JS
Hello,
I just came across ARIS Mashzone around two weeks ago. So, I was curious and checked out, what you now are putting into the market. Having checked the product just superficially so far, I admit that it interests me quite a bit. So, in the next days, I guess I will start to examine it a bit more in depths. But so far I would like to make the following annotations and arising from these I also have some questions:
1. The product - in a certain way - partly offers capabilities like a true datawarehouse solution e.g. data sourcing capabilities usually known in the industry as ETL (Extraction - Transformation - Loading). At least your datafeeds look like this. It was a kind of "Wow" - effect when I found out that your product even enables the user to combine data records from different sources containing common fields/columns via a "true" JOIN - and not only a UNION statement. This is indeed very remarkable given the low release level of Mashzone. I even know fairly mature data warehouse solutions which still struggle to offer this capability in an easy (!) way... ;-)
However, I was wondering whether you will in future offer an interface to extract data from MS Access solutions. The reason for this idea is fairly simple: In the last two years I worked more on the business side instead of IT departments and I was fairly surprised to see the amount of MS Access solutions which are still used by business users even in case that there have already existed incredible powerful Data Warehouse solutions since many years in these companies. I was so surprised to see such sophisticated and amazing MS Access solutions (in which much effort has been invested) that I tried to understand the reasons why the business created and maintained them. I finally realized that - despite of all IT efforts to overcome these bypass solutions via central datawarehouses - there will always be such solutions on the business side. And I also started to understand that it sometimes indeed makes sense which is very tough to admit, especially for me being a convinced BI consultant. ;-) However, even the most sophisticated MS access solutions lacked the capability to be distributed in an easy way via the intranet. I therefore could imagine that a kind of MS access connector in Mashzone could stimulate your Mashzone sales. Any plans regarding such an interface for the next releases?
2. The interactivity of Mashzone apps - according to my current understanding - is limited to filtering. Any plans on your side to offer some kind of "Pivot" functionality also known as "Slice" and "Dice" in future relases? I know, this is not so simple due to the necessary underlying data model (Star Schema) you then need to consider. But this leads me to my next, more important question...
3. How does Mashzone manages/keeps/stores the data extracted via data feeds? I guess, there is database file underneath. Or do you just keep the data in a simple file (hopefully not...)? If there is no database underneath, any plans to do this in future?
4. The product surprised me. However, I guess the success clearly depends on a clear positioning in the market.
What do you try to target? I do not assume that you try to compete with powerful datawarehouse solutions such as SAP BI, IBM Cognos, Oracle Hyperion... I also assume that you do not try to create yet another kind of Portal functionality... What you call "Self service BI" for the Business user, that sounds interesting at first glance. "Quick and agile" BI - solutions overcoming these incredible long lasting project times of BI projects (in average between 6 months and 2-3 years depending on scope and scale) could attract frustrated business users.
A platform offering business departments the ability to realize a limited, but quick solution for analytic applications/questions which usually do not get enough awareness and high priorities from IT side (due to the very specific business question which might be not so interesting for other departments)? Is this what you want to offer? If so, fine. There is indeed a need in the market.
Integrating internal and external source data quickly on department level without the need to create a time consuming IT investment proposal and - in case of approval - to wait a couple of months, if not even years for the realization. Stunnning idea...
However, the typical "use cases" need to be clarified. A clearer distinction from other, more powerful solutions need to be made, also for the reason that IT is usually looking with very "critical" eyes on such "submarine"-solutions. Usually whitepapers containing use cases could explain theses scenarios in a good way. Anything in the pipeline?
5. "Transposer" functionality for two dimensional excel tables. I read the discussion about this feature and I understand the request of the user who started the discussion. This is indeed a weak point of Mashzone - at least currently. If a business user needs to manually transform the excel tables into a flat file, this reduces the benefit of Mashzone significantly. Expecting the user to write a VBA skript or a VB macro is also somehow reducing the goal of the application to offer a quick and agile BI end user solution and requires a certain development knowledge on business user level. I of course know and agree that even some of the more sophisticated datawarehouse solutions such as SAP BI also still lack this functionality... but some other tools have this ability... Any plans for the future?
6. Files/Annotations. Any plans to enable to attach documents such as Word/PPT/PDF etc. to a Mash? Any plans to enable a kind of functionality to make comments/annotations to specific figures (cell based of course... ;-) )?
