Stop Using Bookmarks - Ep.526 - Power BI tips

In Episode 526 of Explicit Measures, Mike Carlo and Tommy Puglia unpack the latest Power BI and Microsoft Fabric topics from the show. You’ll get a quick read on the episode’s biggest ideas, why they matter, and where to dig deeper in the full conversation.

News & Announcements

• No linked announcements were available in the episode description for this post.

Main Discussion

This episode covers the major themes, opinions, and practical lessons Mike and Tommy surfaced during the conversation. The transcript below captures the full verbatim discussion if you want the exact phrasing and context.

• Mike and Tommy react to the episode’s biggest Power BI and Fabric developments and explain what stood out to them.
• They connect product announcements to day-to-day practitioner decisions instead of treating the news as abstract roadmap chatter.
• The conversation highlights where teams can move quickly, where they should slow down, and what tradeoffs deserve attention.
• They share candid perspective from real project work, which gives the discussion more practical value than a headline recap alone.
• The episode mixes tactical advice, opinionated takes, and a few forward-looking predictions about what listeners should watch next.

Looking Forward

If this episode’s topics affect your current Power BI or Fabric plans, use the transcript and linked resources to identify one concrete change you can test with your team this week.

Episode Transcript

0:02 up the sky. Dance to the day to laugh in the mix. Fabric and A. I get your fix. Explicit measures. Drop the beat now. H feel the crowd. Explicit measures. Hello everyone and welcome back to the explicit measures podcast with Tommy and Mike. Mike. Good morning Tommy, dude. How you doing? Well, I’m here. I’m awake.

0:32 Well, I’m here. I’m awake. You’re awake? It’s been a feverish pace running into summer. I guess it’s just been crazy. Lots of things going on. Things are just wild right now. I don’t know if they’re busy for you, but they’re definitely feeling busy for me. me. Yeah. No,, it’s that we got the kids almost done with school. Projects are crazy. People are asking for fabric. It’s great, man. So,, really, dude, I’m really seeing the adoption now of finally going, “Guys, like literally there are some PowerBI

1:02 like literally there are some PowerBI projects I was working on. They’re like, “Hey, when are we going to do fabric?” I’m like, “I don’t know. I’ve been asking.” And it it’s really one of those where it’s just now the just native experience. And it’s funny because people like, “Well, it’s still PowerBI, right?” You’re like, “Yeah, “Yeah, it’s just now part of it thing.” So there’s still there is still confusion around it around fabric you’re thinking not not obviously for us but is it powerbi is it I’m still getting those questions or fielding questions from people

1:32 from people yeah so again technology goes faster than people are yeah agree and I’m going to ask you a question about that before we got before I ask you the question though Tommy I do want to go jump back over to our main topic today so our main topic today is called stop using bookmarks. That’s our main topic for today. So, we’ll come back to bookmarks and talk about them inside reports. reports. Okay, Tommy, you you’re talking about fabric. my question to you is when

2:03 fabric. my question to you is when you I have I have this question fairly often as well and I have an explanation that I give like an elevator pitch around what is it based on someone and for context here users are typically coming to me in from the PowerBI space. Yeah. Yeah. Asking hey I think we should be using fabric. What should we do to get into fabric? So curious, Tommy, when someone comes to you from PowerBI and they’re like, “Hey, we’ve been hearing about fabric.” What is it? What’s the elevator

2:33 fabric.” What is it? What’s the elevator pitch that you give that is that is that that defines fabric for you? Well, typically I always talk about their bottlenecks and they always have bottlenecks and I say, how you’re always trying to we’re always trying to get your data. You’re working with it. We don’t,, we’re pulling all that data in PowerBI and it’s great and it works. However, fabric is simply this layer for us to finally actually be able to store and hold your data. And it’s and I always joke around like I would not say this two years a year a year ago, but in terms of the

3:05 year a year ago, but in terms of the dead easy solution just to get started to be able to store your data., there’s no better way right now. So, that’s what I explain it because yeah, yeah, that’s a pretty good pitch. That’s all right. So, I’m yours. Let me let me let me sling you my pitch and let’s see what you think about the pitch. is pitch. is you maybe you can take mine. my pitch is fabric is all of the data platform items previously developed for many many years now moving into the same building

3:37 years now moving into the same building space as PowerBI. That’s that’s how I like and then I can if they ask questions from there I usually explain well so yeah so yeah pipelines came from found from Azure data f factories we have lakeouses which are blob storage right it’s just a more consistent package instead of having to do a bunch of Azure deployments I just go to fabric click the item I want and it’s there that’s how that’s my mental model what do you think I actually I’ve done that bit but to me that at least with the clients I talk to

4:09 that at least with the clients I talk to there’s almost a bit of overwhelming like well why do we need that? Why do we need all the things in Azure? It’s like well it’s not that. So that becomes an explanation once we dive in more. Okay. Okay. But again you an elevator pitch is I don’t need all the things in Azure. It’s like yeah you don’t need it but what you do need. So I guess the point I’m trying it’s hard. Yeah. The point I’m trying to convey is fabric isn’t something new. Mhm. It’s a repackaging of something old that is now just easier to use.

4:40 that is now just easier to use. So that’s my main story there is like Microsoft’s already good at data platform things. They’ve got a whole ecosystem. You just didn’t know how to use it. use it. Right Right now now that’s what I need for my clients. Yeah. Now it’s easier for you to like turn things on and and move things around and create what you want to create and you’re not having to provision a bunch of things. It’s just handled for you. So, I think that’s that’s trying to be the pitch, right? So, I want companies to feel like,, it’s not like, hey, I’m going to show up here and get like this half-hearted product anymore, right? I’m

5:10 half-hearted product anymore, right? I’m I’m stepping into a product that’s been maturing for 10 plus years across other stacks in other areas. We’re just bringing it all to the same place, right? So, that’s really the the message I’m trying to convey with my elevator pitch. pitch. what that almost sounds like? What’s that? That almost sounds like you ever heard of data for dummies? That’s you. And that’s what we made this idiot proof for you. you. Yeah, maybe. I don’t I don’t No, I don’t see that way. But But I don’t see the pitches being that way.

