Workspace Roles in Fabric – Ep. 260

Microsoft Fabric workspaces have become the unit of delivery: pipelines, Lakehouses/Warehouses, notebooks, semantic models, and reports often ship together. That makes workspace roles the first governance gate your team hits—before you’ve written a single line of DAX or Spark.

In Ep. 260, Mike, Tommy, and Seth unpack what Admin, Member, Contributor, and Viewer really mean in day-to-day Fabric work. The punchline: you can’t permission your way out of unclear ownership—roles only work when they match how your team builds, reviews, and operates content.

News & Announcements

• Power BI October 2023 Feature Summary — October Desktop is out; the update includes several quality-of-life improvements that matter in real projects (including collaboration defaults and model-authoring enhancements).
• Suggest a podcast topic — Drop an idea or question for a future episode.
• PowerBI.tips Podcast — The full Explicit Measures back-catalog, with links to listen on your preferred platform.
• Power BI Theme Generator (Tips+) — Build theme JSON to standardize colors, fonts, and visual styling across reports.

Main Discussion

The episode opens with a quick pass over the Power BI October Desktop update (including discussion on composite models and collaboration workflows), then pivots to the core topic: how to assign Fabric workspace roles without creating bottlenecks—or chaos.

Fabric raises the stakes because the workspace can contain engineering and BI artifacts side-by-side. Over-permissioning doesn’t just risk a report; it can expand the blast radius across pipelines, storage, and production data assets.

Here are the key takeaways:

• Start with ownership, not buttons: decide who is accountable for the workspace (settings, lifecycle, and recovery) before you decide who can build in it.
• Admin should be a small, intentional group: treat Admin as workspace ownership and configuration—use it sparingly (often via a group), and avoid handing it out as a shortcut.
• Member vs. Contributor is the real design choice: both can build; the question is who you trust to manage broader workspace-level behaviors and content organization.
• Use Contributor as the default builder role: it supports creation and iteration while keeping the most sensitive workspace controls tighter.
• Viewer is for consumption, not development: keep ‘audience’ users in Viewer and publish only what you intend them to discover and rely on.
• Separate workspaces when audiences diverge: if engineers and report consumers have different needs (or different risk profiles), split build vs. consume to keep permissions clean.
• Document exceptions and revisit them: service principals, CI/CD, and break-glass Admin access are normal—just make them explicit and periodically reviewed.

Helpful references mentioned / related to the topic:

• https://learn.microsoft.com/en-us/fabric/get-started/roles-workspaces
• https://learn.microsoft.com/en-us/fabric/data-warehouse/workspace-roles
• https://learn.microsoft.com/en-us/fabric/data-engineering/workspace-roles-lakehouse

Looking Forward

When you debate ‘one workspace vs. many,’ start by listing the distinct teams involved and which artifacts they must be able to change—then let roles (and workspace boundaries) follow.

Episode Transcript

0:30 good morning morning everyone and welcome back to the explicit measures podcast with Tommy Seth and Mike hello everyone good morning all right we’re going to jump in today’s topic is going to be around how to best use your workspace roles all the roles that go along with a workspace today inside the context of fabric so we have pipelines we’ve got other artifacts we’ve got other data engineering things so that is going to be our main topic for today lots of I think debatable points around this one I think we’re going to have some good topic Tommy was already trying

1:01 some good topic Tommy was already trying to debate us before the the show even started so there’s going to be some good stuff there all righty let’s before we do that let’s jump into a quick a couple announcements and things that are coming across that we have found interesting across the webs I think our first major thing we need to talk about here is October desktop is here it’s out you can download it there’s all a whole bunch of new features in the blog gentlemen anything that you found that that stuck out to you what was what did you like from the the most recent blog mon

1:31 blog mon there’s a glaring one Mike that’s has to be a continued conversation on this podcast that we’ve avoided more than I would have liked and that’s the need for composite models they have they’ve continued one it’s generally available and it’s been at GA for quite a while y one of the big items in the October blog was the ability if I’m adding multiple data sets together how do I want to name the Alias or how do I want to name the other model coming in if two tables at

2:02 other model coming in if two tables at the same name like sales and sales so not only does this tell me how much of are they thinking about composite models but they are widely used and these are obviously requests features people have been saying hey we need to compensate for this is it or is this table Stakes or is this this is like basic features that they were missing like I I’m not sure like if you’re really going to get people to adopt you you I think I disagree with your comment that it’s widely being used I don’t think it’s being widely used and actually I find

2:33 being widely used and actually I find whenever I talk to people about composite models I tell them use with Extreme Caution because there’s massive text strings that need to move between models and do filtering and things so I don’t use composite models nor do I build I haven’t really encountered any clients that I use that use composite models what are you finding there Tommy I’m finding a lot with the needs for especially in self-service scenarios where you already have an Enterprise model or you already have the but I need to add an additional table or source to it I’m seeing more I’m hearing more and

3:06 it I’m seeing more I’m hearing more and more of this being a common need and that’s exactly the point of a composite model okay I’m not seeing so much the need where I have two Enterprise data sets and I need to merge them together got it but the fact that it’s GA the fact that I can choose which tables I want from each model and I don’t have to have the whole cabang I can just say these four table keang Kebab Kebab Kebab we’ll say Kebab the whole kebab can’t go wrong with that I know I don’t have to have the whole canoli I like I’m G to start using that actually I don’t

3:36 G to start using that actually I don’t need the whole canoli yeah you do need the whole canoli that’s the problem yeah but that’s the I can’t actually use that because I always do but but the fact I can choose the tables I’m like from this Enterprise model I can choose the fact table and these two Dimension tables from this other one I can use all the dimension tables and oh this is all GA from by Microsoft how is that not a need and we could say that but yeah yeah I could see it being a need I just I don’t see it as a as a strong of a feature as some of the other things that maybe were released last

4:07 released last month so Seth anything that stood out to you any any features that you have you downloaded a desktop played with it a little bit anything that you stuck out to you I think the the one at the top is interesting where the capabilities of one drive to SharePoint integration are on by default now that’s the one I was thinking more like the use the use case presented is an interesting one that I I’d love to get your feedback on because it it’s basically the the bullet points underneath are hey this is great because

4:37 underneath are hey this is great because powerbi authors can now collaborate and I I guess validate a report prior to publishing to the service do you guys do you guys find that a valid use case I don’t think so because I want to look at the model not the nice pretty layout yeah I don’t I don’t think that’s a valid use case I would want to look at the model or the report and say is this the right report should I be publishing this maybe but there’s no automatic method to get from the report