7. Delta mechanism: It sounds a bit nasty/ugly... I know... But are there any possibilities to upload only the changes of a file since the last extraction or does the end user needs to make sure, that the file only contains the changes and not also the old values... ? I know... crazy idea, but a very common scenario...
8. The functionality of the datafeed in terms of data enrichment (e.g. calbulations) looks very promising. Aggregation, concation, calculation, all this is very nice and fine and certainly appreciated. However, in case of many necessary transformations/enrichments, the current way to do this graphically might be a bit time consuming. Maybe a more efficient way to do the mapping respectively to enter the enrichment rules would be good... Any plans?
9. How frequently do you plan to offer new releases? Once in a year? More frequently?
Let me finally say, the product really surprised me in an enjoyable way - at least at first glance. I never expected IDS Scheer to enter the BI market (I am not talking about the PI (Process Intelligence) market). There might be indeed a niche in this huge market to empower the business user who needs a tool to provide sophisticated reporting solutions on departmental level integrating data from different internal and external data sources in an easy and quick way without the help of the IT department. However, there might be a clash with well established solutions on the market... Nevertheless, I wish you Good luck.
Looking forward to hearing your opinion regarding the suggestions/questions made above.
Best regards,
Joachim Schirra
Hi Joachim,
Thank you so much for this very detailed analysis of our all new product. Your feedback is very valuable for us and it contains lots of new ideas that will certainly find their way into ARIS MashZone. I am trying to answer your questions.
Positioning and Target group
These business guys (investing a lot of effort into MS Access based solutions) you mention we had in mind, too. Except the fact we did not took MS Access but Excel into account. We think about it as something that is called “the long tail”. There seems to be a great need for solutions for people that are not in the focus of internal IT. Since internal IT only concentrates on mission critical systems with lots of users involved tons of applications – serving only a few – remain open. But these are still important! ARIS MashZone tries to address them from an analytical point of view.
As we propose self service BI software we want to free users from handling SQL-scripts. But as still requested database connectivity (ODBC/JDBC) remains on our agenda.
Storage & delta mechanisms
ARIS MashZone itself does not store the data you enrich and analyze. Replace L with P and you get extract – transform – present: ETP instead of data warehouse’s ETL. ARIS MashZone stores only metadata inside its embedded data base, i.e. MashZone only knows where to obtain data if requested. Data feed processing happens on the fly.
Although this sounds strange from a performance and data warehouse’s point of view it makes sense considering our target group. We want users to create quick solutions without dealing with questions regarding data ownership, legal aspects of data integration, security and more. If somebody already decided to publish a report or an Excel file on the network MashZone can consume it. But the ownership does never change. If the publisher decides to revoke the data, it is not available to MashZone anymore. (It is the same with Google or any other service within the Cloud.)
In addition the most recent data is always available to MashZone this way. As long as the structure does not change updates will be presented in a jiffy.
Transpose
We will handle two dimensional (cross) tables in the next release as it has been even more requested than database connectivity. But I think the term transpose is not enough as it refers to rotate the table by 90° only. As data feeds are record oriented each column has its distinct semantics. A sales report for instance where each subsidiary takes a row and each month a column is not record oriented – but of course easily converted.
Annotation
As of today there are only hyperlinks you can embed within the Visual Composer (it is where you design MashApps). You may use them to delegate to your companies’ wiki for example. File attachments would be handled in a similar way.
Comments within data feeds (for documentation’s purpose) or annotations within MashApps (“My sales figures had been this way last quarter because of this or that special situation…”) sound interesting. I will definitely keep this in mind.