5:40 I don’t see the pitches being that way. But But no, it’s it’s not. But to be honest, fabric is very idiot proof coming from Azure., if Mike, if you want to create a database in the next two minutes, I’m not I’m not going to say idiot proof. I’m going to just say it the integration is much tighter right the right the the things that are tricky or I have to go through all this the screen to pick all the different regions and

6:10 to pick all the different regions and pick the different settings on the server and there’s there’s a lot of like your managed keys your managed identities there’s a there’s a lot of upfront like configuring to get a thing out the door and then you’re like well is the lemon worth that squeeze right is it going to be that good you, starting up the SQL server or the server and and then when you create it, there’s like three things created and I’m not really sure what’s going on there. So, where’s the security on each level and the networking? There’s a lot of things and and just having something that just owns

6:41 having something that just owns that that part of it. It’s really the right solution. Yeah, it’s interesting, Tommy. Yeah, I like it. So, yeah, it’s fun. But, one other quick note here,, our conversation with Alex, his CLI tool for the fabric development. Yes. Yes. Well, Mike, I went less and more building. Yes. I went crazy with it because I was looking in the repo and there’s a bunch of skills, ton of skills. Obviously, there’s a lot of Python scripts, but I was intrigued, Mike. I’m like, well,

7:13 was intrigued, Mike. I’m like, well, you’re going to like this because one of my notion things, I was like, well, notion has skills. You can actually use skills in notion. Just create a page. So I’ve basically created a page with all the skills and I said hey just convert this because we’re not going to be building scripts I don’t want you to deploy anything but I want you to all the framework and the architecture use these skills create a fabric advisory agent basically and it it’s like so just now I have this agent where I’m like hey let’s take a

7:44 agent where I’m like hey let’s take a look at this meeting based fabric advisory agent I have the same business question the same way that Alex framed it. So, dude, Alex did a great job with that tool, man. He really did. Well, and I think I think the apply So, one of the things that makes this tool, I think, very useful or very interesting is the reasoning of the agent is applied in the appropriate place to build the system. Right? the agent is applied in a way that is

8:17 applied in a way that is where like helping me decide which parts of the architecture to stitch together agent related right going through the documents and understanding a lakehouse can be a landing zone for these other data sources right these are things that again Alex is like this is just in the documentation it’s just there it just wasn’t all in a way that agent could read you could use it, right? So the repo is mainly around like the meta data or the

8:47 mainly around like the meta data or the core parts of all the components that are fabric. So So the the the part that I think I really resonate here with is once the agent makes the decision or you beat up the idea with the agent a couple times, everything else is just Python. It’s the deployment scripts and putting things together and like how to build things. Like all of that’s just code. And this is I I guess my point, Tommy, right? I’ve been on this kick here for the last week or so around

9:17 kick here for the last week or so around creator agents., we’re talking about agents. Yeah. Yeah. Yeah. Creator agents is really what I think we want. We want agents to create things that we can use. Alex had done the creator agents part of this. Yeah. He’s applying an agent still, but the agents apply in the reasoning portion of the process. process. Mhm. the creator agent made all the Python scripts, made all the things that would be deployable. So once you had the structure laid out, everything from there on out is known technology,

9:48 there on out is known technology, PowerShell and Python, like easy like we already know what that looks like. So this is where I think the real power of agents rely is it’s not necessarily like ask it a question and have it do every single thing only in the agent, right? It’s a combination of look at your process. Where do you need to think about different scenarios or patterns or whatever, what information can you provide to the agent to do the thinking part, which is what Alex provided in the repo. repo. But then after that’s done, make the

10:18 But then after that’s done, make the agent write the code. Yeah. Yeah. And then everything else is just automatic. It’s just build it. So now you could have the agent build three or four different patterns. You can redeploy them as much as you want. Yeah. And Yeah. And and that’s not using any agent. Yeah. And it’s it’s crazy. So, there’s a guy online, his name’s Simon Wilson, and he’s great around the AI space, and there’s this tool he has, which will

10:39 there’s this tool he has, which will actually export your cloud code transcripts into something like that’s really readable. And if you actually dive in, if you ever whenever you use those agents, you don’t realize anytime you ask it to like make it do work, it’s actually using about 10 agents per project. So, and actually will deploy things or it’ll create its own agent based on your request or use the agents that it has and you can actually see in this really nice readable way the agents talking to each other before it gets back to you. The other might do a research one it may

11:09 The other might do a research one it may do a debugging that one say hey you need to know about this and then you’re the one you’re talking to comes back with all the output. So understanding that workflow is so like if you really want to start scaling this out is so important. Yeah, I agree with that. I haven’t I haven’t I’ve done a little bit of exploration around this area here around multiple custom agents. Mh. Mh. but I feel like this is another level up once you understand how to use agents. Once you understand how to use the command line interface with an agent, agent, the next level up is okay, now you’re

11:40 the next level up is okay, now you’re starting to build custom agents that do a bunch of things, very specific things. Awesome co-pilot is a website. I think it’s awesome. Is a great website to go pick up a lot of these custom agents. Hey, you’re the reviewer. You’re the CI/CD expert. You’re the PowerBI modeler. You’re the DAX modeler. Right. those are things that are very useful in those regards. Very, very cool. Awesome. I really like this. Okay. let’s get over to some quickly news items here. Tommy, which what news are you going to start with here? here? Oh man. Well, I’m going to start with

12:11 Oh man. Well, I’m going to start with one that you thought was dead, but it’s not. It’s going to be Tableau Editor 3 and a great article here around AI readiness and the best practice for your semantic model. How do you basically do the AI prep, right? Like Microsoft has that part of the tool, but it’s fine. So really when it comes to we’ve talked about on our podcast and other people have as well that the best framework for data around an a something agent is a semantic model ideally and it talks about really the

12:42 ideally and it talks about really the article is great because it talks about obviously the output’s so much more important than just a prompt. it’s the structure, the documentation, the anchors, and the examples., or you’re going to be very likely to get inconsistent answers. So, it talks about where you can actually understand your data before actually designing. So, how do you set your expectations? How do you gauge this skills that are going to be needed? And then how do you actually build that structure, the data quality, the functional model design?, and going through how do you actually

13:12 going through how do you actually optimize the model the best way. A lot of this is I think universal best practices around semantic modeling in general. Honestly, a lot of this is even from Microsoft’s old AI prep guide when they had to ask your questions on a dashboard. So like for example, hiding the columns, reduce redundancy in column names. And then the only part that’s really unique for the age we’re in now is providing that context. You like how do you actually give it the instructions? How important those instructions are? or what can you give

13:43 instructions are? or what can you give it external documentation? So, it’s a great article and tablet editor actually just came out with their own AI tool and the product too. So, So, yeah, it was about time I was getting really worried that they fall behind here. still don’t think I’m going to write C# on things that’s not but you write C# on things that’s not but having an agent to write it for me know having an agent to write it for me like that’s a much better experience. Again, back to my creator agent experience, right? I don’t want to do all the code writing anymore. That’s not a good value ad for my time to sit down and reason through code. code. The agents are better at that now.