5:08 automatic method to get from the report in SharePoint to power. com in an easy way you still have to download it and publish it from desktop so what’s the point like I’m going to open it anyways so maybe it saves me like a couple maybe it saves me like a minute of like not having to download it first so I know which one I’m looking at before I go it just seems maybe it’s a ease of use UI X validation maybe that has to be a did you get the popup yet for a pro user that would make sense for a premium user that does not make sense because in

5:39 that does not make sense because in premium you get your own deployment pipeline so anything premium I don’t think this makes sense at all well not even even deployment but like the a process that you follow to push a report into a validation whether that’s its own QA environment or whatever yeah but there’s always use cases where if I’m sharing an app I already have that staging environment in the workspace right yeah exactly the workspace is the stage environment so I I don’t really get it so I still think I

6:11 I don’t really get it so I still think I still think someone’s trying to make us use case for this one to say oh this is a feature we can do this is really cool and then we’re just trying to make up a use case yeah well that was an interesting part to me but okay if you’re saving to SharePoint before you get to publish if you say I want to publish to a workspace yep you get to say hey we’re going to publish or basically save upload to SharePoint first you’re like but that’s on my computer so there’s that weird oh interesting so it has yeah it has this weird idea where like you’re not syncing

6:41 weird idea where like you’re not syncing that SharePoint Library which you already should be if you have a good workflow on your computer and it’s going to actually just upload it to it and again it’s really trying to copy or replicate the same way you would work on a PowerPoint file or a Word document if if it’s coming from someone’s shared act like a shared folder so not assuming that you already have that on your computer that’s where it makes the most sense sense but well I’m surprised Tommy you

7:11 but well I’m surprised Tommy you didn’t mention all the features that you love as all the new on object updates oh my [Laughter] J I’ve always been wanting this is the feature that i’ really been needing is I knew a I need a new onplace a placeholder for text in the editing phase so what I did notice is you can now change the total when it says total at the bottom left hand corner of the table the table now lets you change the total name and now you can change the title and there and when you hover or you click on the visual the title shows up

7:43 click on the visual the title shows up and shifts everything around in the visual so visual now gets this here’s this placeholder that appears at random times and adjust your view of the visual I’m like I don’t really no I don’t like this give the people what they want I don’t know who’s wanting this but it’s G it’s being given to you I I still I’m not a feel like it need some refinement I’m not there yet I’m not there calculation groups in the desktop now so yeah but we talk about it on

8:13 now so yeah but we talk about it on Tuesday yeah we did talk about that on Tuesday I did see a really funny meme on Twitter the other day there is a new feature that I really like is inside the deployment pipelines you can now have n number of deployment pipelines so before you had Dev test prod they’ve now just enhanced that so you can have any number of deployment pipeline pieces which I thought was awesome you can now have four Dev test QA prod whatever the heck you want to call it now and someone on Twitter said now I have

8:45 someone on Twitter said now I have enough of environments to match what I want to they had like 10 of them and they’re like the the the the environment is enough done before I had to go home for work on a weekend you to go home for work on a weekend the enough done to like it was I I know the enough done to like it was I I got to pull it up it was really funny all the different I think it was I feel like it was like a oh gosh I have to pull it up on my my Twitter feed here I thought it was absolutely hilarious and I thought it was very comical I think that’s probably the the probably the most either underrated or regardless the most valued part of the October update to me from a powerbi point of

9:16 update to me from a powerbi point of view I hear a lot of needs like why do I need three I only need two I need a Dev and a prod for smaller teams I me I think works yeah yes I I can’t under I can’t overestimate like how valuable that thing is that makes a ton of sense sense yeah oh man I can’t find the note today where did it go anyways how do you like the new home the powerbi home in desktop I’m I’m a little undecided yet I I I’m

9:48 I’m I’m a little undecided yet I I I’m going to going to reserve I think I like it it’s wearing on me a little bit but it’s definitely very different than what I’m used to so we’ll see I I I’ll I’ll I’ll hang out with it I’ll see what H what happens what are you what are you guys thinking about that you like it Tommy you like it just I think it’s fine yeah it it didn’t push it doesn’t push me one way or the other a lot yeah I’m not disgusted by it but I’m not I’m not disgusted by it I don’t

10:18 I’m not I’m not disgusted by it I don’t hate it and I don’t love it yeah it’s it’s fine it’s fine yeah I don’t yes it’s it’s okay how however double check here another feature here what I did that I like is there is now a direct button for settings which was way needed so in the bottom rightand corner buried very very far away in the corner on the right hand side of your report there’s now a little tiny gear and when you click on the gear

10:43 tiny gear and when you click on the gear all of the settings panel opens up and the amount of times I had to go in classes or teaching people all right go to the file go to the options click on options and set there was like five clicks just to get to the menu where the settings lived so now there’s a single button press to get get to all the options I like that a lot and I was probably been out for a while and but that’s to me that’s a much better feature than some of these other UI interactive things that’s going on here a significant update so now instead of seven clicks to get to report level

11:13 seven clicks to get to report level settings it’s now six yes yes anyways good stuff any other announcements or things that we should talk about before we jump in the main topic all right we’re getting the blank stairs let’s move on on to the next topic so let’s let’s talk about so main topic for today I think this is going to create a lot of conversation I think I like this where this is going to go already how do you manage the roles inside

11:44 how do you manage the roles inside your workspace now we have workspaces today there’s four roles only admin member contributor viewer and I always get confused between contributor and member I always get them flipped around member is more has more privileges than a contributor which feels weird to me I don’t know whatever for whatever reason and then you have the viewer roll on the very low end there so now that we’re adding all these extra and new artifacts The Lakehouse the pipelines these notebooks a whole bunch of other things that are coming along with this and there’s actually some

12:14 with this and there’s actually some really good documentation that Microsoft has produced to give us more of a picture around here’s what you can do with these new various artifacts and when someone has access to the workspace what can they do with that workspace all right this is going to be interesting what is your impression on this Tommy I think you’ve got maybe some different opinions than what I’ve got around how these workspaces should work and the workspace is a the workspace is a control mechanism

12:44 a the workspace is a control mechanism to figure out who has build or edit access to pipelines reports data sets and things and let me frame up something here really quick for under for for clarity when you when Microsoft recommends hey you’re going to build a data set and a report if you don’t want people touching the data set and you need that to be separate from what’s Happening inside the the workspace Microsoft’s guidance is build a pipeline or build a workspace for only reports thinly connect them or or live connect

