Skip to Content

Following a recent talk I gave at an SAP event, a comment I made comparing “Delivering Self-service” to “The War on Drugs” was picked up and questioned on Twitter. The 140 character limitation made it a little hard to explain exactly what I meant so I never really answered the question. I had reason to expand on this subject yesterday so thought I would convert this into a ‘blog post. I’m very curious to poll other people’s opinions on this subject so feel free to agree/disagree in the comments or on Twitter (@mcbobj)

Yesterday afternoon I wrote an email to SAP and asked them to stop using the term “self service” as a noun, and start using it as a verb. It’s not just SAP who do this of course but I’d like to see SAP take the lead on this. The reason that, in my opinion, “self-service” has failed to be delivered in countless BI projects I have witnessed over the last ten years is precisely because of what I, and many others, see as the misuse of the word.

Self-service is a verb, not a noun!

Self-service is not, in my opinion, something you deliver, it is something you do with something that is delivered.

Example 1 Analysis, Edition for OLAP is not “self service”, it is a reporting tool (used for drill-down analysis of BW OLAP queries) that can be deployed in a number of scenarios and, in some of those scenarios, a user may also be offered self-service options but also they may not. If I create an IT delivered report in Analysis OLAP (as we do within our organisation) then only some users have the ability to save a modified version of that report – a scenario we refer to as “level one self service”.

Example 2 Lumira is not “self service” it is a product that is used to visualise data. IT could deliver a report, against functional requirements, in Lumira and they may or may not also allow users to modify/create reports with Lumira i.e. allow users to self-serve (a verb) with the product. I make this point because Lumira is always presented as “self service” by SAP and I strongly contend that description, here and everywhere!

Everybody has a (different) opinion

Herein lies the bigger problem – at any give seminar or event when someone talks about “self service” the first thing anyone says is “of course, everybody has a different idea of what self-service is” but then nobody actually creates an industry definition for people to follow or use as a baseline. It would be helpful, I think, if SAP actually produced something that defines the tiers of self-service conceptually.

For example, within the enterprise architecture function of our business, we define four levels of self-service that summarise as follows

Self-service, level 1 IT delivered report that a user has the rights to save a variant copy of to their personal folder (Analyst rights)

Self-service, level 2 IT delivered report that a user has the rights to save a variant copy of to a shared folder (Super user rights)

Self-service, level 3 IT delivered semantic layer (e.g. BEx query/Universe). User has rights to application (Analysis/Webi/etc) and ability to create document.

Self-service, level 4 IT delivered “enablement” for “composite data” allowing user to upload their own data, link with corporate data and then create own report.

In our company’s example of scenario 4 this would be BW Workspaces  + Analysis Office

Interestingly, when we asked our execs what they expected from “self-service” they replied: “Instead of having to ask an assistant, or somebody in an end market, to run a report for me, I want the rights to access BI and get the report myself, when I want it.”. So, from the point of view of the business, just having access to BI and the ability to run the report themselves is “self service”. We may actually include this in our company definition!

We are currently (successfully) delivering scenarios 1 and 2 and are prototyping scenario 4 (we had to roll out SPS08 for BW7.3 to enable BW Workspaces).

But…there is always “a but”

The most commonly attempted scenario in BI projects is scenario 3 and this is the one that I commonly see fail because the project tries to reduce reliance on IT and enable the business to do their own reporting, typically by providing them with a Universe and a copy of Webi. Even our report writers find writing reports difficult and this is why, potentially, the business never really achieves self-service with scenario 3, the business simply doesn’t typically have the resources to create reports from scratch, even in Webi, let alone build their own semantic layers. However with scenarios 1 and 2, any product could be used to allow self-service, even Crystal – with the right degree of training and governance.

I’ve left the word Tableau until the end (so those at SAP with a nervous disposition can look away now!). I mention this because we deploy Tableau but for  a complex supply chain diagnostic tool and it is very definitely IT delivered with no element of self-service attached. However Tableau keep telling us (and the business!) that users should be creating their own visualisations by creating their own semantic layers. I, er, disagree strongly but I am open to being convinced.