14:14 The agents are better at that now. Period. Full stop. Like don’t you can’t argue with me anymore about this one. Like this is something I never argued with you. I’m just saying. I’m just saying people in general are like no but I want to touch the code. I’m like but people but pe like no wrong. Like I can outcode you in circles. Like there’s there’s no reason why you need to be writing code. If I’m looking at Fortune 100 companies fully ripping code with their agents and we’re saying I’m hearing numbers of I

14:44 we’re saying I’m hearing numbers of I don’t know why I was reading this one but I was reading numbers of like maybe from Uber is an article I was reading up to 75% of their code now is all written and managed and owned by agents like if they’re finding like success with that there’s no reason on earth we need to have this conversation. My job has substantially changed from writing the code to now to your point time earlier building custom agents figuring out how the agents can work together. How do I get the agent to do more self,, discovery of the things that

15:14 , discovery of the things that it built? Like that’s where it’s at right now. Yeah. Yeah. And that’s the same thing with time editor, right? I don’t want to write stuff. I want to have a nice graphical like that’s the whole reason why we like Power Query. The whole thing was so graphical. I’m not writing M code. No one wants to write M code. The reason we like Power Query is because there has a bunch of buttons you can click and I just did things. Well, now I don’t even want to click the buttons. I just want to tell the agent what I wanted to do. Hey, take this table, delete these columns, and and, make this unique, whatever. Like, I just wanted I

15:45 unique, whatever. Like, I just wanted I don’t even want to like click buttons now. Like, I’m becoming such an agent prick at this point because I just don’t want I just don’t I don’t want want The hardest word has ever been used on the podcast. Oh, man. I I just I don’t want to even click buttons now. Like it’s it so but this is where this is why I think this is so revolutionary I believe Tommy because this is we’re we’re physically moving into a new space like we’re we’re actively moving from the where we were

16:15 actively moving from the where we were sit in front of a keyboard and type hands on keyboards to it’s now video it’s now talking. It’s now text to speech speech to text. That’s that’s the integration mode. Like that’s going to shift how this all works. First, I have one problem with what you said. I actually do like writing Power Query or M language. The times that I have done that. Hold on. Hold on. I know. But the times that I’ve done that and it worked when I needed like some looping or some ridiculous thing back in the day. I felt like Tom Hanks and

16:46 the day. I felt like Tom Hanks and Castaway like ALL RIGHT I MADE FIRE KIND Castaway like ALL RIGHT I MADE FIRE LIKE just but no but I think it’s an OF LIKE just but no but I think it’s an important distinction because I think people listening oh because you’re saying code and they’re like well that’s not powerbi oh yes this is when we’re talking about this like you got to mention here we’re also talking about DAX and we’re also talking about VM language here three years ago the only way you could have anything write DAX was chat GPT but it was maybe right right because it didn’t understand the context text the fact now that you can

17:16 context text the fact now that you can have things pour over the semantic model so it understands the context with the MCP server skills like there’s a whole bunch of extra knowledge we can now give it to help it understand things the first time I used the MCP over a model I was weary I was very weary because I wanted to go from soup to nuts it was raw data y and I said but I had a pretty intensive instructions for that particular project it wrote all the power query created all the relationships in a great Hey, when I said, “Hey, there’s actually two tables.

17:46 said, “Hey, there’s actually two tables. I need them to actually talk to each other. Is there a way?” It’s like, “Oh, yeah. I think we can make that work.” It like,, all my DAX measures that were and I’m like, they work incred like so it’s this is not when we say code here. I just want to make sure that we’re emphasizing that we are very much talking about PowerBI and the DAX and the M world as well. well. Sure. Yep. Exactly. Well, and the only code that we really touch inside typical

18:16 code that we really touch inside typical editor is yet another language,, C# script that are running like,, automations that they built,, tab editor built a a,, a language that they can use to manipulate the structure of the tables and measures and things. This was required at a time where we didn’t have things. I’d even maybe argue Tommy, do we even need the C language in tablet edit anymore because we have timal? Like timal’s the language which we would program a model with anymore. So like

18:46 program a model with anymore. So like it’s just now automation on top of Timle. So while I understand why they have it and that’s like a legacy part of their tool, I’d rather just talk to the agent and have the agent just figure it out. I know it is frustrating that I’m throwing that skill away because I spent many hours trying to understand the Tom language back in the day because I was so so intent. I’m like I want to write C I want to write macros and yeah so but hey anymore you got to move on. also in this article just really briefly here Tommy it talks a lot about

19:16 briefly here Tommy it talks a lot about like structure and content and like you like structure and content and like searchable names. I’ve gone through know searchable names. I’ve gone through a couple recent projects around this one. This is like, okay, let me this may this may be a bit opinionated an opinionated option here. Preparing your model for preparing your model for AI isn’t what we’re doing here. Yes, it’s part of it, but we’re actually just making really good models better. Right. So you you

19:47 good models better. Right. So you you give a user,, 15 dimensions and six fact tables and you put all the measures in a measure table and say, “All right, user, go build it. Go build your reports.” They don’t know what to do. They don’t know what they’re doing. They don’t know where the data came from. They can’t trust it. Right? This is the same approach with the agents. So like I understand the point here, but there are so many bad models out there that don’t have any descriptions, don’t have any measures,

20:18 descriptions, don’t have any measures, don’t have any relationship. So users are trying to go out and build things and and it’s easy when you’re the one who made the model to use your own model. Of course, perfect. The whole game gets harder when you’re trying to make the model for another team. Mhm. Mhm. You’re going to build reports with it. I’m going to make the model. I’m going to hand it to somebody else. So now we need to have communication between me the model creator and builder maintainer and you the report user or report builder. This is where the exchange of

20:50 builder. This is where the exchange of information from me to you has to happen. Either you’re building separate documentation about it or you’re putting as much as you can into the model. And I think what we’re seeing here is people have traditionally just made separate documentation. Now it’s let’s make all the documentation in the model where it should belong anyways. And that one makes a great benefit for AI, but two, that’s what your end users want anyways., can you can you trace

21:17 anyways., can you can you trace the lineage of this column in the semantic model back to the source data table that came out of SAP? So that way, if I’m running a query in SAP, I know that column should match this data in my model. model. That’s important., nothing here’s new to your point. point. Nothing here’s new. This is not this is not new information. And this is just your model probably isn’t ready for regular users either, but they can learn other ways and have other supplemental documentation. This is just taking all that and putting it in the model. Anyways, really like the article. I

21:47 Anyways, really like the article. I thought it was good., definitely agree with the a lot of the sentiment here. There’s a lot of g great tips or tricks in here.,, thin things out, remove a lot of stuff., add annotations, add your descriptions, make sure you have synonyms. if you have DAX measures or things that you’re you’re generating put comments in code around DAX measures it can read that stuff as well. So really good good article here it had a lot of great key examples and and very thorough here.