13:15 thinly connect them or or live connect them to a data set that’s in a different workspace so the control surface area or the controls around who can edit and modify that data set those now live or reside within the workspace this is the the tricky part the permissions on the data set are independent of the workspace so I can use the data set in the workspace that I don’t have access to which I think confuses a lot of people all right introduce Fabric and

13:45 people all right introduce Fabric and add to the confusion oh I think honestly before that let’s understand roles as we know them today that’s a good idea before we get to fabricor in in a traditional powerbi workspace if you had just your data sets report the four roles and there’s always been the blurry line between admin and member the top levels admin can add other members delete the workspace yes a member can and and obviously an admin can do literally everything else in a workspace publish update delete add other admins like that

14:15 update delete add other admins like that a member can do everything but add other admins and or delete the workspace but they have the same cap capabilities as an admin except they can’t add an admin yes so really from there there not a lot of distinct differences it’s basically an admin without the deleting the workspace privileges the contributor can’t just add people but they can still update publish Etc they have all the capabilities to work in the workspace no limitations they can publish an app and then you have the viewer so I think

14:46 then you have the viewer so I think we’re already lacking a bit with roles in just your data set report environment so I want to tease on that like okay I want to go down so as you’re looking at the Microsoft documentation you have the viewer role they basically can’t do anything you can just show up and look at things right B yeah what what is what is the missing features between what I see is viewer and contributor that you’re saying need to be there like what’s what’s the missing link that you’re saying you need in that space you need someone to be able to publish to

15:16 need someone to be able to publish to the workspace but they won’t be able to share or deploy a content in my view I need someone who yeah you can publish content to the workspace but that does not mean it gets to the app it doesn’t get to the actual sharing to Consumers yet okay so that’s the difference between member and contributor a contributor cannot publish to an app but a member can so you just described a feature that already exists yeah yes okay so I’ll go back to my question again from the difference between viewer and contributor what’s what’s the Gap that’s

15:47 contributor what’s what’s the Gap that’s missing there now this is in the current world let’s thinking only about current powerbi like that’s I think there’s a smaller Gap so the reason I’m asking this question is I think there’s a very small Gap if not at all all between viewer and contributor in the context of a current powerbi workspace right where I think things get more interesting is when you start adding fabric but I’ll go back to my question again what what is the missing feature between viewer and contributor still I still don’t think I see it you obviously have something in mind

16:17 it you obviously have something in mind but if I’m gonna read your brain okay no I don’t have anything in mind I’m just I’m just trying to understand like what you think is so I don’t think there’s a gap right now yeah it’s not contributor viewer you said it’s member contributor where I think the biggest gaps are no I don’t I don’t think those are gaps I I think the Gap is for me right I think the idea is what’s confusing here is for me or at least particularly for users right the when you think about a workspace everyone assumes if I’m going to touch or read or edit it or data set

16:47 to touch or read or edit it or data set that those permissions to the data set because I’m potentially going to build reports or whatever link to analyz and Excel the data set permissions are independent of the workspace permissions so there’s like a there’s like another layer here so if I had to add another role in this list it would be data set like what can you do with the data set so the reason this is confusing is because the data set is the permissions there are independent of

17:17 permissions there are independent of that workspace so the workspace is only about the things you can see in power. com then you have this whole another layer of can I even see the data set or work with the data set independently of all that that’s what I’m getting hung up on because I feel like this is going to be now the same thing for like lake houses when we start adding other data sources of information yeah it was manageable though in in our current world of data sets reports like it was maybe a little frustrating but it was manageable like we could deal with it I don’t think it caused too many

17:47 it I don’t think it caused too many significant issues but I agree with you Mike we’re moving over to now a whole different type of environment where it’s almost SCAR or or hard to call that them just a normal workspace yeah again I just yeah so let’s keep going down this route so what any other points you want to make so let’s talk about the the existing environment right yeah when so another layer of permissions is maybe I want to give people access to the reports but they can’t edit them or I want to have a

18:18 they can’t edit them or I want to have a collection of reports and I want to give them out so that is the appside so the app side is basically it’s a duplication of the reports the data set still stays the same the app is now independently copying those reports and essentially what’s happening again in my mental model here the reports are being copied from the worksspace landed somewhere else in powerbi. com and so the as you change your report the thin reports if you change your thin reports in the workspace those can be different

18:48 in the workspace those can be different from what’s in the app they’re not a the same thing they’re literally physically two different objects that is my favorite yeah good if you go look at the usage metrics you’ll see there is the report name or the report thing and there’s another Report with the word app at the end because it’s the same report but not the same report it’s linked but it’s not it could technically be versioned differently you do remind me of my favorite thing to train yeah what was that Seth it’s a version yeah yeah yeah it’s like a version of what the and but there’s no indicator you don’t know

19:19 but there’s no indicator you don’t know in the usage metrics like you just know that it’s linked based on the the ID and you don’t know what’s published either this is my favorite thing to train like oh yeah if you publish a report and if you don’t up the app not updated correct update the data set updated everything to teach people now trippy what’s trippy is when when the app already exists and you’ve already shared it and you’re adding a new report to it it’s a weird experience to me where you have to update the app but and then add it but then you have to unhide it like the

19:49 then you have to unhide it like the action of adding it wasn’t enough to just default default show not default hide yes cuz I actually got tripped up a a couple times like this I do it yeah yeah where I’m like okay I oh I’m like chastising somebody on my team I’m like oh you just didn’t update the app updated the app no you should see it and the guy’s like I don’t see it other guys are like no no no no no go to the other tab now and unhide it unhide the thing you just added I was like okay yeah I forgot about that it’s weird well so

20:21 forgot about that it’s weird well so Mike now that we’re dealing though without from the the app point of view which again one of my favorite things to try to teach people exactly where you had the app audiences you’re like oh I didn’t hide on all my audiences because it’s hidden by default sometimes we’re dealing with fabric now and the roles if you look at the documentation I don’t know if we shared it yet but the roles in fabric still have the four main roles admin member contributor and viewer I have some major problems here I’ll just walk through what the

20:52 I’ll just walk through what the normal permissions are an admin and this is now Fabric in the world of fabric an admin can up update and delete a workspace and add or remove other people including admins and do everything else in a fabric workspace bark notebooks lake houses a member cannot add other admins and cannot delete the workspace but can do everything else in a workspace a contributor cannot share other items and cannot add other people but can do everything else in a fabric workspace so

21:22 everything else in a fabric workspace so you have three roles that can literally

21:25 you have three roles that can literally do everything else delete update publish create any type of fabric artifact that there’s no distinction between what they can create update or modify and all the artifacts engineering factory data science and then there’s viewer well so let’s let’s talk about what Fabric’s adding here right so fabric is not adding a frontend side to powerbi no fabric is adding a lot of backend pieces

