives here.

49:05
You’re doing a live or an extract against that. It’s communicating back.

49:09
Returning a dataset back here.

49:11
When you publish a dataset to Tableau server, I can use Tableau Desktop to connect to a server based data source. So then, my desktop is no longer going through here.

49:23
It’s going to route through Tableau server, right? So that’s ideally what you want to do is, you’re not going to have all your Tableau Desktop redoing this.

49:31
They’re going to use that, that that source and you can no, you can govern that within Tableau, Right?

49:38
You put your, your Salesforce within the sales project, only the salespeople can see that Finance force, Finance, et cetera, et cetera.

49:47
If you’re using the recommended client server model, it doesn’t change a whole lot. It just introduces this server here.

49:54
And instead of having the full stack over here, I have these very lightweight clients.

49:59
Or if you’re using SQL Server, it’s the SQL Server connection. It’s really going through that and going directly to the server.

50:07
But if you’re using the client, it’s going to behave similarly.

50:11
But you’re centralizing the connection through the server, right?

50:15
This is dedicated environment, again.

50:17
That small VM that has the licensing in the configuration, so you’re not, you’re maintaining this, and this is super lightweight, or maybe non-existent in the case of the SQL Server connection.

50:29
For Power BI, it’s slightly different in that you have the Senturus.

50:39
These are the full client server, or the server, but the difference is that you have an on premise data gateway on premises. Data gateway is what Microsoft snap leave it to Microsoft.

50:54
Give it a name that’s cumbersome.

50:56
And, that’s your gateway that, that allows you to talk through the, the firewall back to on premise data sources. Right, Which, in this case, your Cognos environment.

51:05
So, the Connector lives there, as well or, communicates to the Senturus Analytics, Connect or server, communicates Cognos. Renders that backup, same thing where you start with the desktop.

51:16
Create a connection back to the packager, the data module, publish that up.

51:21
And then it’s going to route that way.

51:24
And then, likewise, your Power BI Desktop can always connect to Power BI, publish data sources, in which case, then it’s going to go through that server and through the on premise data gateway worth mentioning with Tableau. If you use Tableau Online, they have something that’s similar that’s called Tableau bidge.

51:43
And so, we can use the Tableau bridge, and it will do the same thing as the on premise data gateway, and then the Connector has to be able to communicate with that, or rather, the bridge needs to be able to connect through that.

51:57
All right, so, if you’re kind of going, Hey, this sounds like a something I could really use in my environment, a quick summary of what are some of the, what’s the return on investment, one of the big benefits, you could save tons of time and money.

52:12
That’s, that’s really the big one, because you people aren’t recreating metadata from scratch, which is, you know, creating all kinds of problems.

52:22
You’re not creating reports and dumping them out. In terms of flat files.

52:26
You’re not maintaining different metadata layers between tools, a really big one, that’s almost impossible to quantify as, reducing that, data security risk.

52:34
Not getting all those flat files, Excel spreadsheets floating around.

52:38
You get to leverage, from that, really maximize your cognizant investments.

52:40
You get to the reason you bought Cognos and the reason you’re still sticking with it in All likelihood is because you have that, you know, you’ve got a lot of great metadata assets there that just, you know, they work against this govern data. And you’ve got operational and other reports that you’re knocking, pixel perfect reports that you’re really not going to do in Power BI or Tableau in all likelihood. So, you’re keeping the tool, and you have these assets. You may as well use them.

53:07
And you can better ensure accurate and aligned business metrics.

53:11
So you’re not compromising the data integrity because you’re avoiding that data remodeling, and you’re avoiding the pitfalls of poorly modeled data.

53:21
So, stick around for the Q&A here. I’m sure there’s a few questions in the questions pane here.

53:26
Couple quick things about what’s coming up here.

53:28
And a few things where Senturus can help you.

53:31
So, whatever your organization is doing, if you’re, if you’re cutting the cord, right, if you’re migrating from Cognos to Power BI or Tableau, We can really help you here, because we are strong in all of these tools are our people, oftentimes, no.