22:19 key examples and and very thorough here. So anyways great great article from Taber definitely worth a read. I’ll make sure I put this in the chat. Tommy we have a little bit more time maybe quickly. quickly. Do you want to do that or do you want to just go to the topic? You can go to the topic that’s fine if you want to go there. All dude, I’ll let you drive. You drive. So today’s main topic is probably going to be a bit controversial. Are PowerBI bookmarks even worth it in desktop anymore? And what? And and maybe it’s maybe the answer is no. Maybe the answer is yes. But are bookmarks even

22:49 answer is yes. But are bookmarks even worth your time to spend on bookmarks? Do we use them? Where do they find value? And maybe there’s a there’s a use case or a special scenario where you should be using bookmarks and you should maybe pull back on bookmarks in other areas. So, let’s go there. What do you think, Tommy? What’s your initial reaction to this this topic? Do you like bookmarks or do you want to get rid of them? what? It’s a this is a lovehate relationship. And I think I’m not the only one who feels that way around bookmarks because again just to

23:21 around bookmarks because again just to explain bookmarks have been around in PowerBI nearly since the beginning. if the first time I think was actually at their first conference the data insight summit was when maybe then they announced it so 2017. and it’s an integral part. takes the snapshot of a page, whether it’s the data, the visuals, the sorting order, what’s actually being shown, what’s being filtered, and you can create as many bookmarks as you want. There’s a lot of great features around grouping things. They have I finally added things

23:51 things. They have I finally added things for buttons., it’s been around for a while, and it’s a great part to make it an app-like experience in PowerBI., however, for those who have intensively looked into bookmarks and built them, they can be incredibly hard to manage, especially when you have a lot of bookmark what does what, organizing them because you see all, no matter if you’re creating bookmarks for a single page and you have a lot of bookmarks on a page, anytime you go to the bookmarks view, you see all the bookmarks everywhere for

24:22 you see all the bookmarks everywhere for everything., so it becomes hard to manage what they each do and when to how to actually update them. So I think the question really becomes really Mike, we’ve been so focused on our data and just the last year and a half or so around providing the great data and we’ve talked about the report design just comes with that. I guess the question for you is are bookmarks still one of the first tools that people use when they need a navigation when they need to help update

24:54 navigation when they need to help update a report or a model it? Let me give some additional mental context that I’m relaying this question through. through. The first part of this is the bookmarks were made in a time where a lot of other navigation elements did not exist. There’s actually a button on the a button you can now use. It’s called the bookmark button navigation or bookmark the navigation button I guess is how you would say. I don’t remember the name of it exactly. But what you do now is on that button, every

25:25 what you do now is on that button, every page of the report is just automatically added as an item on that button. And if a page is hidden, the button can be shown or not shown on that one single object on the page that now exists as a separate button. When I was building bookmarks originally to navigate different parts of a report page, page, that was one of the reasons why I made bookmarks because I needed the ability to navigate between pages. Oh, yeah. Oh, yeah. Using the report, right? So, that now

25:57 Using the report, right? So, that now exists that never existed before., I’m trying to think about other options. Buttons weren’t weren’t a thing at all at the time. There was maybe a couple buttons, but the action was there was only like an action on the button. there there’s there’s a lot of other elements on the the the report page. We didn’t have a filter pane. I remember before filter panes there was none. So we wanted to not remove page space for a bunch

26:28 remove page space for a bunch of slicers that felt like I only use them once. I set them and then I remove them. Right? So, how can I have an experience where I show something on the page for a period of time while I’m interacting with it and then when I’m done with that interaction actually allow that object to be removed? You could have built a custom bookmark, a custom visual, all these other things, but it was just too difficult to do those things. So, what we did instead is we cobbled together bookmarks and all these visual elements, showing, hiding, all these things. At the end of the day,

27:00 things. At the end of the day, it did a lot of things, but it was also really confusing to ever edit and maintain them. So, yeah, yeah, that is I’m going to say if if bookmarks had an Achilles heel, it’s the I don’t know what this bookmark does after you make it. Unless you have a really good name on it, it doesn’t have a description. You can’t see what items it’s affecting. Yeah. Yeah. You can’t see like how it’s attached to that. Like, you can only click on it and then see the page. You don’t know what actually happened. You have to really dive in each time. Yes. Now, with the adventation of the

27:31 Yes. Now, with the adventation of the new report format, PBIR, you can actually make a bookmark and go see what it does. You can actually look at the information in it. Like that’s also helpful, too. So, let’s let’s step back now, right? When you look at a report and all the objects now, the bookmarks existed in October 2020 2017. I was Okay. All right. You were very close, Tommy. That’s I just Googled this one. So, it’s that’s when they came out. that was in a very raw state of PowerBI desktop desktop and right around that time is when I

28:03 and right around that time is when I decided to start building a lot of scrims and backgrounds and layouts and templates and like I’m I can build full experiences on the reports and so for me bookmarks was an enabler to build a lot of data app like experiences inside PowerBI with data visuals I loved it right later on you saw that A couple years down the road, it was very difficult to maintain, very hard to get that to work. And the people that built the bookmarks were pretty pretty much the ones that had to maintain them

28:34 much the ones that had to maintain them verbatim. Now, Tommy, I almost never use bookmarks on anything. Ah, Ah, so it is while it was a good feature for the time, I think there’s been enough other supplemental elements that are now supporting the bookmarks. So, yeah. So, like what Yeah, I’m intrigued to hear what you think is in a sense replacement. It’s just these other elements. It’s other visual elements, right? The filter pane is now something you can turn on or turn off. I like using that more than I like having a bookmark to hide and show filters, right? bookmark navigation. If I want

29:05 right? bookmark navigation. If I want page navigation, I want to hide the tabs at the bottom of the page. I now don’t need that. Right? Also, page navigation is no longer at the bottom of every page. You can actually move page navigation to the lefth hand side. So, if you’re doing PowerBI embedding, you can actually move the page navigation to the left side of the report page. also feels a lot like the the the apps that you build, right? I build an app. you get navigation there. So if I have navigation in an app where the actual pages are listed on the lefth hand side, I don’t actually need the navigation at the bottom of the page. I don’t actually need buttons on the page.

29:35 don’t actually need buttons on the page. I can allow other PowerBI core features to just carry that information. So those are the things I’m replacing it with and that’s why I don’t use them nearly as much as I used to. Let me pause there. What do you think? And I’m I’m unfortunately I’m with you here because here you’re meant we’re talking about the complications here. Unfortunately, why is that? Well, because I like bookmarks and I think that’s why like I it’s not that I am it’s unfortunate that I agree with you. It’s unfortunate about what we’re agreeing on. So

30:05 agreeing on. So ah I see. Okay. Yeah. And it doesn’t it’s more like I don’t want to agree with you. It’s unfortunate that I’m actually agreeing with you on I’m always actively trying to see how I can go with the other side. So,, but no, I think that, but here’s the thing, and I I want to at least make this note. Is it really complicated for the developer? Yes. But when you really do set up bookmarks and using buttons, the consumer doesn’t know any of those complications and it makes it dead easy for them, especially with buttons. So, there is that. Yeah. So, it’s not so

