THE SQL Server Blog Spot on the Web

Welcome to - The SQL Server blog spot on the web Sign in | |
in Search

Jamie Thomson

This is the blog of Jamie Thomson, a data mangler in London working for Dunnhumby

Geoflow? Is that it, Microsoft?

I didn’t travel to the SQL Pass Business Analytics conference this week but I keenly followed what was going on via the #passbac hashtag on Twitter. Seemingly the big announcement was Geoflow Preview for Excel 2013, an add-in for Excel that visualises data over space and time (read more at Public preview of project codename “GeoFlow” for Excel delivers 3D data visualization and storytelling). Geoflow certainly looks compelling at first glance though I must say I found it rather strange that it got top billing given that Microsoft were talking about it at some SharePoint conference five months ago but nonetheless the keynote demo was apparently very impressive indeed. Here’s a screenshot of Geoflow:

I think Geoflow looks great, I really do; the questions I immediately had about it were:

  • Can I share my Geoflow’d Excel workbooks and have people view them on SharePoint? Answer: No, there’s no SharePoint collaboration story.
  • Is Geoflow part of Power View? Answer: No, its a separate installation.
  • OK so I have to install it. I presume then that its available in the Office App Store given that’s the new model for distributing Excel add-ins? Answer: No, you have to download it from Microsoft’s download site.

So the big reveal from Microsoft at this conference was an Excel add-in that does something very very cool but had already been announced, is only a preview, doesn’t fit with Microsoft’s BI collaboration strategy, doesn’t use their modern distribution platform and isn’t part of their Excel-based data visualisation tool. Well thank god I didn’t stump up the cost of travel, accommodation, loss of income and time away from the family for that! Doubtless there were a lot of other good reasons to go to the conference but I would have been going with high expectations of news from Microsoft that is going to be compelling and help me sell Microsoft’s BI offering to my clients – Geoflow doesn’t do that, not by a long chalk.

What I was hoping for, nay expecting, was a concrete announcement regarding Microsoft’s mobile BI strategy. We first saw Power View demonstrated on an iPad at the PASS 2011 conference and I assumed that in the intervening eighteen months they might have built something we could actually install and play around with. Apparently not.

Microsoft are getting killed in this area. At my current client all the management folk walk around with iPads glued to their hands – this is the tool on which they consume information and Microsoft doesn’t have anything for them. I was working for a client two years ago that had just invested in a product called RoamBI  because it enabled them to view Reporting Services reports on an iPad. Two years ago for pity’s sake, and Microsoft haven’t released anything mobile-BI-related since!


Published Sunday, April 14, 2013 5:42 PM by jamiet

Comment Notification

If you would like to receive an email when updates are made to this post, please register here

Subscribe to this post's comments using RSS



Allan Mitchell said:

Agree wholeheartedly JT.  Given Sharepoint is the BI collaboration tool I would expect Geoflow to be integrated.

I am also not a fan of separate tools for everything.  PV + Geoflow would be good/Better.

Other people are doing Mobile BI (Tableau for example) and not doing badly either.  MSFT at the moment are talking the talk but not walking the walk.  They are already playing catch-up but if left much longer the gap may be too large for them to make an impact.

April 14, 2013 12:08 PM

Mark Stacey said:

Jamie, I absolutely agree with you on this:

Given the hype, I'd have expected a lot more from MS on spatial visualization. Between pie charts on maps in PowerView, no HTML5 support for PowerView, and the whole "PowerView can't be used with the other SharePoint components, it only really works fullscreen and can't accept parameters" story, I really think MS has lost their way on this.

We (a very strong Microsoft Gold partner in BI) have actually gotten to the point that we're writing our own SharePoint BI webparts using D3.js to get the BI we want. And we're going to put them on the App STore as well :)

April 14, 2013 12:20 PM

Toufiq Abrahams said:

This was exactly my expectation. Two weeks after the Data Explorer release, I did a demo at a SSUG with "teaser trailers" of GeoFlow and machine learning. I mentioned #passbac and boldly voiced that I expected mobile to be announced because I was so sure that PASS could not launch a new conference without an announcement of this scale and impact ... GeoFlow does not measure up. It's a "cool" product, but doesn't help the MS platform anywhere close to the gaping hole called mobile.

April 14, 2013 12:56 PM

Will Riley said:


As you know I agree entirely with your sentiments. In the time it's taken Microsoft to release nothing useful that's mobile BI related (Geoflow looks cool but its practical use is within the BI stack is virtually nil) and at the same time manage to piss off the Excel community by effectively restricting the adoption of PowerPivot/Power View, most of the competition have not only released BYOD-friendly mobile BI apps but the most forward thinking have completely re-written their offerings in HTML5 (QlikView .Next for example)

So still we wait, and at every pre-sales meeting where the "Can our execs use MSFT BI on their iPads?" question comes up (80% of the time now), I have to shift uncomfortably in my seat and say "No, and what's more, even as a MSFT partner, we have no clue when we will be able to say Yes!"


