Microsoft Power BI Reviews

This morning I was on the train going to a briefing session and I was compelled to look again at the Gartner Magic Quadrant paper on BI – in the same way as mid-exam you might go back and look at the question to make sure you are answering it. Here are the things I pulled out for my slides. You might find them useful.

I see Gartner as the arbiters of good-taste in matters informatics. They explain the market and solutions, they rate vendors and they offer thought-provoking insight to people making technology choices – whether you are buying or making. I love ‘em. I’m making no apologies for my promotion of Microsoft. I believe it to be the most complete in terms of the company’s vision, the easiest to execute and I buy into the visionaries in Redmond and beyond (especially Cambridge in the UK) as Microsoft tries to lead the market. I bet my house on this a few years ago and I still live there. Phew.

Thinking about what BI is. It’s really about getting people with the right tools for their job to work effectively and collaboratively in managing the flow of information across an integrated infrastructure (so the flow doesn’t break), an integrated data architecture (so that when you blend the liquids flowing through the pipes they taste nice), without IT being constantly in their homes / offices / cars / clients houses. It’s about delivering the information to people who need it to make good business and clinical decisions in the right way at the right time. It’s about being able to find information and getting information to find me – I want to hear the erudite information shouting loudest at me amid the tumult of data chatter. It’s about the information being structured so that I can plug tools into it and predictive model, run SPC and do all the other things that I want to do in order to improve the safety, quality and cost-effectiveness of my services.

The Microsoft stack does this for me – see previous posts. This is recognised. Gartner points out that the Microsoft solution set is wide in scope – there is something in the toolset for everyone, however the set is integrated and so it works. See my article on why you wouldn’t buy reporting solutions for example – in and of themselves they don’t solve your problems.

Clearly the Microsoft Bi stack is designed with Gartner’s feedback in mind, he said smilingly, as we can directly map what they have done, to the above description of good BI.

Microsoft BI is recognised as being wide in scope and deep in functionality so that it ticks all of the above boxes and the UI has something in it for everyone in terms of the abilities of the combined tools to enable access to data. Some might say they have too many tools – see previous post – however the partner eco-system of people like us in Ascribe should be able to line up features and functions to roles and so that shouldn’t be a concern. The eco-system is actually another reason why people buy Microsoft. As the technology giant creates a giant platform niche (and even scale) vendors build targeted solutions on the platform – which is why it’s as good for banks as it is for hospitals. Giants feed themselves on R&D and Redmond leads the biggest R&D budget in the world which means the platform that Ascribe work upon is always the best. The scale makes it cheap – particularly if you invest in Microsoft across your enterprise and then sweat BI out of the asset with marginal cost. You can also use a range of resources to help, whether its software vendors with Microsoft powered software or consultancies who configure BI solutions or contractors or your own staff. Finally there is the architecture. The software is designed to align with industry standard methodologies such as Agile, so you can build solutions quickly, and Kimball so you can have a concrete data management strategy but a rubber implementation plan. Thanks Simon M for the concrete and rubber….

The other big play is cloud – I’ll post on that later. All in all then it’s easy to see why I bought into the platform, as the foundation to my business. It should have clear benefits for you too.

I am quite excited about the launch of SQL2012 and in particular PowerView, or Crescent as some of you may know it as. I am pleased that Microsoft are sharpening their In-Memory BI story and they have a drag and drop user interface that can compete with the likes of Qlik-View et al. Blimey, this has started off like a techy post – didn’t mean to. I’ll write more about our use of PowerView on a really interesting project, next time. Let me get to the point.

Microsoft now has Excel, ProClarity, PerformancePoint, PowerView, PowerPivot, Reporting Services, Visio and BingMaps interfacing with its dimensional model (Analysis Services) and now its BISM (BI Semantic Model) which seems to have replaced the Report Model. I am confused and so are my customers. This is also an issue that Gartner picked up on when they did the last magic quadrant review. In fact I remember being at a presentation on SQL 2012 (Denali as was) last year and a poor guy from Microsoft was mullered by the audience of technical guys who berated him for the lack of coherence in Microsoft’s BI message.

I wasn’t that worried actually because, as a partner, it’s my job to take the platform Microsoft gives me and manipulate it to meet my customers’ needs and vice versa – in fact, probably more vice versa.

In my mind I have this sorted out. This is what I do.

Firstly, I talk about the health and social care BI portal as a gateway to all the knowledge assets the organisation holds and my customers shout out things like EDRM / Collaboration / Search / BI / Unstructured Content / nice-looking web-site. We don’t really talk SharePoint. I don’t talk about the different platforms and their naming conventions. For example, trying to explain the evolution of Performance Point only distracts from the need it serves. The need it serves is to provide people who live in a one –five mouse-click world to go from a macro to micro view of organisational performance using a scorecard / dashboard. I think about Public Health Maps, organisational strategy maps and caseload reports (Reporting Services) in the same way – how many clicks does it take to get the information need and how can I, as an end-user be best connected with my data.

I would then think about Excel meeting the needs of analysts by providing direct access to data and I would tell the story of in-memory BI using PowerPivot.

Then I have to think about PowerView. That’s okay – in my first sentence I articulated the value to people who sit between Excel Pivot-table Gods and people who consume data via dashboards. So individually I can map each sort of user profile to a solution and to an underlying Microsoft technology. The problem comes when you step back and think about this strategically. I don’t mean as a programme of work because things like the UI are very similar and so the training overhead isn’t a problem. I think more about the coherence and I go back to that very hot room and the hot talk that made my mate at Microsoft sweat.

I don’t think that has been figured out. Maybe in the next iteration of SharePoint all the BI will be brought together and made into a seamless application so the alignment of function to “user need” doesn’t jar but emphasises the richness of the platform. Let’s see. Microsoft friends if you are reading, what do you think?

For now, I’ll keep on telling my tale – looking into the eyes of each of the different users that I pitch to and pointing out which application is exactly for them and emphasising how we, at Ascribe, understand that this can appear confusing but actually isn’t. So does it matter that when we step back it looks a little messy, when we are actually meeting the needs of our people. I don’t think it does, yet, but I think it will as the BI becomes more embedded.

Because that is the point of BI – to a large extent. You want people to come together to look at information and make sense of it and use it – we may be victims of our own success if we solve the “one version of the truth” issue (so they are all looking at the same data) but we create confusion through the range of tools we offer.

This one will run and run.