30:37 So, there is that. Yeah. So, it’s not so much that it’s complicated also for users. it’s actually in a sense a better you wouldn’t make the argument bookmarks enhance your experience it enhance the app experience when you’re in the reports agreed agreed but again the but again I think to your point the value ad here right and unless where I stand is unless they’re simplified like the value ad isn’t is to your point the lem is not worth the squeeze if I need some simple bookmarks

31:08 squeeze if I need some simple bookmarks maybe show and hide some information or show the show the data off of this time frame over this time frame. These quick things that they want to see snapshots of that’s where bookmarks can be pretty handy and they’re again don’t really lead to that complications. Yes. But I think to your point when you start building rather than building bookmarks around the report when you’re building reports around bookmarks so to speak where your reports depended on it

31:38 speak where your reports depended on it that’s where again it’s you’re taking a lot of time for something I don’t think is providing that much value to users. There is a place for it absolutely and I still tend to use them. But let me pause there. Do you agree with at least a simplified version of bookmarks or

31:56 simplified version of bookmarks or I I want to that’s where I was going to go next Tommy which was you going to go next Tommy which was where is a good use case for know where is a good use case for bookmarks where do we see this being applied correctly I guess I would maybe say and I think I would argue Tommy the the value of bookmarks come a lot more from personalized bookmarks right now so when you have a report and it’s published and some user is clicking around the report and like I want to always see this year’s data 2026 I want to make a bookmark for myself taking snapshots of the data,

32:27 taking snapshots of the data, highlighting some filters, snapping that filter context, and being able to come back and set that as my default filter selection value when I show up to the report is extremely useful. That’s very helpful. So, when I look at that part, I think to myself, oh, personalized bookmarks makes a lot of sense. I really like that one. And I think also if you are starting to embed if you if you go use PowerBI in teams I think one of the mechanics that Microsoft uses

32:58 one of the mechanics that Microsoft uses when you select something on the page and then chat to somebody about it right in the in the chat experience you actually communicate and talk about the data or highlight things right that’s also I believe using bookmarks so it behind the scenes there’s a mechanics there so hey I want to highlight this visual this table this bar and this chart start and say, “H, this doesn’t look right. Can you go look at this information?” Right? If you want to communicate about the data you’re seeing in the visual, that’s where bookmarks are applied. That’s where you’d use them. Right? I find a

33:28 where you’d use them. Right? I find a lot of times, Tommy, when I build applications now, so I’m I’m building you’ve you’ve heard about this project before, Tommy, in the past. I’ve I built this thing called Content Nudge. And it’s this idea of a program that’s going to to collect a lot of data together. So now I’m collecting all of the data for all my social media activities in one single location, but the back end of that is all in fabric, but I’m building basically visuals and interaction layers on top of it in a,, a reg a

33:59 on top of it in a,, a reg a regular web app. So great, I love that I have this SQL server and I can get the data out into my visuals. But the downside is Tommy, there’s no like saving of the filter context if I had something I have to say. Everything’s for me for me right now it’s a lot of like go to this page I have to click the same things and so sometimes I can save the defaults but I usually just tell the app okay when I go to this page it should be defaulting to the last seven days of data right so this is a feature that I don’t have other places and to be honest I miss

34:31 other places and to be honest I miss bookmarks in areas where I had to rebuild them in other visual experiences because there’s no context to that inside inside my other apps that I build, right? So, in that regard, having quick selection of filters for a particular page of data, that’s that’s where the value I think is. think is., this may be something that can be a hot take. But, do where I’ve actually been leaning more around the bookmark experience is personal

35:01 the bookmark experience is personal bookmarks? Yeah, Yeah, I’ve talked about this before and you’ve turned me on to those a lot, Tommy. Actually, I I highlight those a lot more now than I ever have in the past. past. Okay, let me ask you then. Actually, why have you adopted that or really found the value around personal bookmarks? Cuz you told me to and I said, “Okay, cool.” So, that was it. That was it. You never deal with it. You never know what I say., no. I I think think I’m starting to see a lot more regular

35:33 I’m starting to see a lot more regular interactions with people when they are looking at reports, questions. The questions I’m getting as I’m working on reports with them is how do I get back to this information? What do I see when I show up? Right? And I and and now that you brought that topic up, I’m like, “Oh, well, here’s a great way for you to personalize the view of the data and this is just for you.” And so it’s it’s less about the feature and it’s more about did education is what you need to what you

36:04 education is what you need to what you need to observe, right? So it’s more about an education issue, right? Same thing that around the filter pane, right? right? I like using the filter pane. I’m okay with having a slicer or two on the report page because that’s what you may need for generalized filter context transitions, ones that are used very frequently. But if you get all these nuances, like I’ve seen people with like 10, 15, 20 filters on a page to like get down to the data they care about. Okay, that’s fine. You can have all that. I don’t want 15 slicers on the page that I’d have no room for any visuals or tables or data or anything. So, I

36:35 tables or data or anything. So, I actually prefer those extra slicers going over in the filter pane. Again, it’s not an issue. It’s an education problem, right? People don’t know how to use it and they’re like, “Ah, it’s too complicated, too.” No, it’s not. Take the time to learn it. That’s what we’re going to build. People are pretty happy with it. I think it’s pretty good. You with it. I think it’s pretty good., personal bookmarks for me have know, personal bookmarks for me have really served two, objectives for me. I think to your point, the thing with a bookmark is again, you really have to create universal bookmarks that everyone would

37:05 universal bookmarks that everyone would use. Usually people ask for like a subset of it and that’s when I think things get complicated. It’s like well we just want to see the northeast region. Okay. They have to reconfigure all that. And to your point personal bookmarks are incredibly simple for someone to do just what are you looking at? Save it so you can go back to it. to it. And again I don’t have to worry about managing or creating that. That’s per user. I wish you could share them. You can do that with comments. But Mike, the real value ad for personal bookmarks for me is when I create a

37:37 bookmarks for me is when I create a bookmark, I do not know like on desktop, I really don’t know if anyone’s using it unless I’m using some pretty,, fancy tooling out there. And again, it’s serving a generalized purpose. any bookmark that you create. Mhm. Mhm. With personal bookmarks, the reason not just that I don’t have to do bookmarks that I found the better reason is I’ve talked about this before too. The fact that you’re now understanding how report consumers are trying to bake or pre-bake

38:08 consumers are trying to bake or pre-bake the data. How are they actually trying to view the data? I use personal bookmarks and I push my consumers to to it so I can build a better report or I can create the right measures for them. I cannot tell you how many times that we go through and it’s like okay have you created any personal bookmark show me what you’re doing like yeah we’re looking at our team but we’re looking at it only three months and then we have another bookmark for the previous three months because we’re d whatever the case may be. And then in your head it’s spinning because you’re

38:39 your head it’s spinning because you’re going I can create a measure that does that. Like I can have a few measures and that’s actually going to just show that on the page. And all of a sudden now I understand the context of really again not just the data but how what framework are people trying to view this report. What framework are they actually using the data for? And for me that is a completely incredible way that personal bookmarks have become an integ a value ad in any work that I’ve been doing.