April 14, 2013 1:12 PM

Derek Goodridge said:

I agree as well.  Twice over the past week I've been asked about best BI tools for iPad, also seeing lots of interest in Tableau, and their new version 8 is very compelling. Getting increasingly difficult to mount an argument in Microsoft's defence.

April 14, 2013 1:24 PM

mick horne said:

And so having spent 2 years pushing the SharePoint message, despite huge reservations from the partner community, here we have a release of a front end tool that isn't supported through SharePoint.

You want to put the fun back into BI? It might be fun producing flashy but ultimately pointless demos for serial conference attendees but it certainly isn't fun trying to position a Microsoft BI strategy to enterprise customers.

And how many different but overlapping tools do we need? I now need two monitors side by side to view my Excel ribbon.


April 14, 2013 2:39 PM

Frank Szendzielarz said:

I don't know. I think one sign of the sickness pervading the last decade or so is, as you put it, 'management folk walking around with iPads glued to their hands,' and such phenomenon. The so-called communication revolution has gone too far.

Ideally, what I'd like to see is all these management folk glued staring into their ipads wandering around continue to do so and wander straight off the nearest cliff.

Honestly, these days most people are struggling to keep a job, find credit to start a business, or otherwise cope with an ongoing financial crisis. If anyone thinks they need geo-flow on their iPad then they're probably more in need of putting their pennies into a saving account and a long holiday in Tibet or something.

April 14, 2013 2:54 PM

Will Riley said:

Hmmm...I think the point most people contributing here are trying to make is that it would be nice to get ANYTHING related to Microsoft BI on to a mobile device. Those of us trying to sell BI to the enterprise are being severely compromised by a total lack of support, information and any indication of a roadmap and timeline from Microsoft.

On the other angle, I'm not sure that mobile BI and its adoption by end users has got anything to do with the global financial crisis. I just don't see a correlation at all.

April 14, 2013 3:55 PM

TheRomit said:

I have to hope the thing was not in the Office App Store because it is preview tech. No?

April 14, 2013 9:45 PM

Mike said:

Finally someone understands my Microsoft BI pains. I too followed the PASS BAC tweets looking for good news. Fun and easy, ok but BI = Basic Intelligence is terrible. My customers want their competitors to have Basic Intelligence, they ask me to recommend the Best Intelligence.  My customers need and will pay big bucks for strategic BI tools...not shiny toys.  The MVP attendees lost some credibility in my mind since they sounded like they were being paid by Microsoft to hype Excel and cloud versus provide the BI professional, thought leader updates they historically have provided.  The GeoFlow Excel add-in is interesting but it is yet another silo tool that does not really fit in anywhere except for rare, niche use cases. I see Data Explorer as a response to Qlik, Oracle Endeca, and other data discovery tools but why Geoflow?  Who asked for it?  What about the millions of customer requests for Mobile BI?  And for cloud, come on guys we really need solid BI on-premise for a long while. Many of the customers are just upgrading now to Microsoft Office 2010 and SQL Server 2008R2 or 2012.  At this point Microsoft BI has also lost credibility after years of lagging and bumbling around.  Microsoft should just buy a BI company that has a mature, modern, mobile capable, front-end BI solution.

On a related note, am I the only one noticing that there have been little-to-no real enhancements to classic Reporting Services, PerformancePoint, or Data Mining in the past few releases?   Power View was the exception but Microsoft chose Silverlight (a dead technology) and Silverlight is still in Excel 2013. The new 2013 licensing is keeping PowerPivot and Power View out of reach. Report Models were quietly deprecated in 2012 without a migration path. PerformancePoint death is next. I can't recommend PerformancePoint in good faith to any of my customers knowing that the engineers for it were laid off a long time ago. It seems like Microsoft has been planning to leave the BI market since they forced the SharePoint requirement. They chose not to build APIs for PowerPivot or Power View, forced many software vendors to look for other vendors embedded BI solutions, chose to rely heavily on partners fill voids, prolong the ongoing silence about BI product futures, and all this while shoving select BI components into Excel to focus entirely on Microsoft Excel and cloud futures. Most customers call for help wanting to get away from reporting in Excel. As soon as I open Excel 2013 to show them the cool Power View demos, I often hear "what else do you have that we can use"?  Microsoft BI pales in the face of the BI competition. Microsoft may have become experts at wooing the analysts and MVPs. They clearly are not listening to their customers. Look I know this all sounds bad but I am not a hater. I miss the Microsoft BI of the past. Who can help Microsoft see the light before it is too late?

April 14, 2013 10:51 PM

Josh Ash said:

For what it's worth I'd like to chip in that I agree completely with what you guys are saying.

What Mike said - "Microsoft should just buy a BI company that has a mature, modern, mobile capable, front-end BI solution."

I think this is a brilliant suggestion. Tableau or something similar would be a good acquisition for Microsoft as it seems, at least up until this point, that Microsoft are not capable of producing a coherent BI solution that includes mobile delivery.

