Planning for Microsoft Fabric: How To Prepare Your Team

Ian Wellman:Hello everyone and welcome to our Microsoft Roundtable, talking today specifically about Microsoft Fabric. I’ll let everyone introduce themselves in the call, but for me, I’m Ian Wellman, Director at Nigel Frank and I’m very keen to understand how Microsoft Fabric impacts the talent markets that we have in the Microsoft ecosystem.

Prathyusha Kamasani: Hi everyone, I’m Prathy Kamasani. I work at Avanade as a Director, basically doing all the Microsoft BI things like Power BI, and nowadays very excited about Fabric. I’m not really working on a Fabric project, but I am playing with it a lot!

Chris Webb: I’m Chris Webb, I work on the Fabric Customer Advisory team at Microsoft. Previously, I was on the Power BI Customer Advisory team, but everybody’s a Fabric person nowadays—and as the job description suggests, a lot of my time is spent advising customers about Fabric. A lot of my time is also spent finding out what the feedback is and taking it back to the rest of the people who build Fabric. I’m based in the UK and cover customers in the UK.

Vicky Simpson: Hi everyone, I’m Vicky Simpson. I’m a Principal Consultant on Nigel Frank’s Business Intelligence and Data Recruitment team. We recruit for all sorts of BI and Data roles across the UK. We do this for Microsoft partners, for large enterprise organizations and for end users too. And obviously, recently we’ve started learning more about the changes that Microsoft Fabric might bring to the work that we’re doing—so excited to talk more about that today.

 

Chris Webb: So I think there’s a lot of different ways of answering this. One way of answering it is to say that it’s Power BI, but with a load more new capabilities. So everything that Power BI was is still there, except that instead of just having data sets and reports and Paginated reports and data flows, there’s now a greater variety of things you can have in your Power BI workspace. You can have warehouses with a proper full-featured relational, highly scalable warehouse engine. You’ve got data pipelines, you’ve got a lot of Azure data factory type functionality, you’ve got machine learning functionality, you’ve got Spark, you’ve got Notebooks, you’ve got alerting. So in one way, Power BI has grown incredibly—there’s loads more stuff in Power BI.

Another way of looking at it is that it is the future of the whole Azure data platform. In the past you’ve got Power BI on one side, then you’ve got Synapse, you’ve got Azure Data Factory, you’ve got all of these other different Azure services—Fabric is all of them coming together into one software-as-a-service platform where everything all just works together seamlessly.

Prathyusha Kamasani: I think if you are a customer and looking at it, trying to understand why it’s there—mostly what we do is move data from one place to the other place. And without Fabric, we worked with different clouds and then moved the data from there to Power BI. But with Fabric, there are fewer moving parts, so we get access to our data much quicker.

There are tons and tons of new, interesting, innovative features inside Fabric as well. I’ve seen some people say, oh, is it just rebranding of existing stuff—and it’s not. And it’s not even just bringing everything together in a more integrated way. There’s a lot of really new, interesting technology in there as well. Like the brand new warehouse engine, the Direct Lake mode for Power BI and things like that. It’s also very exciting for Power BI people and for me, coming from the Power BI world. So many Power BI people may think there is nothing new for them, but there is—there are quite a few new things they could enjoy. Like I’m enjoying Notebooks, even though most of the time I don’t know what I’m doing, but still, I’m enjoying it!

 

Prathyusha Kamasani: It’s in public preview now, yes.

Chris Webb: And so I would say public preview means we’re showing it, you can play with it, but not everything is available yet. We’d love you to try it out and play with it and start to build things with it, but the performance isn’t there yet. There are a lot of features that are missing that will be coming later on in the public preview, so don’t judge it yet. Don’t do any performance testing, for example. But it’s there in a stable, usable state.

 

Ian Wellman:From a business’s point of view, have you got training courses or is there a new skill set that you’re looking to hire? What’s the plan?