21:56 fabric is adding a lot of backend pieces so do your Point Tommy you’re exactly right a contributor can edit create delete all of the artifacts that are inside your workspace so the lake houses the pipelines the data flows like but all to me those are all things that happened before the data set so if we think about the artifacts that we had in powerbi previously we had at the very beginning we had just a data set and just a report and so we were through trying to manage how do we get the data set or the data engineering portion separate from the reporting side so to

22:27 separate from the reporting side so to solve that that our initial solution was put the data sets in their own workspace then make a a different workspace where the report thinly connects to it I can control permissions down to the data set level again because that doesn’t have to be linked to the workspace itself all we’re doing now with fabric is we’re adding just a whole bunch of more a whole bunch more items that are all backend processes these are all data engineering tools so I don’t see the gap between like yeah I’d want a contributor

22:57 between like yeah I’d want a contributor to do any of that stuff if you’re in the workspace You’re Building all the things here’s where I think where I think things are going to go now previously when you building a data set the the way the reason you wanted to keep control away from the data set is because the data set is being mod modeled or created by your your basically your your data modeler right so there’s someone who’s doing the modeling of the data in that Cube anyone who can’t edit that should not be a part of that workspace so that’s why you’re moving that away what I see happening now with fabric is

23:28 I see happening now with fabric is there’s now another layer deeper than the model potentially if your data model or personal personal personality or Persona is different than your data engineer well now you need two workspaces so now you need a workspace for lake houses pipelines anything that comes out of that raw you anything that comes out of that raw Enterprise Lake type edw why do you know Enterprise Lake type edw why do you need multiple workspaces so if your data engineer is separate than the data modeler oh and you don’t don’t want the

23:58 modeler oh and you don’t don’t want the data modeler to go back and start mucking with pipelines fabric notebooks or whatever you like so if you need physical separation between those roles and you have people that only are supposed to touch the data and Engineering of the data essentially making lak houses and you have different people that are managing a data set then you need another layer of workspaces so you’d have a workspace for lakehouses you’d have a workspace for data sets and you’d have a workspace for reports because of

24:28 have a workspace for reports because of the current infrastructure because of the current roles because of the permissions the permissions don’t give you can’t say right but there’s I don’t think I don’t think Microsoft would ever do this I don’t think they’re going to add like contributor for data sets contributor for pipelines contributor for lakehouses all in the same context of the same workspace that there’s going to be too many roles and it’ll just be overwhelming it doesn’t need to be that granular I agree you don’t want if Mike’s work on a lake house he doesn’t want time to go and go what does this do click on something he wants to separate a but I disagree that the Microsoft

25:00 a but I disagree that the Microsoft wouldn’t go that far because they kind wouldn’t go that far because they already do with scorecards and with of already do with scorecards and with apps with in terms of the Gran how detailed you can get with sharing certain artifacts and that’s on scorecards I can share and or create audiences for individual scorecards and I can create a group of people that are going to see individual metrics in a scorecard in apps I have the app audiences I’m not saying per Lakehouse but on I but I am saying from a contributor level role there is the data

25:31 contributor level role there is the data engineering side people working in lakehouses people working in notebooks there’s the data science side where it’s going to be the obviously the machine learning and certain types of notebooks there’s the data Factory side data flows and pipelines why couldn’t these be separated within a workspace rather me having to create I literally wrote down as you’re talking how many workspaces do I need to create then because of the role permissions now yes that’s my and that’s my beef my my my design so this is what I’m saying when I was I me I’m

26:01 is what I’m saying when I was I me I’m talking to this a lot with people now I’m saying right your governance and your design of the system just got a whole lot harder because now you have to think about what does the data engineer and data science do what what does my data modeling team do and what does my report development team now in smaller organizations no problem you can probably get away with in small organization you could throw everything in one workspace and only publish via an app that’s probably fine because the the number of people that need segmentation

26:31 number of people that need segmentation you don’t have a clear person that’s only focusing on building reports because that person is doing the data engineering that person’s loading up the lake with data that person’s building your data model and they’re building the reports so I see a lot of the times in smaller organizations it’s totally fine to have just one workspace because when you add permissions to that contributor level it meets all of it ticks all the boxes and then the only way you share things is is via the app because then you can give report consumers or people

27:01 you can give report consumers or people who are not allowed to edit things access to the data set and or the reports as they need I feel like you guys are Su suspect of me on this one no no I’m not and I get I get like what you’re saying and why to some degree but I guess I guess the challenge question is why like for instance you have a data modeler and a data engineer and you’re saying I don’t want that person to have access to my data set

27:34 my data set why that again so why why like what what what is you’re just because something lives in the same the same workspace if if you have an individual that shouldn’t be touching the raw data set and you’re just modeling on top of it that that that’s a kind are are are you trying to control things that shouldn’t happen versus like teams having access to things that they don’t manipulate

28:04 don’t manipulate anyway it’s the argument between process and and Technology I think at this point but the reason I asked this question is there’s nothing stopping from anyone who has access to any of these workspaces to create and do whatever they would want anyway totally true yeah if ex this is where the monitoring solution comes in and makes a lot of sense now so because so company comes in and says we’re going to set up a policy that’s going to do this the policy of the

28:34 going to do this the policy of the company is I’m again looking at the world today our policy is there’s going to be a team of people doing the data modeling independently of the team that’s doing the report building right so we’re going to have two works so that’s a company policy and what you’re doing is you’re using the workspace control mechanism the technology portion to influence what’s the right permissions for those teams of people because we we know the skill set of the report Builders is similar but different than what you would have at the skill set that does data modeling

29:06 the skill set that does data modeling the thing that blows my mind and doesn’t make sense to me is is this this container of a workspace across everything without the ability to just toggle some of these services on or off for people like yeah I I want to I want a contributor that does not have lakeh house access right I want a but they can do everything else to your point right like I’m a modeler like just just get rid get rid of the hardcore engineering part because they don’t need to care about it correct make them make them a

29:37 about it correct make them make them a like make like make them in that realm and a contributor everywhere else instead instead we’re we’re having you’re forcing me to create workspaces like completely different containers where it’s not going to stop them from doing stuff anyway but I I get on an Enterprise level like you’re going to be pulling from all your data sources anyway but it’s just yeah I I I agree with M like what I don’t agree with Mike is exploding all this just because I don’t have more granular permissions

