Skip to Content
Author's profile photo Graham Robinson

Changing perception of usability at SAP

I was fortunate to get to SAP TechEd in Madrid this year. It was the first time I had been to one of the big SAP events in Europe for more than a decade and was a great reminder about some things that are particularly European.

For example, a lot of people wear suits. Sure because TechEd and Sapphire are combined events in Europe you expect the larger “business” audience to outgun the “techies” – but there are still a lot of people wearing suits. Also, a lot of people smoke – noticeably more than in Australia or North America.

Percentage of Males Smoking by Country – Source Wikimedia Commons

This file is licensed under the Creative Commons Attribution-Share Alike 3.0 Unported

Attribution: Emilfaro

One of the big messages for me from the TechEd season was the renewed focus SAP has on the user experience. This, of course, is not the first time SAP have attacked this issue.

While working for Oracle in the early 1990’s they came up with a plan to counter the SAP graphical user interface. The Oracle technology, Oracle Forms, was designed to work on terminals – typically the DEC VT series or similar. While Oracle Forms was very capable the visual experience between a VT220 terminal and the WIMPS interactions of SAPGUI on MS Windows was stark. So Oracle came up with a solution. They used a windows terminal emulator that included scripting capabilities and supported windows style rendering to take the fixed-font character experience of the terminal and transform it into what appeared to be a windows application. They called this solution “Client/Server Lite” to position it against SAP’s 3-Tier Client Server architecture. If imitation is the sincerest form of flattery then Oracle were acknowledging that SAPGUI was the benchmark for user interfaces in the enterprise.

Times have changed and there is no doubt that the dynpro programming model and the SAPGUI screens that it produces are pretty boring by contemporary user interface standards. Over the past 15 years or so we have had EnjoySAP, SAP@Web, SAP ITS, SAP Mobile Engine, BSP, CRM UI, Web Dynpro, SAPUI5 and probably several others that all in some part seek to provide better experiences for SAP users. Yet SAPGUI is still the mainstay of SAP user interface technology.

Recently Sam Yen took up the role of SAP’s Global Head of Design and User Experience. Sam has a very big job – and it would be fair to say a job, or at least some tasks, that other people have failed at before. I was fortunate to spend some time with Sam in Madrid and get his thoughts on his new job, what he hopes to achieve and how he will try to do it.

One of Sam’s primary goals is to change the perception of the usability of SAP. To me this is an important difference from previous UI renewal attempts that tended to only focus on new UI technologies. The simple facts are that SAP delivers to its customers about 16000 transactions – comprising as many as 30000 dynpro screens – as part of their applications suite which are used by the vast majority of SAP users. And these dynpro screens are not going away anytime soon – none of the new UI technologies SAP has released has ever made any significant impact on these numbers.

Adoption of new technologies in the SAP customer base is painfully slow. Some of the reasons for this include:-

  • Enterprise software is critical to the day-to-day running of most businesses across the globe – customers are very reluctant to change anything that might effect their enterprise applications because of the very real possibility of significant business impact if things go wrong. That is why the Vishal Sikka “Innovation Without Disruption” mantra is so important.
  • SAP applications are linked to a specific NetWeaver release and new technology is delivered in the latest NetWeaver release making it unavailable until an applications upgrade is done – no trivial project.
  • SAP have done very little to take existing dynpro transactions and port them to newer technologies. If SAP’s commitment to WebDynpro ABAP was measured in the number of traditional Dynpro screens migrated to WDA the results would be quite damning.

So the uptake of new SAP UI technology has been very slow – in fact almost non-existent. Take the example of a customer just now upgrading from ERP 4.7 to ECC 6. They have not yet done anything on WebDynpro ABAP because their current technical platform doesn’t really support it, and if they were to honestly look around now at the most appropriate UI technology to use for new user experiences they would be quite justified in choosing something else because WDA certainly does not deliver the consumer-grade user experiences that are expected in 2012 and beyond.

Sam intends to tackle the negative perception of SAP usability on two fronts.

Firstly, he is identifying the top transactions that pretty much every SAP customer uses. He sees this as being about 20-30 transactions in areas such as Purchasing, Financials, etc. He foresees that SAP will redesign and redeliver these transactions making full use of contemporary technologies and techniques to provide a much more engaging and effective user experience. My assumption here is that we are talking about replacement transactions built using SAPUI5, NetWeaver Gateway, etc. but it could well be something else altogether.

Secondly, Sam believes that each individual SAP customer has a further subset of transactions that are critical to their business. Again this might be around 20-30 transactions. For these transactions Sam sees a role for SAP Screen Personas. He believes customers will be able to use Screen Personas to “re-skin” standard dypro screens to provide better experiences for their users.

This all makes perfect sense – but for me I still don’t yet see how the slow adoption problem can be solved. Sam is in a hurry – he wants real movement on this issue in months not years so he needs to solve all the problems inhibiting adoption that already exist plus a few new ones like…

  • If SAP build 20-30 completely new transactions using contemporary technology isn’t that just the same as new UI technology solutions that have failed to be adopted before?
  • If SAP view Screen Personas as the answer to the 20-30 customer-specific key transactions why don’t they see it as the answer for the 20-30 universal key transactions? If it is good enough for their customers why isn’t it good enough for SAP themselves?
  • Even something as non-disruptive as Screen Personas requires a certain kernel level. This presents the same application/technology change management issues I mentioned above.

