Most people on our team completed a "Dashboard in a Day" training held by Microsoft or one of their partners, which was great for getting the basics down, but we also build most of our reports off our data warehouse so Power BI is used more for just visualizing data rather than modeling. We also make it a point to have a regular "intra-team training" for team members to share things they've been working on and highlight any helpful technical tip/feature they figured out.
We've tried to shift some of the report building away from our team (centralized data and system support) to some stakeholder teams over the years as those teams have shown a desire to learn and the commitment to building that capacity internally. In that situation, we've encouraged them to attend the same type of training, but also provided some direct training and ongoing coaching as they build their first report or two.
We've restricted access to who can build mainly because we can't afford to give everyone a Pro license, but I do think there's some value in keeping the number of designers down in general. There's been a lot of points raised in this forum about building trust in the accuracy of data across this forum and I think keeping the number of designers on the lower end makes it easier to ensure that you have consistency in what data and assumptions/logic are used across reports. Another reason to keep the number of designers low is to better curate your Power BI online space. If you have too many designers, you end up with a lot of workspaces and apps where reports are published which causes confusion for your end users because they don't know where to find a report. If the number of designers is low, you can discuss where a new report should be published based on the intended audience and make sure that the designers have access to publish reports to the appropriate space.
If you're wanting to build some capacity in your finance team, I'd suggest getting a couple of people up to speed through a free "Dashboard in a Day" training, but also make sure they work closely with your data science team to understand any work done on the data before it gets to the point that they're pulling it into Power BI. Also keep in mind that report design isn't just about data, but also about building something that is visually appealing. I received some feedback early on that my reports were a bit "loud" in terms of my color choices so I then read up a bit on how to choose colors from a design theory perspective. I also have a colleague that said she mimics color schemes from shopping web sites that she enjoys so keep in mind that inspiration can come from anywhere :)
Original Message:
Sent: 03-31-2022 09:46
From: Olivia MacDonald
Subject: How Do You Use Power BI? 📊
Hey everyone! I was wondering how did your team find the move to PowerBi from a building perspective? Right now we have a team of experts who lead our company dashboarding but we are hoping to start training the finance team up to create their own dashboards as required rather than relying on another team. Currently none of us have experience but definitely a strong appetite to learn. I'm wondering where is the right place to start (a course vs trial and error)? Did you restrict who had access to building in PowerBi and if so, why?
------------------------------
Olivia MacDonald
Original Message:
Sent: 03-09-2022 11:46
From: Brendan Eger
Subject: How Do You Use Power BI? 📊
Hi Gus. I was involved with our data warehouse build-out and will admit that nobody on the team was a data science expert at the time so we've probably made some mistakes along the way - like building data transformations that aren't very efficient. Our biggest challenge right now is that we have 4 people who work on data from different departments and the pace of what they build isn't always in line, which leads to some work being ready to deploy to our production environment while other work needs more time. We've been talking recently about adjusting our shared processes and best practices by organizing our work into 1-2 week sprints so that we only commit to work that we can get done and fully test within that time frame. In the early days, this happened less often as we had less data, but as our data warehouse has matured and we've added more data from different departments, we're finding that we need to be much more coordinated so that we don't accidentally deploy something that isn't yet ready.
We haven't spent as much time using Data Flows since we've had a data warehouse already, but we've been playing with it a bit for some unstructured data or data that we our team doesn't manage ourselves - i.e. we work closely with our clinical data colleagues who don't yet have a mature data warehouse so we helped them test Data Flows as a proof of concept. My former boss who's now at a new organization said his team is trying to use Data Flows exclusively and they abandoned their other data warehouse platform (Jedox). That was back in December so I'm due to follow-up with him to see how that approach is working out and could provide you an update afterwards or connect the two of you.
At the end of the day, I think the choice of platform doesn't matter too much, as long as it's something that your team can learn and manage. Having buy-in from stakeholder teams to agree to and create the single source of truth is the key.
------------------------------
Brendan Eger
Original Message:
Sent: 03-08-2022 22:34
From: Gus Arndt
Subject: How Do You Use Power BI? 📊
Brendan, I appreciate your response. The path you have laid out with your data is a path we would like to follow in the future. Part of that consistency problem you mentioned is evident for us as a "grain" problem, specifically, that our hierarchies are overly complex and not always easily aligned at the same grain on the balance sheet / income statement. Were you involved in the data warehouse build out? If so, are there any pitfalls you would point out?
Also, How are you liking the Data Flows? I know that was something I had on my to-do list to check out, as we leverage lots of similar Dynamics views to drive our sales team, and having one centralized version would be a big time saver.
------------------------------
Gus Arndt
Original Message:
Sent: 03-08-2022 14:07
From: Brendan Eger
Subject: How Do You Use Power BI? 📊
We push our Vena data (budget only) to our SQL based data warehouse (Jet Data Manager) on a nightly basis where we're then able to combine it with data from other systems (ERP, CRM, HRIS, Supply Chain Management System, etc.). Within the data warehouse we normalize all data into standard reporting tables and cubes for Power BI to pull from. We've found that using cubes is most efficient for reports needing to pull a lot of data (millions of transactions with 50+ columns) and/or is used by a wide audience (hundreds of viewers) as the reports perform better when using a live connection rather than trying to load a set of cached data to the Power BI online service. With that said though, we do still have an occasional use to model within Power BI itself for Proof of Concept reports or data that isn't as easy to get into our data warehouse. Our Power BI journey started in 2016 with monthly financial reports, but has since grown to a broader set of financial reports and wider use by other teams across the organization. We've recently started playing around with Power BI Data Flows as well which can be used to do some of the modeling in a central location (similar to a data warehouse) rather than within a .pbix file.
I think my biggest tip would be to try to centralize standard data transformations as much as possible so that your reports are all sharing the same data and you have a single source of truth. If you create multiple Power BI files/reports where you have to deploy the same logic, consistency becomes challenging and you risk losing your audience's trust in the accuracy of the data if they see different numbers across reports.
------------------------------
Brendan Eger
Original Message:
Sent: 01-27-2022 14:42
From: Irene Tang
Subject: How Do You Use Power BI? 📊
Hello everyone!
I want to start a new thread and see who in the community uses Power BI in their organizations?
➡️ What do you use it for and why do you love it? Is it used for data visualizations, reporting, and/or data modelling?
➡️ Are there any tips and tricks you would like to share?
➡️ Do you have any questions to ask the group?
💡 Want to see the Vena Power BI Connector in action? Check out this video!
🔗 This thread is a great spot to see who else in the community is a Power BI user and is open to networking! Make some new connections and learn from one another on best practices!
Reminder: you can look in the Directory and make connections with other Forum members!
Thanks everyone!
------------------------------
Irene Tang
------------------------------