"The MVP attendees lost some credibility in my mind since they sounded like they were being paid by Microsoft to hype Excel and cloud versus provide the BI professional, thought leader updates they historically have provided." - This too is an accurate observation. It seems the main benchmark to become an MVP now is touting Microsoft's products on your blog in a positive way. There are some great MVPs out there still, but it has changed.

"Microsoft may have become experts at wooing the analysts and MVPs. They clearly are not listening to their customers. Look I know this all sounds bad but I am not a hater. I miss the Microsoft BI of the past. Who can help Microsoft see the light before it is too late?" - HEAR HEAR!!

April 15, 2013 12:36 AM

T said:

Unfortunately any company MS buys would end up dying once absorbed, its poeple leaving to built another startup and its technology refactored to fit Sharepoint and Excel molds, so it would just be a waste of a valuable alternative for customers... Yes I too miss the MS of the past. Very Nostalgic.

April 15, 2013 12:55 AM

jamiet said:

Hi Josh/Mike,

"It seems the main benchmark to become an MVP now is touting Microsoft's products on your blog in a positive way.

For what its worth, I am currently an MVP (perhaps not for much longer after writing this blog post :) ). "



April 15, 2013 1:24 AM

Josh Ash said:

Hi Jamie

It is true, you *have* sinned.

Say ten "Hail Ballmer's" , and confess to your MVP lead!


April 15, 2013 1:34 AM

Rob Farley said:

I see lots of MVPs being critical about Microsoft technologies. I think if someone is ONLY critical or ONLY flattering, then that's more likely to hurt a potential MVP award. A mix is just fine, as it helps everyone to realise that when praise is given, it has probably been well-considered.

April 15, 2013 2:29 AM

Charlie Maitland said:

Ah, the eternal hunt for a BI Strategy from Microsoft. This total fragmentation is nothing new. Just think KPIs. People are suggesting the Microsoft should buy a BI vendor, well remember Proclarity. That worked out well!

At the end of the day the reality is that there are 4 competing teams all pushing to get budget and attention for their pet projects and therefore their promotions. In the Red Corner you have the Office team with Excel and SharePoint. In the Blue Corner you have SQL Server and Analysis Services. All of these teams have very different views on what BI is and how it should be delivered. Sometimes personalities can cross the divide but for the most part  it is every man for himself.

If you are waiting for a coherent, joined up, and user centric BI Strategy from Microsoft you may want to check your pension plan first.

April 15, 2013 2:34 AM

Mike said:

Mark Stacey, Rob Collie,and Chris Webb have all called out Microsoft at least once in public for the good of the many. If I went back to the PASS event when SSAS Tabular was introduced, there were a lot of ugly gripes. They lived through it for the better even if some hands were slapped.    

The MVP community does need a few voices of reason to lead us safely into the future. MVPs should be able to call out issues and help Microsoft make course corrections. As an implementation partner, our group has been begging for answers for a long time. I do know the "uncomfortable seat shifting" the other partner mentioned and it is costing us deals. Thanks for being strong enough to point out issues allowing us a chance to beware of them, discuss them, and learn from each other.

April 15, 2013 2:38 AM

Josh Ash said:

It's true there are a lot of great MVP's out there. And it is understandable that Microsoft looks to its MVP program to promote its products. After all, they are not a charity.

I know there are different teams vying for budget and attention etc but surely there would be an overarching strategy so these teams can work together rather than endlessly compete ?

April 15, 2013 4:23 AM

Chris Webb said:

Welcome to your "SSAS is dead" moment, Jamie!

I agree with pretty much everything that's been said here about GeoFlow, and also about the buy/build question. Despite all the valid criticisms of GeoFlow here though, I do like it and it does demo extremely well - as anyone who saw Amir's keynote on Thursday will testify.

Regarding the criticism of MVPs, I don't think it's completely fair to say that we've been co-opted by MS into hyping their products or that the benchmark for becoming an MVP is to do this. There's a constant stream of public and private feedback that goes back and forth between MS and the MVP community, and MS does listen even if it can't always take immediate action, or disagrees with what it hears. I've made some very public criticisms of MS BI strategy in the past and I'll do so in the future, but one thing I have learnt over the years is that the result of *constant* public complaining is that your complaints end up getting ignored.

I don't know exactly what's going on with Mobile BI and why it's so late but I'm 100% sure that the relevant team at MS don't need us to tell them that they are in deep trouble. It seems that they do need to be reminded of the need for communication with partners and customers though, even if the message is bad, and that's why I'm glad Jamie wrote this post.

April 15, 2013 5:11 AM

Calvin Ferns said:

Another Partner here agreeing that it does not help us sell a coherent enterprise BI strategy.  This is what we need to see from MS far more than yet another tool.

April 15, 2013 5:13 AM

Will Riley said:


I agree with your assessment of MVP independence, and I already advised Jamie to get a thick book to stick a book down his trousers at SQLBITS :-)

Microsoft rely heavily on both the community and their partner ecosystem. I'm convinced that their current struggles to hit sales targets are (in part at least) directly linked to their lack of communication with partners and the community in general around Mobile BI.