Prathyusha Kamasani: That’s a very good question. Talking from my experience, when you are a partner, you already have a Data Engineering team and Power BI team—and Fabric isn’t doing much, just combining them together into one theme. This used to be the case traditionally, and that’s what it’s going to be now. So if I’m looking for a new person to join the team, I would probably look for somebody who is also interested in learning new things like Notebook or Python. Or if we have somebody who is doing Power BI, but already understands Python and those kind of things, that would help.

From the training point of view, there’s plenty of training on the Microsoft site itself—there are quite a few happy paths and stuff. Internally as well, we are spending quite a lot of time and effort to train people, making it available to understand, because I think while the languages and tools may not change completely, what would change for a client is the architecture. Now we have to update traditional architecture with the new components available in Fabric—I think that’s where the change would be from a partner’s point of view.

Chris Webb: Yeah, I think there are lots of interesting implications for hiring and for skills. I think in terms of skills, when I first started using Power BI, it was possible for somebody to know everything about it. Now, so much has changed and happened in Power BI. It was already at the point where it was impossible for somebody to know absolutely everything—there were specialisms—and you might be able to know a little bit about everything, but there’s no way that any one person could really know everything about Power BI because it was so large. Maybe even within some of those consultancies like where Prathy works, there would be people who are better at Data Visualization or better at DAX or better at Power Query and things. Now, with Fabric, that’s got 1000x more. If somebody comes and says I want a Fabric expert, you’re never going to find that. You’re going to have to look inside Fabric and see what are the different happy paths that Prathy mentioned, or the different skill sets. You’re going to want to find somebody who’s got those Power BI skills, that’s going to be great. You’re going to want to find people with data engineering skills, people maybe with Python skills, people perhaps with SQL skills. There are going to be different ways of using Fabric, some of which use those skills and some of which don’t.

If you’re looking for somebody to learn or skill up on Fabric, apart from the Power BI skills, if you’ve got somebody who’s got some Synapse experience, that’ll be great for large parts of Fabric. If you get somebody with Databricks experience, they’ll probably be able to learn and get up to speed quite easily with Fabric as well. Anybody with SQL Server or kind of good strong TSQL skills, they’ll be good at other different parts of Fabric as well. I think it’s still a little bit early—nobody’s really worked out how a Fabric team would be structured, but it’s not one person, and you won’t have one person knowing everything.

Prathyusha Kamasani: Yeah, and I just want to add one more thing to what Chris said. There is a low code, no code approach to Fabric as well. It’s like just like how in Power BI, if you are a power user or end user, you could go and create your own reports—the same type of elements are available the whole end to end. There are lighter elements to Fabric as well!

 

Ian Wellman:Is it the large enterprises? Are we looking at the smaller businesses, or is it available to all?

Prathyusha Kamasani: Everyone!

Chris Webb: Everyone. Now I know that sounds like a kind of cop-out answer, but it is genuinely everyone. Synapse basically was an offering aimed at enterprises. There are some medium-sized businesses using it, but it’s basically an enterprise offering. But Fabric aims to take the same kind of approach Power BI had, in that there are some of the largest customers in the world using Power BI. There are super large customers who’ve got more people using Power BI every month than live in some of the biggest cities in the world. But on the other hand, there are tons and tons of companies using Power BI where there are one or two people who use it, and they’re used in very different ways. You could get some people who just connect to an Excel spreadsheet, build a basic report and publish it for two or three colleagues. You’ve got some people who got super complex skills using CI CD, connecting to gigantic databases, building reports conceived by thousands of people. That’s the strength of Power BI and that’s what I think we want to bring to Fabric.

We want it to be absolutely everything to everyone. Irrespective of what stage you’re in your analytical process, there is something in Fabric that you could use to make it better. I think that’s what excites me, as somebody who is passionate about analytics. It doesn’t matter if I want to do very raw data processing or machine learning or data science; I could pick up anywhere in this story and use Fabric.

 

Ian Wellman: Vicky, from the candidate side of things, what questions are you getting from the candidates at the moment and what are the pressing demand that need answering?

