5 things to consider before choosing an ideal Microsoft Power BI partner

If you have just embarked on a journey to pull out meaningful insights from your data using Microsoft Power BI and wish to work with a partner who can help you out, then you need to ensure, then you need to ensure you take some right steps before choosing an ideal partner. One of the first steps you need to take is look for the partner’s or the Power BI consulting service provider’s track record and successful customer case studies. When you engage with a partner, you should even ask them for past portfolios and few customer references to gain better clarity on the experience and expertise the partner carries.

We even have few words of caution for you when you initiate discussion with a prospective Power BI partner.
DO NOT get enamoured by FANCY graphs.
Don’t get us wrong. A sleek presentation is important, but the journey of any data analytics project doesn’t (at least shouldn’t) start with pre-prepared graphics and already available insights.

It all starts with data. What’s available to you, in what form and format, how accessible it is, or in other words is it readily usable or too raw to be consumed directly. Specifically in the context of Power BI, go beyond sample dashboards. Assess the prospective partner on these 5 parameters (among others that your organization might follow as part of your procurement process).

1. Data modelling across multiple data sources

You may start with a small dataset and try out a few limited dashboards. But if you are successful in your objective, the scope is bound to expand over time. In fact, if you are the project owner, you’d like the small project to be a showcase and encourage more teams, departments, people to jump in and give their inputs, to make it bigger and better. Maybe you are starting small because you do not yet have a data warehouse in place. Even if you have one, it may not have all the data you need.

Our experience tells us that most of the Enterprise Data Warehouses are good at housing transactional data coming from many different systems but are not designed to handle the recent trends. For example, it would be interesting to combine the data coming out of Heap Analytics for your website traffic with subsequent order conversion. Targets against the same may not be available in any formal system (and that’s when we use good old Microsoft Excel). Or say, you want to analyze sales, then you need a data model which is optimized to query and aggregate data in large volume. In such cases, see how expert your partner has created a data model for their past customers.

2. Power Query and Transformations

In case you are not familiar with Power BI’s capabilities, you’ll be surprised to know the levels of transformation that can be achieved by Power BI. For many organizations, this may help avoid instituting a separate, parallel data warehouse project. The easiest way to create a data model is to get data from the sources, define relationships and then design reports. However, this method carries some issues when you have multiple tables, or different values in the table representing the same entity. Hence, you’ll need to modify your tables using Power Query before creating reports. All this is possible, but only if you have someone who understands data and business equally well.

3. DAX Expertise

What is the point of any analytics project if monitoring and comparing trends have not been carried out across various time periods? It is a rhetorical question but Depending on your domain, running aggregates and comparisons of variances can be even more exciting and fruitful to make informed business decisions. Temporal data can be a difficult beast to manage. While Data Analysis Expressions (DAX) can be used for a variety of purposes, they are especially useful to create calculated measures from base data to bring about a totally different level of insights. Not only the partner, but the person who’s going to work with you too needs to be very comfortable with time series data, complex calculations, and DAX expressions.

4. Custom Visuals

If you have seen any fancy (eye-popping colours, graphs of all shapes and sizes) visual in Power BI, you can safely assume that it is some custom visual. Options to control and modify the look and feel of visuals in Power BI are very limited. But what works in its favour is the ability to code any custom visual and use it as a widget on top of the Power BI platform. Using R or Typescript along with D3.js, you can extend Power BI to do pretty much anything. What it requires is excellent knowledge of coding in TypeScript. If the partner is good in developing custom visuals, you won’t need to restrict your imagination about the great visuals that can be created.

5. UX/UI Specialization

Last but not the least. It’s very difficult to find people who really enjoy being submerged in data and have tremendously refined sense of good UX and UI. If you know someone like that, please pass on the reference to us and we’ll hire them immediately. For lesser mortals, collaborating with a UX/UI specialist is the only option. It does mean additional cost, but it’s worth the investment. Overall cost-wise, it may not matter whether you employ the specialist, or the partner does it for you as an additional billable activity. We recommend letting the partner manage them for 2 simple reasons. First, partner cannot complain about delays in inputs from the UI end if there are multiple iterations. Second, it also avoids the trap of a partner using ‘technical feasibility’ as an excuse to justify its delays.

In conclusion, here are the 5 points for your quick reference:

✓ Data Modelling across multiple Data Sources
✓ Power Query and Transformations
✓ DAX Expertise
✓ Custom Visuals
✓ UX/UI Specialization

Motifworks is a certified Microsoft Power BI Gold Partner on grounds of offering expert Power BI consulting, development, implementation services. And you are reading this on our website so it’s a no brainer that we claim to be superlatively good on those 5 points discussed above.