Every business should go through a pre-implementation planning stage prior to fully implementing new software. It’s a critical stage that involves making sure that all of the features and data connections you were expecting to be available are setup and working,
Companies that skip the pre-implementation stage risk a Power BI deployment that doesn’t meet their full business needs or creates a rocky start for people who are already skeptical of using a different software package than what they’ve been accustomed to.
Microsoft publishes a number of guides related to Power BI implementation planning, but we find it more helpful to pre-plan. This is the stage where you validate if Power BI is going to meet your specific business needs, integrate with your existing systems and determine who will actually be building, publishing and sharing dashboards.
Table of Contents
What is Power BI Pre-Deployment Planning?
The difference between a smooth deployment and a rocky one often comes down to how much preparation was done before hand. Power BI is deceptively easy to deploy, the platform mostly includes Power BI Desktop and an online service at PowerBI.com used to share reports. The difficulties arise when attempting to connect Power BI to a live data source, such as various accounting software packages, or CRM’s.
The Pre-Deployment stage is where you can get a cross-functional team together to help setup data connections to avoid a lot of re-work. Many consulting firms will recommend using Excel as a basis of Power BI reports, but if you go down this route there’s no guarantee that you will ever actualize live data connections.
The other part of this phase is determining who will build reports and manage them on an ongoing basis. Consultants and developers for hire can be a great resource for getting a quick start to deploying Power BI, but as business needs change, reports will need to be updated.
You will often also need the help of IT to establish connections, manage data governance, and setup connections between various systems that consultants will not have access to.
Here are some of the pre-deployment steps that we recommend.
Make Sure that Power BI is the Right Tool for the Job
If you’re reading an article like this, we can assume that you’re serious about deploying Power BI. However, we caution people to make sure that Power BI is being appropriately scoped for their needs. While it can create beautiful interactive dashboards and replace a lot of manual work preparing financial statements or KPI’s it’s unlikely to replace all of the Excel spreadsheets your company currently puts together.
The best way to make sure that Power BI will improve your business is to involve as many of the downstream report viewers as possible before finalizing the selection of a business intelligence tool.
Involve Managers Early to Scope Reporting Needs
Online forums like Reddit are littered with disgruntled business intelligence developers who spent weeks or months on building highly interactive, beautiful and explorable dashboards to only be told by an executive, they just want the Excel!
We are big supporters of business intelligence tools, but we totally get it. At the end of the day, the goal is to provide managers with the information they need to make the best business decisions possible regardless of the format or technology producing it.
There are many tools that integrate popular accounting packages with Excel to automate reporting. In a number of cases these tools can be a better fit than Power BI to meet an organization’s reporting needs.
Involve management early to understand what they want and whether Power BI is going to be the right tool to get there. Power BI consultants and developers often have a vested interest in getting you to go with the platform that they support regardless of how much time and money it will cost you and may not even be familiar with the other reporting options that are out there.
It’s worth noting that Power BI has a feature called Paginated Reports that lets you create highly formatted traditional reports. It requires a different skillset to setup than a standard Power BI dashboard, has more in common with SSRS and can be difficult to find developers for.
Engage with IT, Business Analysts and Other Stakeholders
Once you have an understanding of what your managers and report viewers want to see, engage with business analysts and your IT department to determine where data actually lives. While not a requirement, Power BI is the most useful when connected to live data sources.
With live connections, you can schedule refreshes and avoid having to manually intervene in a reporting process. Different data sources require different types of connections to be setup and are not always possible. By engaging with IT and Business Analysts early you can make sure that you understand where the data is that Power BI will be connecting to.
It can even be helpful to use a tool like Microsoft Visio to create a map and flow of your data sources.
Validate Data Connection Availability and Reliability
Power BI has built in connectors to over 150 different data sources making it a highly versatile platform. However, not all connectors are created equally and even if you do connect to a data source the specific information that you want may not be available.
Data connections can make or break an implementation and there’s a spectrum of connectors with varying degrees of support. For example, it can be more difficult than you’d think to connect to common systems like QuickBooks, Oracle HFM, or IBM TM1 just to name a few.
Some of the more specialized connectors are supported by third parties but it’s also hit or miss on whether they will work as expected.
You should also be aware that if you’re used to connecting to something like Net Suite it can be a lot of work to rebuild a report using source data. Very few systems will have something that lets you automate pulling a report the way that you’re used to viewing it and involves piecing different tables together.