53:46
All three of these tools, but pretty much all of them know at least two of these And it’s kind of  like having a translator, right? At the UN.

53:53
You need to be able to speak English and Russian or English and Italian, and we can do that.

53:59
That’s really key if you’re doing those migrations or, if you’re trying to figure out how to make these tools, you know, work together and coexist, which, I suspect, most of you who are keeping Cognos, I really kind of faced with the ladder versus the former.

54:12
And I like to have this little picture of our Senturus as digital artisans, who are both the they have the technical aptitude and the that expertise in bringing these visual tools to life.

54:25
So, you can really enable self-service, and take full advantage of these tools and be one of those Gardner top performing organizations leveraging bimodal BI.

54:37
Couple of customer case studies out there, there’s three of them out there that you go look at for yourself, you could see how these organizations are leveraging the connector to successfully enabled bimodal BI.

54:49
And you’ll see a demo similar to what I did. You often see a demo in their environment.

54:54
And if you want to try it out, we’re here to offer you a free 30 day trial that includes customer support, give you a trial key, give you access to download it, and we will walk you through getting it setup Takes less than an hour.

55:06
It’s really pretty easy and then you can see for yourself how it works.

55:10
If you want to do that, hit us up in the, in the chat or the question log, or you can go to the URL here, go under our Products section, they’re Analytics Connector and you’ll see a button Request A trial.

55:25
While you’re over on that snazzy new remodeled website, be sure to bookmark it because we have a ton of great stuff out there and we’ve been sharing our knowledge for well over a decade. So it’s got our great blog. It’s got other upcoming events, I’ll talk about that in a second.

55:43
Demos of are various products and tons of great BI knowledge.

55:49
In terms of upcoming events, we have two of them, one of them coming up in just a couple of weeks.

55:56
These generally occur at the same time, so although I made a letter out of myself here, we’re doing the next one on Wednesday, November 17th, So make note of that. That’s a day earlier than we usually do the same time, though.

56:07
We’re talking about a great new product that we’re developing that allows you to take control of your of your cloud performance and cost tool called it and visor.

56:17
So if you’ve got Azure and AWS and you are not realizing all the benefits of elasticity and cost management, take a look at that and let us, we can show you how we can help you really kind of improve performance while reducing costs. And then, on Thursday, December 2, we have a webinar on Extending Power BI Functionality, leveraging our, the open source statistical package.

56:41
And our, one of our, are, our trainers and consultants, Patrick Powers will be delivering that, so, that will be a really interesting one.

56:49
In terms of, Got just a couple of quick slides here on Centaurus.

56:53
We are, we focused solely on business intelligence with the depth of knowledge across the entire BI stack.

57:01
Our clients know us for providing clarity from chaos for complex Business requirements, disparate data sources, constantly moving targets, and regulatory environments.

57:10
And we’ve made a name for ourselves because of our strength at bridging that gap between IT and business users.

57:16
We deliver solutions that give you access to reliable analysis, ready data across the organization, sheets, quickly, and easily get answers, at the point of impact, the form of the decisions you make, and the actions you take.

57:28
Our consultants are leading experts in the field of analytics with years of pragmatic real-world. Expect expertise and experience advancing the state-of-the-art.

57:36
We’re so confident in our team, and our methodology that we back our projects with 100% money back guarantee that is unique in the industry.

57:45
It’d be remiss if I didn’t talk about our training.

57:47
We offer training across the three major platforms that I’ve alluded to there: Microsoft Power BI, IBM Cognos and Tableau featuring all the different modalities that your organization might need. Whether it’s tailored group sessions, small group mentoring, instructor led online courses, or self paced e-learning.

58:05
And we can customize those to meet the needs of your organization, particularly useful for organizations who are either migrating from one platform to another, or who are, again, maybe using multiples of those platforms as our consultants are.

58:23
And trainers are bilingual or multilingual, if you will.