Vicky Simpson: I think the number one thing that people are coming to me with is, they’re saying: look, I’ve heard about Fabric, it’s everywhere and I want to get involved in it, but I don’t really know how to begin. Maybe they’re a Power BI Developer who wants to see what else they can do in Fabric, for example. What advice would you have for someone like that to get started?

Prathyusha Kamasani: So we’re talking about somebody like me! What I do is I have a personal Power BI Tenant where I enable Fabric. My daughter is really passionate about Taylor Swift, so I decided to get all of Taylor Swift’s lyrics using Notebooks and create a solution: a sentimental analysis of the lyrics. So that’s a project I created myself, and I think that’s what I would recommend to people. Find something that fascinates you. And even if you don’t have a Power BI Tenant, you could open a trial Tenant. I don’t know how long that lasts, but you have time to use it for at least a month or so. That’s what my recommendation would be.

Chris Webb: You may be able to do this in Power BI at work, depending on if your Power BI admin has turned things on or not. If you haven’t done that, it’s super easy to have your own Power BI Tenant that you can set up at home. There you’ll be the admin, and then you can make sure that you enable Fabric and spin up your own trial, assuming you haven’t already had a Premium trial already. But spin up a Fabric trial and you can just start to get going with it.

Vicky Simpson: Another question that I’m being asked quite a lot, is what will the impact of Fabric be on your commonplace data roles? So for example, your Data Analysts and your Data Engineers—will they still exist as separate roles or do you think ultimately we’ll see those starting to come together?

Chris Webb: I think they will still be separate roles. Having said that, what we’re certainly seeing is Data Analysts who are getting more and more technical. The Business Analysts who were only doing stuff in Excel suddenly got a big boost to their skills with Power BI. And you start to see people who might have just been Excel VBA types doing some really amazing stuff with Power BI. I guess part of the whole strategy we’ve got with Fabric is to just put more cool stuff in their hands—things that would previously have been locked up by the IT department, saying don’t touch this, you’re not skilled enough to use it. And things that would have been just too expensive or too difficult for them to set up, just to make it easier for them to get their hands on it.

So your kind of traditional Business Analyst is going to have a lot more opportunities to do stuff, but I don’t think that’s ever going to really replace Data Engineers and the whole kind of analytics engineers—the more IT department type roles, they’re still going to be there. There’s just going to be even more opportunities for everyone.

 

Chris Webb: As Prathy mentioned, there are courses already. But I think the thing is, because we’re in public preview, things are going to change a lot. The user interface is going to change, performance is going to change. There’s functionality—big, big pieces of functionality—that we haven’t released yet, that will be released over the next couple of months. And as somebody who’s done training courses in the past and written books and things, the problem at a time like this when everything is going to change from week to week, is that if you build the training material now, it’ll be out of date next week. Because in training material, you need to be able to say press this button and go to this side of the screen, but all of those things are going to be moved around. And that just makes it super difficult to build any training material.

So there is stuff now, but it’s a bit too early. Nobody’s going to want to invest in building all of that just yet. So we really are at that point where you’ve just got to play around with it and see what’s there. Kick the tires. I personally do exactly what Prathy said. Just get some data you’re interested in and see what you can do with it.

Prathyusha Kamasani: It’s also going to be very interesting to find a certification because as Chris mentioned earlier, there are so many artifacts and there are so many things that you could do. So I don’t think there would be a certification path that would cover the entirety of Fabric. I think there will probably be certifications for each artifact, or they probably would be something like Azure certifications—like a Fabric Fundamentals make sense.

Chris Webb: So I would say, what are customers doing? There are probably two approaches that I see for customers. First of all, there’s the kind of formal, let’s set up a team to evaluate this type of approach. It’s the kind of top-down thing where somebody senior has seen one of the presentations and says, oh, this looks like something important from a strategic direction. I’m going to put together a team of people and we’re going to do a proof of concept and see what it’s like, and then we can report back.