We’ve seen a lot of software implementations go sideways when a business thinks that all of their data will feed into a reporting system, and actually doesn’t. It’s the reason why we recommend putting such an emphasis on making sure that connections work reliably before spending large amounts of time and money on a deployment.
Identify Potential Development Resources
Once you have the basic architecture mapped out of where data comes from, and what managers want to see you’ll be able one step closer to assigning resources to develop reports and publish them to Power BI. At the development stage there are a few different options you could go with, and we’ll provide a brief overview with some thoughts on each path.
Existing Internal Employees
This would be our preferred choice. It’s hard to understand a business process as an external consultant. People that know their current process and reporting needs the best are the people currently doing it. That being said, there has to be a level of technical knowledge or interest in learning how to use Power BI.
Pros: Cost effective, improves employee moral, provide new career possibilities, already understand the process.
Cons: May lack technical skills, can take time to learn Power BI, may not know Power BI or Data best practices.
Consultants and External Developers
Consultants and professional Power BI development agencies come in all shapes, sizes, and price points. You can find inexpensive overseas Power BI developers through services like Fiverr and Upwork, while there are many domestic US based consulting agencies that offer Power BI services. These can be expensive both from an hourly rate perspective, but also because it takes longer than most people think to develop a dashboard.
Pros: Understand Power BI best practices, can create great looking dashboards, and know best practices.
Cons: Expensive, no guarantee developers will be available on your timetable, will need to learn your process.
Temps
Hiring a Power BI developer on a temp basis can be a cost effective in-between of having a full time employee and an external consultant. A temp should be dedicated to working for your organization for a set period of time, usually 3 to 6 months and can give you the head start you need. By having temps work with existing employees they can also help reduce the total learning curve.
Pros: More cost effective than consultants, focused on a single client, will become familiar with your data.
Cons: Can be difficult to interview and find, the institutional knowledge they learn has a time limit.
Of course with all of these options you can mix-and-match them. Much of the consideration also comes down to how much you want to invest up front to get dashboards and reports created quickly.
Put Together a Broad Training Plan
When you start budgeting and putting together a training program for your employees there are some key questions that you should ask your team. Also, you will want to think about training as a regularly recurring event for different skill levels. Many employees who are new to software platform won’t know what to ask until they’ve had a chance to dig in and use it over several months.
- Should we setup a Center of Excellence (COE) or empower more Business Users?
- Who will attending training courses?
- How often will trainings be held?
- Do employees have the time and capacity to attend training?
- Will employees be able to put what they learn to use in the near future?
- What will training look like for new hires?
By thinking through some of these key questions, you can have an outline of an ongoing training program for new and existing employees to continue to improve their skills.
Prepare Stakeholders for Data Security Discussions
Who should have access to what data is an important consideration for businesses of all sizes. The biggest question you should start with is who is going to manage security? From there, you can work your way down and identify specific data tables, or row level security views in individual reports as more dashboards and reports are deployed.
Knowing who will be in charge of data security is the most important though. Some companies will assign security to IT but they may not understand the data, the business needs, or how the Power BI architecture works when setting up and managing security roles on an ongoing basis.
For example, what happens when a manager shifts between departments? Does their Power BI row level security get updated, and if so, who and where is it updated?
Power BI Pre-Deployment Checklist
We’ve covered a lot of food for thought for a system that isn’t even deployed yet, but by thinking through these considerations prior to purchasing Power BI licenses and installing on people’s computers you will set yourself up for a more successful deployment with better employee buy-in.
- Validate that Power BI meets management reporting needs
- Identify and validate connections Data Sources
- Consider consultants, contractors, temps, and internal development resources
- Build continuous training and monitoring plan
- Outline data governance policies
Conclusion
By following these Power BI Pre-Deployment guidelines you will be setting yourself up for success. It’s incredibly difficult to deploy Power BI or any other software without the infrastructure already in place, but is the path that is often recommended.
We see pre-deployment as a step that also helps you make sure that connections actually happen they way that they should and as expected. It’s a vital step to avoid building reports off of static datasets only to find out after spending a bunch of time and money that the platform won’t work as expected.
As you’re pre-planning also make sure to think about the people that you have available, your timeframe, and whether or not you need to augment it with temps, professional developers, or lots of up front and ongoing training.