30:07 don’t have more granular permissions around the services within this ecosystem now for everybody else there’s this Fant and and this is why I challenge the question right because in many cases a workspace may contain everything and all these people are going to be doing all the activities and it’s not going to be a big deal but the larger and larger you go in organizations like to your point delation between this yes there there there are absolutely rules about who who can have what access to what systems because it matters so the fact

30:38 systems because it matters so the fact that we’re just broad stroking all of this unless there’s some way in which like one security or some other method of applying control on top of it is introduced here this is this isn’t great it’s not it’s not a good thing and I don’t like the fact that we’re we’re talking about like spooling up independent workspaces just because I don’t have the ability to say they shouldn’t be able to touch my data model but right now that that’s the only solution and excuse me if I get a little sentimental we haven’t had an IT versus

31:10 sentimental we haven’t had an IT versus business conversation in a long time and it’s really nice but but I think Mike I I’m gonna have to wholeheartly disagree with what something that you said that stands out you said oh well we have the monitoring solution to take care of that no no no monitoring is there so be clear about my statement all right monitoring solution monitoring solution monitors existing Behavior right we’re going to set policies that say no one should be building reports in the data set workspace we’re going to set policies that say we are not

31:42 policies that say we are not supposed to be connecting to data sets and such so the monitoring solution is the ability for you to look and observe are people following the process and do you need to adjust your process because of the way people are abusing or sharing things in your work like for for example everyone’s sharing stuff through my workspace and their report their reports are getting landed there and they’re being directly shared to other people and you’re getting complaints around things not being accessible that’s a

32:08 things not being accessible that’s a policy violation in my opinion you should never share anything with anyone else from my workspace everyone would I think 100% agree with that one but you can’t know that unless the moning Sol solution exists I agree so my analogy here is like you would never drive a car with your eyes closed and you would never run your powerbi environment without having a monitary solution in place it’s just not wise now you can do it but you’re going to hit something I and I so I agree with the monitor solution but in the world fabric

32:38 solution but in the world fabric especially I don’t think it’s a solution to say oh you have hundreds of workspaces it to manage now you have to make sure whether or not someone hits a policy violation good luck no no no hang on it’s not it job to do this this is the this is the point the monitoring solution or these pieces the whole reason that you have admins members and contributors is to give access or delegate access to other teams of people that are responsible so the the center

33:09 that are responsible so the the center of excellence which may or may not live in it is responsible for this their their task is to build the monitoring solution work out the policies the company believes should be right for them and the Coe should also be enforcing them so when people are violating them it doesn’t have to be an it function but then there is you can say I see bad behavior on these four workspaces and if there is like seven admins on those workspaces who’s responsible nobody is so so that’s why

33:40 responsible nobody is so so that’s why you have to very strategically use who is the admin and who is the contributor in those workspaces and I would argue there should be one admin maybe two for a backup there should be almost no members maybe one maybe two maximum but everyone should be a contributor I think you can run most of your powerbi environment on admins and contributors in all your workspaces and then when there’s a problem the center of excellence team has built reporting around what that looks like and can go

34:11 around what that looks like and can go in and specifically say here’s bad behaviors approaching from these teams these departments these workspaces Let’s either change our policy or go educate people to get them to do the behavior that we want and that’s so I’m not trying to use that as a as a hammer it’s more of like an observ observation tool about how your policies work development Hammer the development Hammer well for the in theory I agree but I think in an Enterprise solution if a team said they only had four workspaces out of policy they would celebrate they would buy a

34:42 they would celebrate they would buy a cake because there’s only four workspaces I think the amount especially introducing fabric into this to try to manage this monitor I don’t want to we could put some of this in the parking lot I just think there’s not a lot of automation around here too powerbi admin would be the one that would be control ring all this the policies would be set by a but yeah you’re you’re you would basically need a full-time admin yeah is the admin not part of the Coe do they have to be I think they do a requirement I don’t think well I think I think thether nor there but yeah

35:14 think thether nor there but yeah I think the admin’s part of enforcer he’s the enforcer develop Hammer enforcer he or she is the development Hammer I think to your analogy the solution that needs to be there if if I’m going to be driving with my eyes closed I don’t want to have to look back to see what I did I would like some stop gaps before it happens where because we don’t have that right now again a contributor can do everything and creating artifacts there’s no restrictions I I don’t I don’t agree with you that I don’t agree like I understand you went controls around forward-looking things but I I disagree

35:45 forward-looking things but I I disagree with that because there’s too much to manage there’s too much things there’s too many things that are happening and from point of view yes Microsoft has made powerbi a Comm anyone can build a report for next to nothing you can start using it for almost Pennies on the dollar like it’s just so cheap to get going because the the expense to run the and use the program is now so low that means you’re going to get a whole bunch of artifacts created and a lot of them are going to be junk so to me this is where the

36:17 be junk so to me this is where the policies or your company needs to think through a you actually need to be focusing on what is certified certified becomes much more important because there’s going to be a whole pile of self-reporting and this is what Matthew roach talks about right the Pyramid of reporting there should be a couple key strategic workspaces certified data sets and process-based things that say these are the certified things we care about and I’m becoming more and more convinced the Coe sets the policies ddge education but really manages that Center that

36:47 but really manages that Center that certified elements of context and we allow a lot of the organization to just do what it has to do at the lower levels now there are some security controls you have to put in place place like I’m not going to let you share data externally there are other things there as well but I I think you need a process if you don’t have a process you need a process to figure out how do we find the best content in our organization and then who is responsible to make it certified and maintain the Integrity of the data which is fine and I’m fine if reports are a

37:17 is fine and I’m fine if reports are a commodity what I don’t want to be a commodity or lak houses and Spark notebooks where again a contributor to admin can do the same thing do why because cost a lot of money that’s the everything that’s now running on the fabric capacity if I have a not it’s not yeah like but it’s not like they’re the ones spooling up instantly the the cost centers right and but like the the monitoring part of this becomes extremely important because to your point to your point Tommy if anybody can

37:48 point to your point Tommy if anybody can start spooling up those heavier lifting systems on the back end it is going to chew up your capacity so fast and you’re going to need to figure out and triage right who’s who’s breaking production environments right well and this now speaks to the fact that like technically I wouldn’t I would need a you have a capacity to have a capacity for alling yes and’s and now I have to have a

38:18 yes and’s and now I have to have a capacity only for my data sets and this but this makes a lot of sense to me because fabric goes down in price much lower now right so this this actually came up in my class recently talking about capacity planning so to this point when teams are abusing or abusing certain parts of fabric this is why you need to have or at least think through there there probably will be multiple fabric capacities and you charge back for that reason so hey you have a finance department that’s doing some crazy reporting and data flows and do it meets their needs great