58:26
And we’ve been doing this for a long time, over two decades now, working across the spectrum, from Fortune 500 as a mid market, and across many industries, and functional areas include in the office, finance, sales, and marketing, manufacturing, operations, HR, and IT.

58:41
Our organization is both large enough to meet all your business analytics needs, yet small enough to provide personalized attention.

58:48
If you’re interested in joining the Senturus as team, we are hiring.

58:51
You see the list of different positions that we have available right now.

58:56
And you can see those descriptions, again, over at our snazzy new website. Go ahead and send your resume over to jobs@Senturus.com.

59:11
And with that, I’m going to jump over to the question pane.

59:17
Let’s see. We only have a couple of minutes left here.

59:19
So what if your query subject from the Presentation Layer Model Packages, merge query from two tables, with an under lying, join, in a different namespace?

59:32
So, again, I would call that an ideal situation where the complexity of that query subject husband rationalized in the model, whether it’s you’ve used a bridge table for a multi fact query, or something like that, right, Are again, a complex scenario like that.

59:53
So that presentation layer will just show up. And we’re going to issue a query. Tableau doesn’t know. Power BI isn’t now.

1:00:02
It says, Oh, hey, I’ve got no sales and product line. And whatever complexity isn’t that model, that’s going to be handled by Cognos. Cognos is going to go, OK, yeah. I need to do this complex thing that’s in this model. And it did the tool. The visualization tool doesn’t know or carry. It just gets a result, setback.

1:00:21
That’s really kind of the beauty of it.

1:00:23
Another question, can you connect a transformer key with Power BI and Tableau?

1:00:30
So yes, we do support cubes as power cubes as data sources, so you’re not connecting directly to the MDC files. So sorry, for those of you. Don’t know.

1:00:40
The cubes. Are there, MDC files? We support packages that leverage cube. So if you’ve published a QB package, you can access it, so you need Cognos.

1:00:50
There. We haven’t.

1:00:51
There’s it’s, uh, we’d love to be able to do that, but we’ve never figured out how to actually connect to that. It’s such an old, legacy technology.

1:01:04
I’m going to jump to another question here.

1:01:05
You are assuming that all models are built off a database, do not have outer joins to be used in these other tools. one of your models are built off the transactional system. The Cognos model is outer joins.

1:01:17
So, again, in this scenario where the models are built in Cognos using outer joins, we that’s a great example of why, if somebody tries to re model that in Tableau or Power BI, you got a decent chance they get that wrong. No, you ask the average person that’s using Tableau or Power BI.

1:01:38
You ask them what a joint is and they might say, OK, it’s a link between two tails. You say, well, what’s a left outer join or a full outer join?

1:01:44
They’ll, they will probably struggle with that.

1:01:47
And you do that wrong in your model, and guess what, your data’s romp.

1:01:52
So, if you’ve done that in your model, again, you don’t have to re model that in Power BI or Tableau.

1:01:58
You just pull it in, use the dummy link and that’s all rationalized in the model are relational packages supported? They absolutely are.

1:02:07
Um, you want to use a power BI dataset as Cognos dashboard. We don’t do that.

1:02:14
Is the Analytics Connector supportive with Power BI on-prem deployment? Great question.

1:02:19
Yes, it is.

1:02:21
I’m using the SQL Server connection because it doesn’t support, custom connectors. But it has to be a Power BI dashboard so we can’t use it with SSRS reports, Senturus connected with multiple Cognos environments. You can create different.

1:02:40
You can add, you’d have to use different connector servers, but you can absolutely do that point, those two different environments.

1:02:49
So, I know I’m at the top of the hour a little bit beyond.

1:02:53
Those are some great questions and that answers all of the questions.

1:02:58
So, with that, I want to thank all of you for joining us today. If the Connector is a fit or if we can help you with any of your business, intelligence, please reach out to us.

1:03:17
Contact us at info@senturus.com or 888-601-6010.

1:03:26
So thank you very much for your time.

Read more