The other approach; the other thing that we see happening is that there’s no official team, but some enterprising Power BI user suddenly thinks, oh, this looks really cool. I’ve read some blog posts about it, I wonder how it can change my job. Maybe they turn on a trial capacity, start playing around with things, start saying, well, this is stuff I’m doing today in Power BI or with Dataflows or with something else—can I do that in a different way, in a better way, with Fabric? And they’ll just start playing with it, and then hopefully they’ll be successful. Kind of like how Power BI spreads through a lot of companies—they show their friends, their friends saying, oh, I’m spending hours every week doing this, but if I use Fabric, I can maybe take that down to five minutes. Maybe I’m going to start playing with it, and it grows virally—so I would say that’s kind of what’s happening at the moment.

That’s maybe not answering your question, but that’s the two different ends of the spectrum. And I would say from a company point of view, even if you don’t have the time to officially set up a task force, don’t stop your end users from playing with it. There’s a certain attitude, which is completely understandable, of this looks new, let’s turn it off and then maybe in a couple of months, we’ll have a look at it and decide whether we can turn it on. Don’t do that. Leave it on. Let people play with it and let them explore.

Prathyusha Kamasani: I think one thing I would add again: don’t keep your production and things on the preview. I’ve seen people who want to do that but, as Chris mentioned, things are changing every day.

 

Chris Webb: There are always bugs. We’re busy fixing the bugs, but you may well encounter some. There will be some things that don’t perform as well as they should, especially when you’re working with larger data volumes. Also, because this is all very new, nobody knows what the best practices are yet—we’re all still working it out. It’s not even like anybody at Microsoft knows what the best practices are yet. We’re building it. But once we’ve built it and we’ve got things working and people try and succeed or try and fail, that’s how we work out what the best practices are.

So there’s also that disadvantage—unlike a mature GA product like Power BI, where you know where to look, you know where to go, and there are accepted ways of doing things that will lead to success. Whereas there are obvious mistakes that people make, that people have written about and said don’t do this because it’ll cause problems later on—that doesn’t exist with fabric yet. It’s all still pretty new.

Vicky Simpson: Are there any particular conversations you think we should be having with the market, with either our candidates or our customers?

Prathyusha Kamasani: I’m involved with a lot of recruitment where I work and I’ve done a lot of interviews. I think from the recruitment side, look for people who are passionate about adopting broader tools—not just I’m a SQL developer, I’ll do SQL. It’s important to be flexible, to be able to learn new things, and try to find a different way of doing things.

For example, when we go to certain client places that have SQL Developers, I tend to see them doing things in SQL which probably needs 100 lines of code, but the same thing could probably be done in three lines in Python, and you could use the data pipelines and probably just do it in one task. So it’s important to find people who could navigate between all these things. I think that’s what I would look for.

Chris Webb: I think in terms of the skills, Power BI people probably have heard of it and be kind of interested to expand, and Synapse people will probably have heard of it and be interested to go into it. I think the interesting thing will be, and this is where you might be able to help suggest things: there will be people who have worked with Databricks who think of themselves as Databricks, and maybe don’t realize how much of their skills will be transferable to Fabric. Perhaps you could suggest learning a bit of Fabric—it’s the cool new thing. They might not have thought about that. Similarly, with Snowflake, there could be people with strong SQL skills who have become Snowflake people in recent years but might be interested in taking a look at Fabric as well.

And I suppose that there’s always that problem of early career people, with lots of people leaving college and university with fairly good Python skills, thinking they’re going to be a developer. It’s about suggesting: well, hey, development is fine, but data is even more cool. I think kids maybe think data is the less cool option, but I actually think a career in data is a lot more satisfying than just being a boring old dev—maybe I’m biased! But all those people with Python skills who are coming out and thinking, what am I going to do with my skills in my career—say hey, look at Notebooks, look at Python, look at working with data there.

Prathyusha Kamasani: I initially never liked data because I felt like it was back end. I wanted to be a front-end person. Nowadays I hear my daughter saying the same thing, that it feels very back end. But Fabric does have front-end elements as well. You could create visualizations, you could create advanced ML and data science stuff, which I don’t know how to do, but I think nowadays kids come out of uni knowing all these things!

Chris Webb: What could be cooler than machine learning and Fabric!?