38:49 flows and do it meets their needs great let them pay for it so in that situation your production I would argue anything that’s going out in certified gets its own own dedicated fabric it is designed for that and you let no one else touch that fabric capacity so this is this is why the monitoring is so important because now you can see who has artifacts on the different capacities and you can kick people out you can move the capacities around and when you find people abusing them now you can go buy a very cheap or lower-end fabric skew and give them that and say here you go when

39:20 give them that and say here you go when you run out of capacity you’re now that workspace is now quarantined away from other important workloads that you have so this is this is my point the governance and planning is just exponentially increased with lak houses this still okay no and I agree but it’s also what it’s speaking to a lot to me is the the role of a powerbi admin is going to be a much bigger thing than it has in the past and and I think to your point to

39:52 past and and I think to your point to Seth everything you just said Mike completely agree with but the admin role and all the other roles that still doesn’t solve anything with the current roles that you would have for fabric where again we’re just splitting up workspaces and having Monitoring Solutions where the type of role and permissions should solve a lot of this if they were set up the right way yes but there but there is like some ideas here too right so you some ideas here too right so again I’m trying to think of like

40:23 know again I’m trying to think of like so I’m doing a lot of work with companies right now thinking about the different layers of security that’s occurring inside your organization when do you have multiple capacities I think there was a large barrier to entry before fabric where you’re saying if I want an Enterprise grade premium something it was like five grand a month that’s P1 right that’s what we’re talking about so because that was so expensive it very much limited you to start breaking that capacity down into smaller and smaller increments and you were saying well I I don’t want to I don’t want to pay five grand for this department but they need something to

40:53 department but they need something to be there’s premium features they need to run so what I see now happening is you can now localize and this is I think geni MVPs have been saying this forever everyone’s been saying it’s too expensive five grand a month is not the right solution now we’re seeing we can go all the way down to pricing of 280 bucks per month for all the premium features that you need which is amazing so now you can then really dial in the right spend the right amount of usage for each department and you can go in

41:24 for each department and you can go in and not every data set should be a billion rows in size you’re not loading hundreds of thousand records in every data set you shouldn’t be this this is where the gravity of those bigger data sets should live on your main capacity it should live in your main engineering team and if your data sets start getting really large you have you have problems if that is a self-service user generated unmanaged environment and and you’re not aware of it as the central it team or Central bi team yeah Checker B asks can you have

41:54 team yeah Checker B asks can you have multiple capacities under 110 100% you have as many you need it’s a cluster right it’s it’s the it’s the engine behind the workspace right so SP for all of these things if you have a production reporting model workspace it’s you would you would Silo that and that’s the conversation here is you Silo that capacity and only have it there for that reason and all of like H how do you manage that breakout though right like allow other users like what

42:25 right like allow other users like what capacity you need you just throw a bunch of workspaces on there for data set you of workspaces on there for data set engineering tasks and then that’s know engineering tasks and then that’s what people have to use yep and then and then monitor which workspace is is hogging all the resources and then create another capacity for all these other exactly right and that like that’s what I’m saying the administrative role of this just became a much bigger conversation because there’s a lot of things that are going

42:52 there’s a lot of things that are going to be going on in this ecosystem that might your point Point monitoring policies governance like we need to understand how yes a very simplified version of what we’ve had in in the past which is the struggle struggle bu Point here right of this was powerbi like these capacities were for models and for reports and Report generation and now all of the sudden it’s like and this and this and this and this it’s like okay great but you also just enabled all these users to abuse that per like th

43:24 these users to abuse that per like th those those workhorses and there in lies my major problem here we’re talking about roles and now we’ve reversed engineer go well we need different capacities and we need all this monitoring and the trust system to make sure that people are not doing it because the member the current roles fail to provide automatic standards for what people can do Mike not one role I can’t articulate your feature though you ask for that I don’t understand your

43:54 ask for that I don’t understand your feature articulate the feature where someone can create a lake house some people can control lake houses some people can just do data science break it out into the fat into the categories Microsoft already has data factory data engineering data science activator but that’s literally why they have the button in the bottom right hand corner or bottom left hand corner anyone can do that if I’m a member or that’s why you have multiple workspaces this is turning into a Rat’s Nest of things right this is this is something they’re not willing to control

44:24 something they’re not willing to control if you want someone to own the work of The Lakehouse the lakehouses will be centralized and all the data engineering will be centralized into a workspace if the data science needs to go read that data you provide them Lakehouse level access and you give the data scientist a new workspace they’re not in the same until somebody yeah but here here’s the thing right to to Tommy’s point right this would be very simple if like all your data engineering activities are in the lake house that’s the thing in the workflow that you care about what what

44:54 workflow that you care about what what happens when that person decides to do modeling in their workspace thank you well now I don’t have a workspace just for modeling right now I’m like does the data like you could Grant permissions in multiple areas but I’m just saying like it it is it is not as clean as you’re you’re separating out just because you’re saying oh yeah these people will just do those things in this workspace like eventually policy based that’s a that’s a policy yeah so it’s I don’t see why

45:25 policy yeah so it’s I don’t see why that’s a problem I I I get it but at the same time same time like what’s the what is I see I see the difference from the the standpoint of permissions to do something Y versus where work is done there is there is a difference okay I agree with that but I I would also argue too why not let the Lakehouse developer or the data engineering team why not let them build a model in there now can anyone consume content from that model no right it may be faster for me to bring things to the lake house and use the so by default by

45:56 lake house and use the so by default by creating a lak house you will get a default data set already so as a data engineer I would like to load data in using the Lakehouse I would like to have a a report maybe there’s a QA report for me as I’m doing work in that workspace that I’m going to go look at to verify my data looks the way it should look so yeah I’m not now the conditions would be is should anyone else touch that data set should anything be published and connected to that data set that doesn’t have access my answer would be no so you could have the data engineering

46:26 could have the data engineering workspace where people are building The Lakehouse powerbi data sets and even quality check reports for themselves inside that workspace totally acceptable however you will not give other people permission to go use that data or anything else unless you have you’re not going to put their artifacts inside that workspace so we already like today I already have clients now that are doing reporting around data quality checks on real data flows so if I think about the

46:56 real data flows so if I think about the that I would want that data engineering team to do again if we’re talking about someone who needs to be responsible for when data starts failing now with data activator why not build a couple Dimensions or a couple reporting pieces that actually have the checks that I need to make sure that every time I’m reloading new data I don’t lose sales numbers in a prior month I don’t lose data there so that’s that could be an app you published that is the data Lake quality Report app and then here’s a bunch of reports that live in there but you’re not building like