April 15, 2013 5:19 AM

Jen Stirrup said:

I tend to think that Microsoft can't win.

Excel is the way forward for Microsoft Business Intelligence: hence, PowerView, Data Explorer and Geoflow are both part of Excel. If it had gone into SharePoint first, there would be a furore about the fact that Geoflow had gone into SharePoint Enterprise and wasn't accessible to people without SharePoint. Remember the Power View furore?

The truth is that the majority of BI work is done in Excel, and I think Microsoft are just bringing everything home to where the Business Users do most of their work. It's about working with the Excel users, and giving people opportunities to work with other technologies that are outside of SharePoint or Office365.

Regarding mobile, my ideal strategy is for people to use Microsoft and Tableau to service a BI solution. Excel will keep most people happy, and Tableau is the 'gold standard' for mobile data visualisation and Microsoft are only one of the many vendors are trying to keep up with Tableau.

For mobile business intelligence strategy, my bet is that you should all be watching Office365, which I haven't seen mentioned here in the commentary so far. Excel is currently available on a mobile device via Office365 or even SkyDrive, for example. It would be a nice evolution for Office365 if GeoFlow, Power View etc were available via a mobile device, since you could access it anywhere.

We Business Intelligence people do not like uncertainty since we are fundamentally 'data' people and I'd like to see clarity in this area, for sure.

April 15, 2013 5:33 AM

Preston said:

Regarding the comments of MVP independence.

Noone would deny the great work that MVP's do. Hell, I've got by on their advice, their code and their blogs for years!

But it is hard to claim they are 100% objectively independent. The award is handed out by MSFT, not an independent company.

It would be very difficult to compose a convincing argument that a person is independent regarding their opinions of a company, when they are receiving an award, and benefits, from that company.

In the greater world outside of technology, that is called a 'conflict of interest'. Nothing wrong with receiving benefits of course, but it does make it hard to claim independence.

April 15, 2013 6:11 AM

jamiet said:

Hi Jen,

You said

"Excel is the way forward for Microsoft Business Intelligence: hence, PowerView, Data Explorer and Geoflow are both part of Excel. If it had gone into SharePoint first, there would be a furore about the fact that Geoflow had gone into SharePoint Enterprise and wasn't accessible to people without SharePoint. Remember the Power View furore?"

I don't understand the notion that it has to go to just one place first, before everywhere else. It should go *everywhere* first. Make Power View *viewable everywhere* (i.e. Escel, SharePoint, SkyDrive) and then put new features (e.g. Geoflow) into Power View, ensuring that its *still* viewable everywhere.

The piecemeal approach Microsoft takes to this really annoys me. Fine, if they want to have premium SKUs then make it so that you can only *build* Power View reports in those premium SKUs, but the reports should be *viewable* everywhere. And that means on SharePoint, in Excel, on SkyDrive and also on iOS/Android/Win8 if/when those apps reach us. I'm fed up with conversations that go along the lines of "well, you can use functionality X, but only if you pay for this edition". The rest of the world doesn't work that way anymore, and customers won't stand for it anymore when the other guys make it a lot simpler.


April 15, 2013 6:35 AM

Jen Stirrup said:

Microsoft suffer from the Scylla and Charybdis problem. They can deliver something quickly to show proof of delivery and to 'test the waters', or they can wait until Geoflow (or any other product) is ready in every and any scenario, and then release it. Either way, they get pricked by Morton's fork since the solution is the same: not everyone is happy.

Another way of putting it is that they are essentially a commercial enterprise and it's the usual business problem of you can have something fast, good, or cheap - but not all three.

Power View was put in SharePoint first, and then added to Excel.  GeoFlow has been put in Excel first (after the Power-View-In-SharePoint-Only problems) and then Microsoft can see what the adoption is like. If nobody uses it, then it will naturally have attrition. But at least we have something to play with, for now.

You can't just buy a technology and expect it to solve all the business problems and 'one size fits all' will never work. Different things for different needs, so hence a hierarchy of payment options - you get what you pay for.

I do think, and have lamented long, that the licensing and functionality relationship is never clear to anybody, and hence comes across as piecemeal and fragmented. This is further compounded by the fact that Business Intelligence, particularly mobile, is simultaneously heterogenous and personal, particularly on mobile devices, and difficult to pin down.

I don't think a 'suck it and see' approach is unreasonable as long as its accompanied by strong messaging - and this is the piece we are missing.

April 15, 2013 7:17 AM

AaronBertrand said:


MVPs are MVPs because they have shared their expertise with the community. It's a reward for the previous year's community contributions. It is not a paid position and we do not receive benefits to be blind Microsoft cheerleaders handing out kool-aid.