39:09 ad in any work that I’ve been doing. What would you say Tommy is more valuable to the end user in this regard? Right. Would it be you just described something to me that that pulled some memories here or some thoughts here. Is it so you said like the last three months, right? I could build this just as a measure, but I could also filter the report by that too, right? I tend to heir on the side of how can I build a generic report that allows filtering for what people

39:41 that allows filtering for what people want to see as opposed to building only the measures that do one very specific thing. So, for example, right, you said, “Yeah, I know what I know. Yeah, go ahead. ahead. See where I’m going with this. Like I could build a sum on a column and just now it calculates every like any period of time you want. It’s now handling it for you, right? I built a visual, right? This visual needs to show this week versus last week’s values or data or something like that. But if I expand the time window, that doesn’t really make sense anymore. That that measure on that thing. It now is it’s a bit too

40:13 that thing. It now is it’s a bit too specific, right? if I change the filter context on the report and the measures are already pre-filtering things sometimes you get a bunch of blanks and zeros so that report becomes less impactful or meaningful there. So I I go back in my I wrestle with this one Tommy I wrestle with the idea of do I just give people general aggregations and controls for filters on a report page and then teach them how to filter down to what they want. Now to your point around the

40:43 want. Now to your point around the last 30 days of sales or last 30 days of information, right? I would agree that having a date table that has a specific column that says this is the last 30 days. that’s pro like I I need to be able to select something easy in the report that says this is the predefined filter context or or date range that I care about because right now date slicers are still a bear and they’re not easy to work with right so so for things like that that would be

41:13 so so for things like that that would be useful but that I feel like that should be part of a column of a table and a model somewhere more than I would hardcode a measure to be only 30 days of data data right so this is where I’m I I’m not quite sure. Now, let me let me keep expanding this a bit. Like, if you’re building a table and the table needs to measure in it, okay, that’s going to be a bit harder to do because the table needs to have the filter context of like multiple years, but then I just want the last 30 days on a table. So, then then the measure kind

41:43 a table. So, then then the measure kind a table. So, then then the measure has to exist. But this is where I go of has to exist. But this is where I go back and forth. Tommy, I I don’t know the right blend of this. Why are report developers so worried about getting too granular with their metrics? I and because I hear what you’re saying, but it’s just over. No, no, no. I completely disagree. I and at least hear me out here because everyone is trying to view that data in some prism, some, you

42:13 some prism, some, you some prism, some,, context that they’re looking at. know, context that they’re looking at. Okay, Okay, I’ll give you an example. I’ll give you one of my favorite examples that I can use on error was working with a giant call center and they we were looking at everything on a weekly basis and had all the data but the thing was they were taking this week’s data for each team looking at last week’s

42:34 for each team looking at last week’s data looking at the variance right and that was the really the only way that they looked at it so I could add filters in there or the to have them easily do that but when you realize that that is their primary way of viewing at least this report or that page you at least this report or that page that is their context that they know that is their context that they need. Sure. To your point, do you all of a sudden miss out if you want to look at the last six weeks or we want to look at all time? Sure. Okay. It’s not like that’s not available, but that

43:05 not like that’s not available, but that is their prism on that where they’re making decisions. That is why they come back to that report usually is because those filters that are applying. Anytime you have something too general, I always go back to what’s going to cause someone go to go back next week to that, right? because you have to apply those filters and see compared to me actually creating for them or their first landing page is I’m always I’m always looking at this week I don’t have to click on anything here’s this week last week variance okay anytime I look at I don’t have to click on something to view it because

43:36 to click on something to view it because it’s a dashboard it people just want to know quickly and then if they want to dive in so I will make the argument with you not that you shouldn’t have those generalized views but odds are that if people have to every time go to your report, click on a filter, then click on another filter and calculate the the difference because again usually people are looking at things granularly in terms of what decisions they’re making. So that’s where my argument would be there.

44:08 Yes, I I think that’s a good argument, Tommy. You’re you’re trying to minimize the amount of effort, but it’s not just minimizing effort though either. It’s the value ad. Well, Well, but I but I this is where I feel like I get a little bit more bit more like I I understand people do certain analysis for certain things, but that’s just what they’ve been able to understand to up until this point in time. time. Let me try and rephrase my statement here again. I think that was a bit confusing.

44:38 confusing. You only know what, right? Sure. Sure. This is the way we’ve done the analysis. This is the analysis that that we use to do our work. I agree. Not going to argue with that. If that’s the analysis that you need this week over last week’s worth of data, that’s what you’re that’s that’s baked out of what does my manager require? What are my requirements for my job? There’s a lot of like,, assumptions that are that are being used to get to that is the metric. This is how we look at our data. This is what we do to measure performance. Mhm.

45:08 do to measure performance. Mhm. I have to imagine once that question is answered there’s always peripheral questions that are also answered and I feel like a lot of data and work begins begins with I need to solve this particular problem and once that problem is solved other questions are typically asked of the data to do additional things and so why I’m less inclined to build the exact measures for all all the

45:38 build the exact measures for all all the filtering items and instead use bookmarks or filter panes or other things or actually not hardcode this week versus last week time frames and actually add that to the dimension table so that way users could say well I want to look at this week and it compares to the prior week. Now if we if we step back and say what are the comparisons right this is where I think I go with a little bit this tummy is there’s the what was the data for this week what was the data for last week I want to look at seven days and compared to the

46:09 look at seven days and compared to the prior seven days sure sure of the tools that I use like other tools Google analytics Microsoft clarity other analytical solutions that I see where other companies are producing visuals and things. I tend to like the idea that they don’t tell me what range of time I can build. I can pick and choose what I want when I want to. Now, you’re right, Tommy. There is a case for this is the visuals I need on

46:40 case for this is the visuals I need on the page and this is the time range I’m going to regularly see on the page. Yeah. Yeah. But to hardcode that, I don’t know if I like that. I don’t I I have internal resistance to that just limits the ability of that page to be only for that particular purpose and that’s fine but I have to admit there’s going to be other use cases for ex other items. Now let’s go back to model building here. Right? If I’m going to build a sum on calls received column, right? There’s a

47:11 calls received column, right? There’s a measure. I’m going to make that measure. I’m going to build that in that very simple measure first and then when I build if to your point Tommy, if I’m building the time intelligence on top of that as measures, I’m going to use the simple measure on top of the more complex, right? The filtering. When you build designs like that, then yes, the model does contain the information you would need to be able to build the very specific report for the very specific filters, filters, but now you’re building filters. You’re spending time and effort. And now with