It would help those of us that deliver IT solutions, based on SAP, if we didn’t have to deal with products being labelled as “self service” but rather as products that had a purpose (ideally as part of an overallBI solution) and that could also be enabled for self-service.

Self-service (the noun) is dead! Long live self-service (the verb)! 😉

We have seen it said many times that the ‘War on Terror’ and ‘War on Drugs’ were not successful because of the noun. I’m suggesting we also consign the noun of self-service to the dustbin. Long live self-service the verb.

To report this post you need to login first.

8 Comments

You must be Logged on to comment or reply to a post.

    1. Mark Cooper Post author

      I guess it applies to many, possibly all, products so I wasn’t quite sure where to put it. If you RT the link on Twitter it will reach everyone anyway! 😆

      (0) 
  1. Donald Maccormick

    Mark, I would go further.

    As far as I can tell self-service is a seductive (but false) idea used by most BI vendors to generate more revenue. Unfortunately, it has the side effect of undermining our ability to get BI more widely used in organizations.

    In fact self-service BI (in all it guises through the years) is probably responsible for more shelf-ware than anything else in our industry.

    ALL “self-service” tools are focused on analyst work patterns, leaving the 80-90% of end users high and dry using a tool unsuitable for them.

    Your exec had it exactly right when they say “I want the rights to access BI and get the report myself”.  They didn’t say “I want a tool which I can create visualizations for myself” or “I want a tool so I can do new analyses on our big data databases”.


    The reason is that end-users want BI (sorry Timo) like the BBC weather site. An interface they can just use (without training) to quickly get the information they need to do their jobs. What they need is GIGWYNGOGO (Get in, Get What you need, Get out and Get on) access to information

    Interestingly self-service means this in other areas, e.g. a self-service restaurant is where I select ready made food myself, it doesn’t mean I go into the kitchen and start cooking.

    The sooner we realize this and stop trying to force-feed end-users analyst tools, the more quickly we will realize the full value of BI in our organizations and go a long way to reversing the shelf-ware blight which has dogged our industry for too long

    (0) 
  2. M. Filiberto

    Good article to start clarifying concepts like “self-service BI” and also “Big Data” (another myth in my opinion).

    I would also like to add a very old example: Executive meeting where the Director of Sales says:  “YTD sales 1.2 million and the Director of Marketing or Finance says: YTD sales 1.1 million” and half of the meeting is used to figure out which figure is correct based on what report/variant/Self Service BI action or combination was used.  I saw that happening in the past many times and today is still happening.

    After 20 years of Data warehouse and BI I still believe that a central repository of reports (corporate or template) should exists based on definitions validated by the business and controlled together with IT.  The possibility of creating your own analysis for some scenarios (like exploring data to find new business opportunities or to improve the current template) should exist but restricted to people from the business that is capable of doing that in terms of business, process, and system configuration knowledge and BI maturity.

    “Self-Service BI” tools as they are wrong classified in my opinion are making this last ” analytic process” simpler than before for this advance users (assuming good modeling behind) due to the power on visualization combined with the power that in-memory technology also brings.

    I really like all the new tools available in the market today when you compare to what it was some years ago, but again in my opinion the only way to maximize the benefit is to combine them with the good governance, data modeling and business maturity (right tool for right people).  I completely disagree with the concept “here is the universe / Bex query … enjoy the fantastic new tool and good luck”.

    I do not want to make it longer because there are a lot of other topics that need to be discussed toward a good solution that combines the best of all the capabilities available combined with business needs (BTW: good feedback “I want to run it myself and do not ask to my secretary”)

    Very good article!! and also very nice approach the 4 Self-Service scenarios proposed as a solution.

    (0) 
  3. Mark Cooper Post author

    I have moved this to SAP Lumira section as Tammy originally suggested. Also, after two days at Gartner hearing products being sold as “self service” per-se, rather than self-service offered as a possible/feasible capability of the product I felt the need to vent again. 🙂

    (0) 

Leave a Reply