I am a SQL Server MVP. 99% of my work is done in Mac OS and 100% of my work is done using Apple hardware (with either VMs or RDP to facilitate my work). I am not shy about lambasting Windows in general, criticizing the terrible things they have done to Windows 8 in specific, or even speaking unfavorably about choices the SQL Server team has made in implementing features, ignoring bugs and throwing most of their money into the cloud. I am very happy to say that I have been this way throughout my career. I have been an MVP since 1997, and have always been very honest with my audience - I cheer when Microsoft does good things, and I boo when they don't. The challenge is to boo in a way that is not seen as a direct attack, and to share feedback internally and give them a chance to respond before blasting them in public (you don't have to be an MVP to show them this kind of respect, either). So in that case I agree that we're not 100% objectively independent, but please don't paint us out as a bunch of zombie lemmings who say nothing but good things about Microsoft. That is quite far from the truth.

April 15, 2013 8:53 AM

David Hager said:

The bottom line is this: Microsoft HAS created great self-service BI tools (PowerPivot, PowerView, Data Explorer, GeoFlow). They are the BI leader by far in this category. However, if these tools are not available to everyone on a level playing field, and don't work smoothly across platforms and in the cloud, then Microsoft's BI bet will fail miserably. I have a gut feeling, though, that Microsoft cannot possibly be as inept as they appear to be, and the availability/integration/interoperability issues will be solved very soon.

David Hager

Former Microsoft Excel MVP

April 15, 2013 12:20 PM

Will Riley said:


I'm sincerely hoping you're right. This isn't just some random dig at Microsoft we are seeing here. It's an impassioned critique of a glaring hole in BI strategy that needs to be filled by those that are in large part either "friends of Microsoft" and/or reliant in th getting this right for their livelihoods.

No news, in this instance, is not good news!

April 15, 2013 1:40 PM

Frank Szendzielarz said:

I also see BI going the Excel way. I am also very curious about the role Silverlight plays in this, and I also agree with one or two above that the space to watch is Office365.

The notion of the datawarehouse is anachronistic - there's more horsepower in a laptop Excel installation than there is in a multimillion dollar old corporate datawarehouse. As for front end interrogation of the data, the real question is XAML or HTML5 - and this leads on to questions about Windows8 and Office365.

I think mobile touchscreen is a fad about to die a swift death in the face of technologies like Google Glass and the like.

April 15, 2013 3:29 PM

Frank Szendzielarz said:

@Will Riley

No, the problem is that you are trying to sell BI into the enterprise. a) what does the enterprise need BI for? large companies have all the info they need, and their businesses run on a combination of cultural inertia and international credit flows. b) Small companies need BI - and this means Office 365 - and that's where MS is at. Enterprise is history.

April 15, 2013 3:33 PM

Peter Thomas said:

Bloody hell.  The future is Excel?  What other news exactly have I missed?  I have been helping customers get away from Excel messes.  Has there been deprecation news for SSRS and PerformancePoint?  Is it true about Report Models?  Where do I get the not so warm, fuzzy news to know what to implement and avoid?

April 15, 2013 9:23 PM

Joe Harris said:

@Frank Szendzielarz

I got a good laugh from "The notion of the datawarehouse is anachronistic" and then I saw "I think mobile touchscreen is a fad about to die a swift death". You have taken trolling to a new level.

Re the touchscreen thing - Google Glass is, at best, the Windows CE v1 of this type of interface. Read Halting State or Rainbows End for an idea of what this needs to look like for real use.

Regarding data warehouses, I'm not sure what kind of DW you've been using but I'm the guy that gets brought in to rescue people from that special kind of "We don't need a DW" Excel hell.  Sure Excel has a million rows these days and that's great. It's still slower than treacle to try and do anything of merit in Excel and a million rows is getting to be a rounding error. Yes, even for "small businesses".

April 16, 2013 10:31 AM

Jon Capezzuto said:

Geoflow should be a feature in the next Power View release, not a separate product.  Then it would be supported immediately in SharePoint and Excel, and soon in HTML5.  Having another Excel based mapping visualization is just confusing and unnecessarily overlapping.

While we're on the topic it also seems Data Explorer should be part of PowerPivot. Keeping both of these as separate products simply complicates the development efforts and increases the opportunities for inconsistent functionality going forward.

April 16, 2013 3:08 PM

Josh Ash said:

Hey has anyone been using Geoflow on corporate data.. I am wondering what kind of performance you are getting..

I don't have a dedicated graphics card at work and was wondering if anyone out there is getting much better performance on machines with dedicated graphics cards?

How many data points are you rendering and what kind of hardware are you using?

I'm getting pretty good performance but was wondering how far GeoFlow can be pushed on high end machines before it starts to slow down..

April 16, 2013 11:16 PM

Frank Szendzielarz said:

@Joe Harris - well I think being a little provocative and mischievous could help boost JamieT's blog readership for one thing :-)

On the other hand, no I am completely serious and I think you are wrong. I think the future of BI is MS's Self Service BI and small or medium sized businesses don't need 'data warehouses' and sophisticated ETL as much as they did before, which are in essence a query performance optimisation strategy and nothing more.   Also another angle is the 'information flow' concept which Jamie mentioned in an earlier blog post, which is basically event driven ETL in my opinion or in other words tying up reporting with business process automation.