Prathyusha Kamasani: Yeah, and it has Copilot! See, that’s what I am doing with my Notebooks—I have GitHub Copilot, so I’m writing code using it. It doesn’t always help me, sometimes it sends me in a different direction, but it does help me to at least write a sentence saying this is what I want to do—and it suggests me something which makes it easy to learn things.

 

Chris Webb: Well, I think if you’re in the London area, you should be going to the Fabric user group, really, shouldn’t you?

Ian Wellman: Absolutely, yeah!

Prathyusha Kamasani: We will be having more events around Fabric, and we’re lucky to have Chris in town, and Mark and Casper will be here most of the time as well. So we are very fortunate to be in a position like that. And there are quite a few other user groups, like virtual ones, that are also talking about Microsoft Fabric recently.

There is plenty of information—I think there is information overload, to be honest! But if somebody is like me and gets overwhelmed by so much information, I think I would just go down the route of following certain people. I personally follow Chris and Guy in a Cube, and people like that so I know what I’m doing. And at the same time, also follow Microsoft blogs, which is where every new change gets explained. That way, you’re on top of it and can then figure out what is that you want to dive into, instead of following everything, which gets a bit intimidating.

Chris Webb: Yeah, I think it would be very easy to just get completely overwhelmed by everything that’s happening in Fabric. So pick the bits you’re interested in. Pick the bits that you think might be relevant to your job, where you’ve got skills you can build on, but don’t feel like you need to learn absolutely everything.

Vicky Simpson: I think, like you have both said, there’s just an absolutely massive amount of information out there for people to consume. So I think our biggest challenge is just helping to slowly guide our clients and our candidates on where to actually start. And like you both said, I think it probably is a case of pursuing some passion projects, doing some exploring yourself, and keeping up with all of these resources and blogs. Obviously our user group as well—we’ve seen a massive increase in attendance since we rebranded to the Fabric User Group, which is so exciting. So there is a lot out there to get involved in and I think that’s the key thing that the people I’m speaking to are looking for at the moment—where do I begin?

 

Chris Webb: I think it’s probably worth noting—and I’m from Microsoft, so I would say this—but Fabric is going to be big. It’s going to be really big. If you see how Power BI has grown exponentially over the last couple of years, and it’s gone absolutely crazy—it’s bigger than I think any BI tool has ever been now, and when I look at the internal numbers, the growth we get is just unbelievable. So amazingly successful. But Fabric is just going to turbocharge that—there will be even more users and it’s going to be an even more diverse community.

As a tech professional, you can do a good job, but how well you do in your career is often dependent on whose bandwagon wagon you hitch yourself onto. You could be a great talent at a bit of tech that just doesn’t get used widely, and then you’ll end up kind of boxing yourself into a corner. Or you can be somebody who’s just failing normally. But if you jump onto the bandwagon of a really successful, hot bit of technology, then suddenly you’re going to be way more in demand. You don’t need to have the world’s biggest brain, you just need to be pretty good at a technology that everybody wants. And I think that’s what Fabric is going to do.

That’s what Power BI has been already. The same formula that we’ve applied to Power BI is going to be applied to Fabric. Good, regular investment. Regular updates. If it’s not able to do something now, in a month, in a year, it probably will be able to. And at Microsoft, we’re listening to the customers, listening to the community, finding out what doesn’t work, and then investing to fix it. It’s not like we’re going away thinking of something, releasing it and it doesn’t work, and then three years later there’ll be version two. It’s going to be regular updates, and it’s going to move really quickly in the way that Power BI did.

So learn Fabric, it’s going to be great! You’ll have an amazing career and you’ll be able to double your rates…possibly!

Prathyusha Kamasani: Yeah, like it or not, it’s going to change the world of analytics—so it’s good to be on the boat.

Ian Wellman: Okay, well, thank you very much for joining us today and that’s definitely made my head a lot clearer on what Microsoft Fabric is and how it’s going to affect the talent markets. Thank you very much, Vicky. Thank you very much, Chris. Thank you very much, Prathy. I’m sure we will be hearing more soon from the wonderful world of Microsoft Fabric.