47:26 in there but you’re not building like Enterprise stuffff off of it the the challenge a couple things Enterprise AR says God Bless America or God bless God bless Microsoft for building in job protection with complexity yes I absolutely agree we’re gonna have jobs for years now the the thing that I’m falling falling down on though which I I Microsoft just doesn’t do a good job anyway related to the the growup story how what what happens when I’m a small small mediums side business

47:57 a small small mediums side business and I start creating everything in one workspace yep and now we build policies what am I supposed to like why but why do they do that you’re you’re right though you’re exactly right but how like am I supposed to now like create a new workspace and move all of my stuff into that other workspace so here’s where so certain things are easier to move than others right okay hold on hold on well the data set’s easy to move a data that would be easy to move between workspaces because

48:27 easy to move between workspaces because you can just republish things like that but to your point Seth like this is why I’m trying to harp on this a lot more because your scenario assumes that they and again maybe I’m making a bad assumption here maybe you’re maybe I’m making a wrong assumption your assumption is someone just got in hands on powerbi they’re happily ploted along they made a bunch of stuff and by someone like the vast majority companies right they all threw all their stuff into one workspace because they didn’t know any better they didn’t go through any training they didn’t read the adoption road map and they didn’t

48:57 the adoption road map and they didn’t hire consultant been doing this for years no no no no okay all of those things all of those things true right because not all organizations or people that are responsible for these things are going to go hire a consultant let alone somebody who’s already on top of the fabric preview as deeply as you right so what they’re going to do is yes that like or a business person is going to do this or somebody else but if we’re talking in the realm of Enterprise how do you set things up even in this from

49:27 do you set things up even in this from that standpoint like somebody said earlier in the comments their modeling team their data engineering team like all of it is the same team they build it in the same workspace and and I’m gonna have to speak here for for Seth as well mind the C hold on hold on hold on what blows my mind is that this is a talking point for me because all of the sudden what yeah we built everything in the same workspace and now now the solution the solutions to apply a policy would be

49:59 the solutions to apply a policy would be inordinate amount of tech debt and time to break things out into a workspace what the this is my this is my point and this goes back to how do you how would you solve this with slightly more granular permissions around what people could do how easy it would be to apply a policy that says yeah we’re just going to separate this out into groups it’s all the same workspace but you guys no longer have edit or contributor access to the data engineering Lake housee we just want you in the model area so now you’re two groups did we change the workpace N all

50:29 groups did we change the workpace N all we did is put you in two different ad groups and and Mike the your approach which which it would be right is assuming that fabric is rolling out top down with this already in place with all the planning con I would argue fabric will have to run out roll down pop top down I don’t think you’re G to get a lot of bottom up from fabric because it is like you said a commodity so you’re dealing with these organizations who are like they’re not going to have this planning they’re not going to have the policies in place I hope they have the moner tring in place but they may not so you’re still dealing regardless of it’s

51:00 you’re still dealing regardless of it’s small or Enterprise you’re still now dealing with people who are creating everything in a workspace with no restrictions this is awesome your approach is great however that is not going to be the way Fabrics rolled out so how do you compensate for that okay it’s not compensation it is how people understand what power so both of you are speaking exactly to my point you guys are exactly speaking to my points because people turn on RBI and they think they know how it works they think they know how to govern it they think they know how to do all these things so

51:31 they know how to do all these things so what happens is naive businesses are showing up to powerbi and saying oh it’s just a visualization tool and they turn it on not understanding data flows data Lakes data sets reporting apps no one understands this now when I explain it to them and we say this is how things could be governed these are the control surface areas what happens here’s the understanding of the way the technology built now business make your decision and as I expose the information about what you can control and what are the

52:02 what you can control and what are the options on how to control it businesses come to their own conclusions and say oh yeah well the the lak houses we don’t want everyone to access them okay well what’s our what is the options we have within powerbi to control those now they make new educated decisions about breaking out data sets and reports the majority of customers I talk to don’t read up on this don’t do any homework don’t rearch it I would never you wouldn’t go be able to drive a car without taking your driver’s ed course it’s the same situation the other policy

52:33 it’s the same situation the other policy Point yeah it’s just naive on the business side of things fine but with fabric 2 I think you’re probably introducing not just the phys or the virtual roles in in Microsoft but you’re also according to the policy that you’re putting together the policy that you need you’re also introducing other roles at the organization right this is not just the same group don’t disagree don’t disagree you’re a punch of other roles but that but you when someone like someone in the chat today to invest in those resources someone in the chat today just said hey should I learn

53:03 today just said hey should I learn powerbi and I said yeah they said well I think I need to be a powerbi developer literally I Zen was asking about it I said yeah you should be but I said a powerbi developer is not inclusive enough of a definition of what people can do inside Fabric and powerbi the roles or the personas that live in powerbi are report Consumer Report Builder data modeler release manager which no one talks about and is incredibly important the admin and now we’ve just added two more personas data

53:33 we’ve just added two more personas data engineer and data scientist just two

53:36 engineer and data scientist just two just two there’s no there’s no other new roles there not there’s not another role other than data science and data engineer they’re adding there these are Enterprise things that is the other items and bringing those in there so yeah to to my opinion that those pieces of the puzzle are all like this we’re we have literally pulled a bunch of things that have traditionally done by it and we’re bringing them now to the business and the business doesn’t think about Deb test PR the business doesn’t think about

54:06 test PR the business doesn’t think about all these things they don’t think about Enterprise gr anything and we’re just showing up saying we know what we’re doing we’re just going to build a bunch of stuff this is why we will have jobs for a long period of time Microsoft is trying to let you deliver value quickly totally understand that but without education or or knowledge if you are an admin and running your powerbi thing you better go find someplace and I’ve got a course that does this so Melissa made a great course powerbi adoption and governance it covers everything in the adoption road map that

54:38 everything in the adoption road map that is where you should start for a minimal price if you’re a leader of parbi you should start there because there are so many things to consider and it all rolls around your people your process and your technology once you have that Foundation now we can have conversations around where things go and because this was so easy to get started a lot of organizations are going to need to backpedal and figure out what is their policy and to your point this is going to happen a lot there’s going to be lots of problems with things being published in the wrong places you’re

55:08 published in the wrong places you’re just going to have to figure it out and and I think with from the governance side of things like again if you have the resources from the onset which I completely agree with from the policy point of view from the monitoring like having that rolled out is great but I’m going back to cth point on the idea that what’s the Band-Aid solution or the temporary solution right now that we don’t have where a lot of organizations are not going to have the investment again we don’t have a lot of investment