As for mobile touchscreen - I stand by my point: there is no way I'm giving up my PC or laptop for a tablet, ever.  There will come other UIs, such as Google Glass, or other gesture based or Kinect-type device based uis that might supersede the mouse and keyboard, but otherwise, no. If someone wants to do geoflow type presentations, they should sit down at a PC or throw them up on a projector.

April 17, 2013 3:36 AM

Joe Harris said:

@Frank Szendzielarz

Re touchscreen: 'Serious' computer users felt the mouse was slow and pointless. Serious computer users still use the mouse as little as possible. It's not about replacement, it's about capturing non-consumption either in new users or new usage situations. Touch devices now outnumber PCs so I'm thinking there might be something in it.

Re DW: A DW is the single place where a business is forced to classify and integrate their data across operational silos. That involves making hard decisions and talking to other departments. The modelling method or technology used is irrelevant it's the bringing together that matters. SSAS (or OLAP generally) is a performance optimisation but the DW is actually for business rules **clarification**.

Re ETL: I love the idea of event driven ETL but in my experience it tends to create a hard dependency between reporting systems and front line systems that slows down change in both sides. Working at one remove is certainly a bit slower in terms of data delivery but it adds a lot flexibility that you miss when it's gone.

A wise man once said "efficiency is the enemy of robustness". A DW and ETL process look like inefficient duplication of data and effort on the surface. But look deeper, the goal is to build a robust and reliable source of cross process data. We didn't arrive at the current approach by some sort of historical accident. Almost every imaginable alternative has been tried by someone and this approach can succeed in average businesses with average developers better than any other.

April 17, 2013 5:23 AM

@Joe Harris said:

Re: DW - well no it's not is it? The SOA-style service bus does something similar, and now we have sharepoint and Excel services and PowerPivot models, while the whole notion of having a DW as a single place to organise views into data has been a failure since the beginning. The DW vs Datamarts debate still goes on, and only serves to illustrate the problem.

The so-called operational silos are only operational silos because of reluctance to do things like create a whack-load of indexes on a LOB transactional db, but with newer technologies, architectures and approaches, that problem starts to go away.

I can see quite clearly new small business users simply connecting their Excel workbook to their CRM systems (eg Office365) and producing all the reports and animations they will ever need. If they want to merge in data from some other source, they can do it all using Sharepoint and Excel. Where's the need for a DW and an expensive consultant? Sorry - no business case - maybe for older companies and legacy enterprises such as banks.

April 17, 2013 7:43 AM

Mike said:

RE: Not needing a DW, users make all the reports they will ever need

Respectfully, completely disagree unless the business is doing trivial reporting!  There are many great case studies on how poor "micro-level" reports/decisions are made that impact entire businesses ie. bank that denies micro-unit business loan for what happens to be a very profitable mega-customer across the whole portfolio of businesses.

In this data explosion era, there is a place for self-service BI and DWs and a solid BI architecture. There is more of a need than ever before for "usable/consumable" prepared data sources like the DWs, hybrid-DMs, these business users can efficiently consume to create their own reports with their own BYO reporting tool. Most of the time you can't just point users at the raw ERP and CRM transactional DBs and say go at it. For instance, business users wouldn't have a clue on how to compare Year over Year performance correctly when Business structures change - DWs use Slowly Changing Dimension patterns to accurately report those scenarios.  There are tons of other examples. Check out this recent article from Claudia Imhoff.  

Do We Still Need a BI Architecture? You Bet We Do!

Claudia Imhoff

April 17, 2013 2:09 PM

Josh Ash said:

"I can see quite clearly new small business users simply connecting their Excel workbook to their CRM systems (eg Office365) and producing all the reports and animations they will ever need"

I disagree with that too.. It's never just been about performance - the complexity of these systems doesn't simply allow you to point Excel at base tables.

The advent of new hardware, especially RAM, has made queries run faster though - it's probably fair to say that in a lot of scenarios we no longer need data warehouses for their performance benefits. And you can get away a lot more now with relational reporting.

But it doesn't solve the problems Mike has given examples of. Data Warehousing is still solving those problems.

And that's before we've even started discussing those trends that have made reporting HARDER. One example being the storing of data in SharePoint lists, and in XML format in InfoPath forms that are user designed. Try giving your business users a connection to a bunch of XML files created by other business users in InfoPath that contain critical data - and see if they can get meaningful results out of it. Don't get me started on reporting on InfoPath data...

April 18, 2013 12:48 AM

Frank Szendzielarz said:

@Mike / Josh

RE preparation of data/slowly changing dimension/etc - I think this is a cultural problem. So far, the business knowledge and understanding was put into the BI team because these challenges of organising information were/are considered technological problems.