47:41 spending time and effort. And now with agents, we can do more of this faster, but you’re spending time and effort to build a single measure for a single page that is only used one way. I really like to try and design things that can be used many different ways. So how I would build it is I would not in the calculation write calculate sum the column name. I would write calculate the measure name and then build the basic measure that’s just the sum somewhere else. Right? So again this is mental model. But I see this all the time. I see this all the time where I see a calculate with a sum of the actual

48:12 calculate with a sum of the actual column with the wrapper of the filter context. context. I don’t know why people do that. Well I I think just because that’s again you don’t know what you don’t know, right? I think this is a case of that’s just the way you built it. It made sense to you at the time to build it, but now you’re ripping out all these other calculations that are still sums of the same thing over and over again. So anyways, I I’m just going to try to expand like I like to have reports that are very specific to a particular use case, but I also want to give the model capability to not just serve one

48:42 capability to not just serve one purpose. I want the model to be expanded to other purposes other than that initial one report page that does the one thing that we need for that user or that team that team might yeah but there’s a difference between the model and then again that certain report or that certain page I completely agree with you in terms of if like if you’re not creating your call them foundational measures those basic ones that you’re going to build everything off of you messed up and you should start doing that right away.,

49:12 should start doing that right away., thoughts already? Well, it it goes to the same it’s part of the same vein, Tommy, around when I make lots of bookmarks that are very complicated, it’s a maintenance issue, issue, right? If you’re not building your bookmarks programmatically with like an agent or C scripts or type editor or something that can maintain these things at volume and at scale, you’re in a like it’s it’s hard to it’s harder to maintain when you get all the

49:42 harder to maintain when you get all the things things right. Every time you make more measures, you got to know what they’re there for. Add the descriptions. Make sure the business users know how to use them. Put them in a proper It’s just more things to manage. agree that they need to be there, but I’m always an error on the side of can I just give you the minimum amount of stuff? What’s what’s the bare minimum I can give you that gets you off the ground to get you working and then teach you the things that you need to do to like customize it to your needs? needs? Right. Right. Yeah. Yeah. And that’s where I think I land a little bit more. I’d rather have more

50:12 a little bit more. I’d rather have more enrichment to the core model and tables than to give you a whole bunch of thin measures in the report that are doing a bunch of additional filter contextes. Right. Well, there’s a would you would you disagree or where do you find it from the refinement period? So, it’s generally really hard to going back to the personal bookmark. It’s hard for me to start with those,, granular metrics because a lot of times you don’t necessarily know and I would even make the argument people don’t know how they’re going to be really using that data and that report. So starting

50:42 that data and that report. So starting off with a report like that, yeah, you’re probably going to lead to H, you you’re probably going to lead to H,, a place where you’re dealing with know, a place where you’re dealing with,, maybe not a lot of value ad, but over time, like I always try, especially when I was FTE, we would have, like a monthly call with each sales department and it would just say, what do you,, any questions about the data? How are you using it? What are the next projects? Okay. And in those times we would look at the general sales report and each team I would basically ask what are you guys using with that they would

51:13 are you guys using with that they would show me their personal bookmarks and you realize that oh this entire team could have their own report for each of these things that is unique to them because they’re are the account management team. Okay. And you’re seeing how they’re using and every single time like it’s not like oh sometimes we use the report this way. It’s they go into the report, they click on these three personal bookmarks, they write some of that stuff down to do additional calculation. That is their workflow every single time. Yeah. Yeah. And that refinement process leads to all again those reports where that’s the

51:45 again those reports where that’s the only thing that they use. That’s what they had in meetings. That’s what they every time that they went to. Starting off with that would be nearly almost impossible because again, I’ll make an argument that people don’t know how they’re going to use that until they’re actually using I agree. I agree. But I I’m hearing you say two things that conflict in my mind. On one hand, you’re saying I want to build all these very specific measures that meet very specific needs for this one report that we’re going to use over and over again. Okay, understand that report is incredibly important for how

52:15 report is incredibly important for how your business unit runs. But then what you just described to me now, which is well, we users don’t really know what they want to use and so therefore we give them a lot of general things and say here you go and then play with it. So what you just said I think speaks more to my side of the story which is generalize more. We do want usage on how the report’s being used. What is making it effective for users to leverage and and build with the report? Totally get it. Like agree but you’re you’re making the case now for more of a generalized

52:45 the case now for more of a generalized model than more of a specific model. And so what I will say though I think my reaction to this one is reports are a living organism. The ideas are fluid, needs change over time. And so some of the needs will be very consistent and you build for those consistent needs initially, but what happens is that it’s not always set in stone, right? That stuff can there are things that

53:12 That stuff can there are things that there are things that are very consistent that we do not change, right? How we recognize revenue, how we recognize a customer, like these are things that we should define and land them and that’s what it is. the other things that are a bit fluffy which is comparing customer A to customer B time periods how long of a data range do we look at when do we do comparisons on time is it this week same week last year holidays mess everything up because then holidays push the data around and you don’t have comparable numbers sometimes I think we

53:43 comparable numbers sometimes I think we and maybe this is just my opinion here sometimes we manage down to like the day level level right we manage down to the day but I think if we zoom out sometimes and manage at the month or higher levels, the week or the quarter, the year, you’re able to then take some of that noise out of that, really looking at the signals very closely out of the system, and then you can make broader, bigger decisions, right? Ah, sales are down 3% year-over-year for this quarter. We need to do something to up that for the rest

54:14 to do something to up that for the rest of the year. Let’s really physically reallocate people to do something like that, right? those those larger strategic decisions maybe need a zoomed out lens. Anyways, I it’s very fluid and a lot of this is like it depends on what situation you’re in and how you’re using them. So, going back to bookmarks. Yes. Yes. Right. I don’t think bookmarks have a lot of bearing on like measures, measure calculations or time period filtering for things. However,

54:44 for things. However, however, however, I like using bookmarks to make a more generalized report specific to a user who needs to use it, right? Okay, that that to me makes sense., can I can I make a report that is useful and can be filtered by any time range and then just add a couple bookmarks or let users create bookmarks using a filter pane, personalized bookmarks, get down to what you want. Right. To your point, Tommy, the reason these things exist is to get to the information that you need

55:14 to get to the information that you need from a larger pile of data. Get down to what you want. And And what would you recommend then I guess moving forward in terms of to provide that without using bookmarks? if you didn’t have bookmarks at all and I I think we’ve talked about this a little already in terms of how we’re actually building that report but I think does fabric change anything or more importantly with where the influx that we talked about with the AI prep with for your semantic model how much has will co-pilot change that as of right I guess

55:45 co-pilot change that as of right I guess right now is there currently an alternative for you that you’re using today instead of bookmarks and where do you see in the future I think I’m using bookmarks the same I was earlier except I’m not using them for navigation experiences. Right? So the I like bookmarks. They work well. Saving states of the data or the report page or filter context. Great use case. Like it still use it. We we’ll still continue to use it. Using bookmarks to hide and show visual

56:15 bookmarks to hide and show visual elements. Move things around the page. change a bar chart to a line chart or change the line chart into a table. not not not as inclined to do that anymore, Tommy. I I just I just don’t think that’s I think the maintenance of that has gotten so high. I’m going to move away from that navigation things. I’m going to use the bookmark slicer, the bookmark button,, that I’m going to use that, I’m I’m going to try to publish more of my reports into apps, power, organization apps and workspace apps

56:48 organization apps and workspace apps because then I have the navigation on the lefth hand side. I don’t need to have any navigation at all. It’s all there., so those things. I think I’m moving more towards these other newer experiences of like navigating moving. I’m not going to put a bunch of filters in a flyyou filter panel and then hide the filter panel. I’m going to use the filter pane. So, a lot of the experiences that I think I was initially building with bookmarks or they’re dead. They’re no longer required because other parts of the tool has taken over those things. So, that’s that’s I’m going to still use bookmarks, but it’s not in the same way.