55:38 again we don’t have a lot of investment already at organizations for the powerbi admin and from the release manager down where they’re not even willing to invest in that type of people I’m not even saying the technology just the people that’s probably another conversation we could put on the parking lot but the people involved in this do drastically change from the governance side would you agree or disagree with that I agree I agree I I think I think I I I would tend to agree with Mike in that one the

56:10 tend to agree with Mike in that one the business the business’s priorities are not Technology based fabric allows us to work very closely with them but to to the point of today’s conversation the fact that the roles are this generic is frustrating and I hope they fix it because you you the like the the solutions that we’re talking about aren’t Solutions they’re they’re they’re hacks to A system that like it’s great I don’t want to control everything for everyone but at the same time like there

56:41 everyone but at the same time like there business is not going to be interested in building techn technological solutions for data that last or can be reused or can be built on Etc that is what bi teams and it is for the fact that we’re bringing all of these groups together is is great I think because there are there are business side people that we want to rely on and include in this process the data stewards the owners of the data itself like so it’s a a mish mash of these two groups together in these workspaces and that’s what I’m

57:12 in these workspaces and that’s what I’m most excited about in fabric but to not have granular controls for these services and these things makes absolutely no sense to me you’re not turning off the capabilities for people especially as we we we we rank up in the the severity of what could happen in these places like to it to have to create all of these net new things all the time and that’s how you manage this

57:42 the time and that’s how you manage this ecosystem I think is ridiculous to your point Seth how many set tenant level settings are there right now for fabric my son can count to it two turn fabric on turn fabric off that’s it so we don’t have nearly the even the tenant level capacity to manage it again everything’s policy based or please don’t do this like at my two-year-old’s birthday party like don’t touch the cake yet that’s basically what you’re trying to say the actual

58:12 you’re trying to say the actual technological restrictions are not there yet we will get there I’m sure we’ll get there but that has to be part of it from a 10 level settings that’s an interest that’s an interesting point time that you make there so there is two settings there’s one turn fabric on turn fabric off the second setting is can you make fabric artifacts or not and the the latter or Security Group it’s globally but it’s a security group that you can enable to do that but that’s yeah but here’s my argument though the education of the members of people in your

58:42 of the members of people in your organization that you’re going to light up and turn on that again this is all about policy at this point there is a flowchart that says someone requests a fabric workspace or the ability to edit fabric I think you should turn off off all the ability you turn on fabric but you leave everyone out of the group of people who create fabric artifacts only after a organization has thought about their policy gone through what they what it means to them who can do it who’s going to be responsible for it once you have that Foundation figured out

59:13 have that Foundation figured out you run those people through your training figure out who those people are once they’ve gone through now you’re able to control it with policy so now you turn on the fabric piece and this is an epiphany that you said Tommy I’m just literally coming to this realization right now that’s a great way to control it yeah because we’re even we’re even saying again with our skills Matrix around what skills do you need as a powerbi developer I don’t want to give people build access or making powerbi stuff you shouldn’t have desktop unless you’ve gone through a minimum of here’s how to build build a report here’s how

59:44 how to build build a report here’s how build a model here’s how to publish things so my my point here is there are so many things that are now appearing the leadership needs to really understand there’s a lot of stuff coming and you need to plan for it and now more than ever it’s going to be important to have your Coe it’s going to be more important than ever to talk about what you think this needs to be happening with your data because when this stuff lights up organizations who don’t plan for this are going to get a lot of rats nests all over the organization

60:15 of rats nests all over the organization of really bundled up weird data that it’s going to be you can public I can literally create a pipeline that takes data from one workspace and lands it in a different like house in a different workspace I can move data between workspaces which is one amazing but two super scary because what happens when someone says Ah we’re not using this work case anymore and they delete it and all of a sudden someone else who’s relying on data in a different workspace somewhere else totally differently is now the policy handle that well it is a policy like that’s know I’m saying it’s

60:45 policy like that’s know I’m saying it’s a after the fact it’s after the fact are you going to fire that person the lake create create as many workspaces as you want for all all the things so that the only way you can understand what’s going on is by using our new tool called purview yeah right right I’m not not wrong but but that’s that’s Microsoft on the on a base on a base level I do like the point that you called out which is yeah like turn the

61:15 called out which is yeah like turn the business user off from fabric yes but it’s still doesn’t address the the roles within those ecosystems if somebody train agree with you and I think it’s just I’m going to go back to it’s about education I think you can get people educated I think you get you need to get people on the same path and what has traditionally up until this point the business is running many different directions all I’m saying is we all got to start thinking about how do we run in the same direction at least on some things yeah my last Point too is I we want people to again we do want this commodity based we do want people to try

61:46 commodity based we do want people to try things out but there are certain workflows in this same environment that we don’t want oopsies we don’t want mistakes just like accidentally deleting a table regardless of his policy base they’re but how do you recover though it’s going to happen you just got to it’s G to happen but we’re we would like to prevent it in some pipelines in some workflows more than others we just need to go buy buy powerbi Sentinel and back everything up and then you’re good perfect perfect you you would do and you wouldn’t do it in the first place if you just purchased

62:17 in the first place if you just purchased Mike’s training in on all this and you’d have policies in place I’m telling you it’s it’s a good place to start I you it’s it’s a good place to start for it’s the the minimum dollar mean for it’s the the minimum dollar investment to get going I I didn’t I didn’t mean like let’s keep talking about it no I’m just saying it’s it’s a good place to start anyways with that I will say there’s a lot of good things coming I think this is a really great conversation and this is just a start of many more conversations that are going to go deeper around how do we build that relationship between the business and

62:47 relationship between the business and what is governing because the tool is doing things that we may not be approving so we got to figure out how that all works inside our organization with that thank you very much for listening to podcast this was a heated discussion one of our more heated ones recently which I think is pretty good so hopefully you found some interest from it hopefully it definitely turned on some knobs or or some lit up some ideas in your head around what’s going on there and then we would really recommend if you like this please share it with somebody else that’s very helpful for us and to continue to grow the podcast Tommy where else can you

63:17 the podcast Tommy where else can you find the podcast you can find us on Apple Spotify or wherever get your podcast make sure to subscribe and leave a rating it helps us out a do you have a question an idea or a topic that you want us to talk about in a future episode head over to P powerbi. com we thank you all very much we appreciate your time today we hope you have a great

63:47 your time today we hope you have a great week and we’ll see you next [Music] you

Thank You

Thanks for listening to the Explicit Measures Podcast. If you enjoyed this episode, share it with a teammate and subscribe so you don’t miss the next one.