The fundamental shift is not a technical one but a cultural one that is made possible by new technologies. While data preparation and organisation of data structures that facilitate reporting (or 'BI' / 'Datawarehousing' as you are now putting it), will continue, the people (you) doing it will no longer be considered technologists. You, or your successors, will be working in operations teams, in analytical teams, superusers who will be responsible for using the existing platforms for creating and maintaining these structures within the Office context.  

So if you are saying that Data Warehousing is really no longer so much about the technology, but about the process of understanding the information and making it available, then yes Data Warehousing will continue, but the people doing it will be members of the business departments, using Office, Office365 and plugins in the great majority of cases I think.

April 18, 2013 3:33 AM

Mike said:

@Frank  Alrighty then, you have had far too much Microsoft Office 365 Kool-Aid... your signature should read "now a word from our sponsor, Microsoft Office".  There will be less IT infrastructure roles and DBAs in cloud world but DW/BI development will thrive.  The DW location simply moves to cloud but the design patterns and complexities of querying and reporting across multiple systems remains the same as it is today.

Let's hear what the those of us that live and work in the real world have to say about that.  

April 18, 2013 11:54 AM

Josh Ash said:

An Office plugin that does data warehousing and can be operated by business users.. cool

Does anyone have the URL for it?


April 18, 2013 9:19 PM

Frank Szendzielarz said:

Perhaps the best place to continue such discussion would be here:

April 19, 2013 3:37 AM

jamiet said:

Jon Capezzuto said:

"Geoflow should be a feature in the next Power View release, not a separate product. While we're on the topic it also seems Data Explorer should be part of PowerPivot."

Even better, what if all four (PP, PV, DE, Geoflow) were all just one product, with one installation?

April 22, 2013 4:05 AM

Jean-Pierre Riehl said:

I am a SQL Server MVP and I'm not happy to read these comments.

We are all passionate by Microsoft Technologies and we share that passion.

That's why we are MVP.

But remember we are also all professionals, consultants, developers, managers...

As you, we are embarassed with some customers questions and more than you we are embarassed to tell them that Microsoft is not clear on some points.

Please consider all of our work before critize them.

Also, I'm totally agree with Jen. You cannot make everybody happy. Nothing is perfect and even is Microsoft.

2 years ago, we all complained about SharePoint. Now you are whinning about Excel. What's next ?

I share the pointed problem about a unique BI story accross all devices and bound in one tool. And I tell it to Microsoft. But I'm not Microsoft, don't blame me.

I also think it take a long, a very very long time, to preview some Mobile BI features...

I like Geoflow and many of my customers said "wow" when I presented it.

I guess it will be integrated in SharePoint, Power View, Office Apps etc. I just tell it to the right person and not throw the baby out with the bathwater.

PS : i'm pleased to see that it is not a French guy who starts complaining :p.

April 28, 2013 9:31 AM

jamiet said:

Hello Jean-Pierre,

Thank you for your comment.

"Please consider all of our work before critize them."
Not quite sure what you're saying here. I will assume from what you said in the rest of your comment that "our" refers to all MVPs. Are you saying that by criticising Microsoft I am reflecting negatively upon the work that other MVPs do? If so, can you explain how?

"But I'm not Microsoft, don't blame me."
Don't blame you for what? I wasn't aware that I was blaming you for anything.

"i'm pleased to see that it is not a French guy who starts complaining"
Why is a person's nationality significant when that person is voicing a personal opinion?

Can you please clarify something for me. Are you suggesting that due to my receiving of the MVP Award that I should not voice any of my opinions that are critical of Microsoft?


April 28, 2013 10:12 AM

AlejandroL said:

I have been following this conversation with the most interest. I left BI consulting almost a year ago, and have moved into corporate BI Management. And if MS strategy looked not so good from the partner perspective, from a large organization point of view, is simply crumbling.

I used to be a SQL Server MVP. I know for sure that (at least SQL Server ones) are very vocal and critical to Microsoft.

However, Microsoft execution is poor, bad designed, and completely utterly ignorant of real market needs. We have had the conversation with Microsoft about BI consistency.

And for the last 5 years, it has been a non-stop SNAFU after another: Proclarity ,

PerformancePoint Planning,

Data Mining forgotten,

no serious management tools for troubleshooting SSAS Cubes -Only what good Mosha left behind-,

poor Reporting Services integration with SSAS,

Zero Microsoft BI Mobile (and the paper on using BI on an Ipad is just for fools, totally humiliating),

PowerView STILL (after 2 years) is not able to access SSAS Multidimensional (we only got a CTP, an old one by current IT standards), and now,

Excel licensing isolating BI from the mainstream.  

Just one question: WHAT IN THE WORLD IS MICROSOFT THINKING?... I seriously respect Amir Netz, but man, you gotta recognize he has only done good demos for the past 2 years... And none of those promises/vision have made it to the hands of us, the consumers, nor with the bare basic quality expected from a company like Microsoft.