57:18 but it’s not in the same way. Not the same way. Not the same way. I I I really completely agree and I I think unless you are building those reports where people have that big insistence on th that design like and you’ve seen this too and there are some companies or I probably when you have larger clients where they want this really special design, right? They just don’t want that in a sense what people would call a normal dashboard. And I I think McDonald’s actually showed one example once of their report. I think it was in the gallery at one point. and it’s an

57:49 the gallery at one point. and it’s an intensive report and it’s pretty amazing what you can do with bookmarks but I think for my role odds are I am do if my focus is on delivering that information it’s not to wow people from a design there I know there are people’s roles who depend on bookmarks and and providing this really worldclass experience but again to to your point what I’m trying to to how do I provide the best

58:19 trying to to how do I provide the best skill or value for someone for their data right now? Do I want to spend six to eight,, six hours building bookmarks that people will sometimes use? Can I pre-bake that? And to your point, if I can, especially now, Mike, use CloudMCP to pre-bake those things., I think that’s really where I’m at. And so, like I said, it’s unfortunate to say that because I do like the bookmark experience and I I think it’d be very helpful. So, So, what what I I I think I understand what

58:49 what what I I I think I understand what you’re saying, but what I don’t get, Tommy, is I can’t interrogate what a bookmark is doing without a lot of code. Mhm. Mhm. Right. That’s that’s and I honestly I don’t think I think Microsoft was actually a little bit surprised on how many people used bookmarks to create visualization you bookmarks to create visualization data app experiences when it came know data app experiences when it came out initially and they were shocked they were actually thinking bookmarks are going to be used for snapping states of data and navigating a

59:19 snapping states of data and navigating a different page like that I think that’s what they were thinking it was going to be doing right that functionality and I think when people came in and built all this visualization stuff with the bookmarks and made it way over complicated. It it we’re just moving away from it now. And so I I think that I think that surprised Microsoft a bit on how it was being leveraged. And I think in order for them to solve like so I think what they did look they made bookmarks. Everyone came out and built visual data app experiences and

59:50 visual data app experiences and Microsoft who wa we don’t like this this is not good and people are complaining about bookmarks not working and like this is not what we want. And so therefore, Microsoft’s came in and said, “Look, what are the main use cases that bookmarks are building? Should we just have it part of the product?” And I’m pretty sure bookmarks drove the filter pain pain initially initially 100%. 100%. I I people built it. Microsoft saw it and said, “Oh, why why isn’t that just part of the product? We should have that.” Then the filter pane came out, right? Oh, people are building navigation with buttons all across the

60:21 navigation with buttons all across the report. That makes sense. We should just have a button for that. It makes it easy. So they Microsoft I think the bookmarks re-imagined a couple additional features that was implemented to take the weight off of bookmarks and now bookmarks are actually doing what they should have been doing what they were designed to do in the initial days which is click on this data save a bookmark click on this data navigate to this page and go to the bookmark that make go to teams talk about something highlight some data

60:51 talk about something highlight some data that’s your bookmark that makes sense to to interact like when you’re talking like app integration with the bookmarks and then the report page this makes sense to me so so that’s why I think that when I look at this I go that the bookmarks are now able to be used in their truest form of what they were I think were initially designed to build initially because other experiences now evolved into it 100% dude I I I think we will have to do another episode about do you still use drill through because I think I’m going to have a different argument there it’s amazing

61:22 different argument there it’s amazing these old features they call it a leg. I I’m not going to call bookmarks a legacy feature at this point, but I think we are moving away from a lot of those kind are moving away from a lot of those integral features that were there in of integral features that were there in PowerBI back in the day that again they ran a course. I will argue drill through is still essential and I I think that’s still there because to me when you talk bookmarks you’re there’s a huge performance story that goes with drill through. I agree with you there Tony.

61:52 you there Tony. Yeah. So, maybe we’ll have that conversation later. But yeah, bookmarks great for n but yet to your point, the navigation one’s dead. Mhm. Mhm. I’m curious as we get near the end here from a button point of view because buttons do more than bookmarks. And we haven’t talked about the button experience in the visual side. I think we leave that for another episode. I think that’s another episode. All about buttons sounds like another episode. episode. All about buttons is a great top name, too. So too. So that’s great. Sounds great.

62:22 Sounds great. That being said, that was our,, deep dive on bookmarks. Things you never thought you were going to need to know and talk about is now here. So now, now that we’ve gotten this one, so stop using bookmarks for v for for UI and visualization layer. Stop doing it. Go use other things. There’s better stuff to do filters. Use the filter pane. So if you are using bookmarks for selecting data or highlighting on the page, I think that’s still a very appropriate use of bookmarks 100%. That being said, thank you so much for listening to the podcast. If you like this podcast, if you enjoy what was going on here, please do us a favor,

62:53 going on here, please do us a favor, give us a thumbs up, a like down below. it really helps the podcast and it actually helps get it out to other people. So, we really appreciate your engagement with with the podcast that directly. Well, with that being said, Tommy, where else can you find the podcast? podcast? You can find us on Apple, Spotify, wherever you get your podcast. Make sure to subscribe and leave a rating. It helps us out a ton. Do you have a question, idea, or topic that you want us to talk about in a future episode? Head over to powerbi. tipsodcast, leave your name in a great question, and finally, join us live every Tuesday and

63:24 finally, join us live every Tuesday and Thursday, a. m. Central on all of PowerBI social media channels. Thank you all so much, and we’ll see you next time. Dance to the laughs in the mix. Fabric and AI, get your fix. Explicit measures. Drop the beat now. H feel the crowd. Explicit measures.

Thank You

Want to catch us live? Join every Tuesday and Thursday at 7:30 AM Central on YouTube and LinkedIn.

Got a question? Head to powerbi.tips/empodcast and submit your topic ideas.

Listen on Spotify, Apple Podcasts, or wherever you get your podcasts.