Regards
Stephan
Hi Stephan,
thanks again for your kind, interesting and informative reply and sorry for the long delay from my side to adress this. Just some annotation to some of your points:
"Transposing" two dimensional Excel files :
You are of course right. The term "transposing" which I used is indeed misleading. Not only a 90 degree transformation is required. It is indeed a functionality which converts a two dimensional structure into a flat file - similar to a big hammer. ;-) I know that this sounds easy but is indeed one of the more complex things to solve since every two dimensional Excel file is unique regarding its structure. Some are very simple, otherones are designed really complex. There might be more simple tables and otherones which consist of even different tables in one and the same worksheet, having even maybe text or graphics between two tables which makes an automated transformation nearly impossible. I thought about that problem and I came to the conclusion that the best functionality inside a future realease to adress this problem is a kind of interactive visual tool which opens the excel worksheet and offers the end user the possibility to highlight the columns to be flattened. In case there is a table which is more complex, but the data of a flattened record consists of the columns of two tables which might be seperated by something (texts, other unimportant information, graphics...) then of course it is necessary to highlight not only one block, but even more blocks separately. I guess this is the most secure way to make sure that - unimportant from the logic and design of a two dimensional matrix - the user gives the correct feedback how the structure is to be read. If then there was also a kind of preview functionality where the Mashzone flat file editor shows the potential result of this transformation to enable the user to evaluate whether this transformation is right, then it would be perfect.
Storage & Delta mechanism:
Reading your comment and also the follow up comment from Tobias I learnt something what I was not so aware about before: Mashzone is a tool which does not store data and is heavily dependent on the availability of the underlying source systems and their (sometimes not existing) capability to provide historical data.
Don´t get me wrong... Having studied Marketing myself during my university studies (but unfortunately not worked in this area ;-) ) and knowing the target group concept all too good as wells as keeping in mind the positioning you & Tobias mentioned, it sounds reasonable not to do this at first glance...
Don´t ask me why, somehow I understand the rationale & the arguments, but my stomach tells me something different. I have the feeling that in the long run, users will request such a storage facility. Coming back to the examples of incredible powerful and impressive Access databases which I could oberve as still being the regular case in large organizations in the business departements despite of all existing enourmous investments in powerful enterprise wide data warehouse systems, I am pretty sure that the same (business) guys who developed all these access solutions will have a problem (i.e. being reluctant) to push the same source data in Mashzone. Just because they cannot provide source queries which take all the data from the first day again and again with each extraction run. Sooner or later that would lead to big problems. As a result of this they store the data in their Access databases today. I saw one example where they kept around 1,9 GB in Access and the frontend design was quite impressive - given the fact that the needed to use Access 2003 which was orders of magnitude away from the possibilities you now have with Access 2007 and the even more impressive capabilities of Access 2010.
On the one hand you mention the target group and the resulting feature set which fits to this positioning. But on the other hand you also offer quite sophisticated joining possibilities in the data feed (inner join...) which also requires a sophisticated level of database knowledge from the business user in case he wants to use this. This is to a certain degree not consistent with your argument given above why not to offer data storage facilities. On the other hand, maybe we all underestimate the IT knowledge of business users in the business departments. There are quite a lot of people out there which have a fairly good knowledge of how to join tables, who also have a certain basic SQL knowledge and some of them even do VBA programming. Not because they love it, but because they need to. If they rely on IT and accepted the long lasting leadtimes to solve analysis problems, they mostly would be lost. I know it, because I was involved in many DWH projects. The minimum project time is around 2-3 months. More often we are talking about 6 months and in some exceptional cases even 1 or 2 years project time. That´s the reason why they are looking for tools to help themselves.
How about the following idea: Understanding and agreeing with your concerns due to the fact that you want to adress the occasional self service BI user (however that acutally means in detail in terms of target group description), it might be helpful to also adress the users I described above looking for a more sophisticated data storage & retrieving possibility. If there was a kind of API which enables to store the data in an external database such as MySQL and MS Access (the first one freely available also for business users, the latter one being standard in nearly all organizations I saw in the last 12 years) and if there also was the possibility to read the data from this database, then such a solution could be offered on a project basis if somebody requires such a feature. As a result of this, ARIS Mashzone requires not only the API to export data, but also an enhancement in terms of reading the stored data from this database. What do you think about this suggestion?
Hope I do not bother you too much. I like the Mashzone concept and the funtionality - even if there is a competitor coming up now with an also impressive functionality (Microsoft PowerPivot). Keep an eye on them,... their concept is also pretty good & very powerful.
Finally a question regarding a potential bug...
However my own trials with Aris Mashzone suffer from severe starting problems. Currently I experience problems to start mashzone. Once I installed the product, it runs perfectly. Having shut down the computer and restarted, the Tomcat Server seems to have problems to start... Do you know this problem? The only workaround right now is to permantly remove the software and reinstall it again (or not to shutdown the computer...) Both is only a short term solution... I am using Windows vista including all support packages and IE08.