April 28, 2013 10:52 AM

Eric Lawson said:

Wow Jamie. This sure has kicked off considerable and passionate debate.

Notwithstanding Jamie's excellent analysis of GeoFlow, for those bemoaning the lack of a good mobile BI offering from MS, it would be good to keep in this in mind.

It is a big ask for any organisation like Microsoft to deliver upon every niche they operate in as well as the core competencies (and I think most of you would agree that in the database arena they are doing very well). You could argue mobile BI is not niche, but not so long ago it was.

Anyway,  as the wheel of the IT industry inexorably turns through the same cycles (only the themes are changing), we continue to have the same choices to make.

So on the BI theme, the choices are

A. Get in bed with with a huge, apparently complete solution / partner (eg. Informatica, Business Objects) and then find out (often too late) some of the components are square when you expected round.

B. utilise the core capabilities provided by your chosen hardware vendor and custom build your solutions.

C. grapple with hybrid's of the two approaches as your organisation moves through the cycle

D. if you reach sufficient age, REPEAT.

The choice between A & B is that age old conundrum of “will the retrofit of the 20% of the solution in option A be better or worse than the ‘do it yourself’ method of option B”?

I have always tended to lean towards custom build, largely because of distrust about silver bullet solutions, offered by teflon coated sales people, that all too often, disappoint.

However, if you are a small IT department, big IT projects can be the undoing of what was a good, capable team.

The good news is, we work in a rewarding and challenging industry and get paid well to confront these challenges. The Microsoft community in particular is very open, productive and supportive and I personally benefit from this support every week ( from contributors who are MVP’s and not). That support depends upon the freedom to express opinions in a respectful way. Jamie certainly ticks that box.

As soon as constraints are put upon that freedom (implied or directed) then the community support will suffer.

The freedom to speak frankly (and also the desire) within large IT projects is crucial as issues cant be addressed if they are not acknowledged and addressed. You cant behave one way in good project teams and another just because you have been endowed with an MVP award. The programme will suffer if the good, straighforward professionals refuse the accolade if they feel they will be censored.

So please keep the Microsoft Professional support community horsemeat (and BS) free.

April 28, 2013 11:22 AM

David said:

Tablets like surface pro can handle easly geoflow

April 29, 2013 3:32 AM

Jean-Pierre Riehl said:


I answered to some comments, not to your article. You started a debate and it is your/our job to do that. Reading comments, I feel like a Microsoft mercenary because I/We are an MVP.

"Please consider all of our work before critize them."

--> I meant a MVP is doing more than writing articles in favor of Microsoft. We all works for CUSTOMERS, not Microsoft

"But I'm not Microsoft, don't blame me."

--> I meant that if Microsoft do something wrong, as an MVP I am not responsible of that, don't blame MVPs

"i'm pleased to see that it is not a French guy who starts complaining"

It was a joke about French people complaining about everything, forget it. My intention was to unpassionate my comment.

Can you please clarify something for me. Are you suggesting that due to my receiving of the MVP Award that I should not voice any of my opinions that are critical of Microsoft?

--> surely not, that debate is important. I share that Mobile solutions are missing from MSBI, I share that 1 tool is better than 4.

My point for people in comments was :

  the world is not divided into 2 categories : MVP blessings Microsoft and Others criticizing it.

I've just defended MVPs. Don't misunderstand my comment Jamie.

April 29, 2013 6:57 AM

jamiet said:

Hi Jean-Pierre,

Thanks for coming back to comment. I admit that when I saw your first comment I was rather put out as I thought you were basically stating that you thought I shouldn't have written the article. I now realise that that is not the case and I apologise for any misunderstanding on my part.

I certainly agree with your assertion that "the world is not divided into 2 categories : MVP blessings Microsoft and Others criticizing it" and understand now that you were merely defending what you saw in the comments as unfair criticism of MVPs - I definitely agree with you here.

Thanks again for taking the time to comment. I'm glad we cleared up any possible misunderstanding.



April 29, 2013 7:09 AM

Jordan Martz said:

MVP and SQL people,

I understand the comments and the awareness with regard to the answer to that solution.  After getting heavy into the mix on MSBI pre-sales and development, I started looking for a better way than what we have today.   I've been heads down looking at how to fix that problem on an SSIS and SSAS (Multi-dem and Tabular) generation framework, as well as, a super powerful front end, called Prism.  I'm the Chief Product Evangelist @ HaloBI.  We're got 400+ installations of SQL Server that sold to create an an enterprise solution for the MSBI stack.  Please take a look at, and I'll demo the product for you, if you email me at  

This product is the real-deal.  I've got answers for every question that has been approached in this conversation, and we're fighting answers from Oracle, IBM, SAP, and Qlikview every day.

I'm sorry to shamelessly plug in the mix here, but we're doing something to address this direct problem and gap in the market and would welcome your feedback.  We'll invest the product and make it happen, so real business analytical process and problems can get moving.


May 3, 2013 1:34 PM

Leave a Comment


This Blog


Privacy Statement