Best regards,
Joachim
Hi,
In addition to the Stephan’s answers some remarks from my side:
Target Groups:
More on the positioning and target groups you can find in the fact sheet on ARIS MashZone
Slice&Dice:
MashZone is not meant to be a generic OLAP tool. However, you can build you “personal” OLAP by building a Mashapp that filters data and retrieves data (parametrized queries). I build already an example for that where ARIS PPM is the data source and you can drill down on process instance level from MashZone.
Release cycle:
We aim to be flexible here. There will be a service release in the next 2 months and we plan an “Enterprise Edition” for the second half of this year.
So thank you for your input. And keep us informed on the results of your first own MashApps….
Tobias
Thanks for your feedback. The handling of two dimensional spreadsheets will be improved. But please always keep in mind that there is a difference between presenting data to humans and machines. So some spreadsheets might be structured for the sake of presentation only.
Regarding storage: Each data source is free to provide its own storage. There is even a product with in the ARIS platform doing this job for you. It is ARIS Process Performance Manager which implements a process warehouse. It is able to remember all your process’ data and corresponding KPIs for many years.
Your server starting issues: During installation you are asked whether MashZone starts automatically or not. Depending on what you selected you might start it manually after reboot. Shortcuts for this purpose are located within the startmenu.
Hi Stephan,
I know of course that the ARIS PPM has such storage possibilities. In a certain way it is a data warehouse with a strong focus on processes. And this exactly is the problem. How should anybody store the results of a mash up containing not only process data, but also some other data which is not process related in ARIS PPM?
As long as ARIS PPM focuses exclusively on process related data (maybe I am wrong regarding this assumption...?) it is a closed & proprietary application. Indeed quite impressive especially regarding the look & feel. ;-) One could even draw the conclusion that the look & feel of ARIS Mashzone is highly inspired by the very impressive user interfaces of ARIS PPM...
Anyway, I understand the reluctance regarding this storage topic, not only from a marketing perspective, but also from a technical perspective. Once you start to build a database below the mashzone, it becomes quite complicated to support the goal to deliver an enduser friendly tool which is easy to install, to configure and to maintain without the help of IT specialists...
... which by the way leads me to my very last point... I still encounter problems regarding the server starting issues. I was aware of the fact, that I need to manually start Mashzone as application via the shortcuts after a reboot of my machine. But trying this, the applications still is not starting. There seem to be problems to start the server. The server is showing a message that he is waiting to start in a certain amount of seconds and waits and waits and waits ... Surprisingly that has worked always perfectly the first time after a fresh installation. Once I reboot, the same procedure does not work again.
I tried to attach a screenshot here showing this Tomcat server message, but also without success. ;-) The first time, I tried to attach a jpeg file (size: 163kb) and got the message, that there is a limit to a file size of 125kb. So I saved it as gif (size 87kb) and then got the message that I am not allowed to attach pictures with more than 700 pixels. :-(
Your ungentle ariscommunity server seems to dislike me... ;-)
Since I am not an expert in picture editing to reduce the pixel size, I unfortunately cannot attach the screenshot for now. Sorry for this. But any ideas how I could solve the starting problem?
Thanks in advance for your help.
Best regards,
Joachim
Hello everybody,
I'm a heavy user (from process customizing to dashboarding and reporting) of Aris PPM and I'm studying about Mashzone's features. I find very interesting this discussion.
Of course, I'm really interested in integration with PPM data. I would like to use Mashzone as a fast and flexible Dashboarding tool, where I can merge data from PPM and other data sources. In this way, I think I can take advantage of PPM storage capabilities.
I'm now using the trial licence, so I didn't try to integrate Mashzone with PPM. Could you please provide me some extra informations? How is intergation performed? Does Mashzone directly connect to Aris PPM DB? Do I have to export PPM data (e.g. in .xml files)?
Thanks in advance
Marco
Hi Marco,
MashZone's PPM connectivity is part of the Professional Edion. So in your free Edition the ARIS PPM data source is not available.
If you are already familiar with PPM: To connect to PPM MashZone utilizes a feature of ARIS PPM called Query API.
In reply to Hi, I'm not sure what by bstowman