I wish Sam and his team all the best in the coming months as they work to change the perception of usability with SAP customers. I hope they are phenomenally successful and prove doubters like me wrong. In fact I hope in my small way to be part of their solution as much as I possibly can. But they have a huge task and will need to execute perfectly to achieve it.

Assigned Tags

      You must be Logged on to comment or reply to a post.
      Author's profile photo Jocelyn Dart
      Jocelyn Dart

      Hi Graham,

      Yes it's always a challenge moving the UI as fast as customer's wants and needs.  Wouldn't it be great if we could get more of the partner ecosystem providing modernized multi-device UIs for all those that don't make the top 20-30? Don't think I envy Sam right now but I applaud the ambition. Rgds,


      Author's profile photo DJ Adams
      DJ Adams

      You've highlighted very well one of the ironic downsides of pace layering - the plethora of layers, at various rates of change, at the top of the stack, and the impedance mismatch between the top and bottom. That said, it's almost a given in innovation and a (sensible and unavoidable) split between core and UI.

      Personally I'd love to see the commitment to SAPUI5 that is so obvious (and encouraging) to see at various SAP events (like TechEd Madrid) retro-fitted to some of the 'core' transactions of which you speak. There's a counter-argument to that of course, that the focus for SAPUI5 (and other newer UI tech) is less on the 'power user' and more on the frequent-to-casual user.

      Perhaps that's where some of the inertia comes from - the desire or assumption (right or wrong) that power users need 'all the features, all the time' as made available in good old classic dynpro based transactions. The 'Microsoft Word syndrome'?



      Author's profile photo Mark Teichmann
      Mark Teichmann

      One of the advantages of SAP Gui being around for such a long time is that most of the users of our various customer base are using the same screens for their work. This is a great help for our helpdesk supporting many users at different customers.

      Now if every user uses his own screen design for the same transaction (one using SAP Gui, the other using WDA, the next one using Screen Personas and the last one using a subset via SAP UI5) it will be much harder for the supporters out there to provide efficient troubleshooting here.

      The more complex the screen landscape is the more expensive the support will be.

      For sure the growing choice on UIs for the users is a nice thing, but everything comes also with a downside.

      Author's profile photo Former Member
      Former Member

      Hi Mark,

      I think you are making a very valid point here. Multiple UI frontends will drive support effort and cost, no doubt.

      Having said that, I think there could be technical solutions to resolve something like this. Solutions that wouldn't be beyond SAP's capabilities. If the front end layers use the same backend then you have a way to monitor "UI traffic" which could be used in a support scenario. Not easy perhaps, certainly a challenge. But not impossible to overcome.



      Author's profile photo DJ Adams
      DJ Adams

      Hi Mark

      Good point about supportability and loss of consistency for look and feel. That's partly why I do like the SAPUI5 Shell ( - provides some consistency not only for the support / helpdesk but also for the actual user (i.e. the 2nd and subsequent apps / transactions are 'familiar').

      Author's profile photo Former Member
      Former Member


      Thank you for bringing UI back into the picture. Despite the furore at TechEd around the announcement of putting UI redesign much, much higher up on the agenda, it's become usually quiet again around this topic.

      My sentiment is that amongst all the mobile talk pundits have lost sight of the "natural habitat" (to quote Thorsten Franz) of the enterprise user: the office desktop/laptop environment.

      In terms of redesign and Sam Yen's massive tasks, I think there are short term and long term goals here: In the short term, tools such as Personas can help to simplify and appease user bases that are becoming increasingly fed up with GUI for Windows/Java. In the medium to long term, I agree with DJ's comment in that we should have a clearer commitment to UI5 by SAP, who has to lead the way here.

      Kind regards,


      Author's profile photo Mark Teichmann
      Mark Teichmann

      Maybe in the near future a large userbase is using Windows 8 with SAP Gui (some of them using Touch Screens). Will there be an optimized solution for Metro style SAP Gui available?

      Author's profile photo Former Member
      Former Member

      You are making two massive assumptions here:

      1. a large userbase of Windows 8 in the near future

      2. an optimized solution (metro-style) SAP GUI

      No one will be able to predict the Win 8 adoption in the enterprise space. Gut feeling tells me it will be sluggish, but I insisted an iPad Mini is wrong, so who am I to tell?

      Generally, I do like the idea of a light-weight, metro-style SAP GUI though.

      Author's profile photo Steve Rumsby
      Steve Rumsby

      I would agree with Windows 8 roll-out being sluggish. We've not yet completed our Windows 7 roll-out, and so have some machines still running Windows XP.

      And as for touch screens, I doubt many people see a point in those for desk-bound activities. I'm guessing they'll become more common, possibly even the default, on laptops over the next couple of years. The vast majority of the machines at my organisation are desktops, though. Will they get touch screens? I seriously doubt it. So a Metro-style SAPgui just doesn't make any sense.

      Author's profile photo Former Member
      Former Member

      Would your organization look into screen simplification a la Personas ?

      Author's profile photo Steve Rumsby
      Steve Rumsby

      Up to a point, yes. If you allow a total free-for-all, though, you get the support issues mentioned above by Mark Teichmann. So long as the screen re-designs are implemented and managed centrally, I can't see a problem.

      I have, over the years, experimented with doing screen simplification via GuiXT, which I know isn't as capable as Personas, but even that has the potential to cause support issues if users are free to do what they what to screens.

      This is a really difficult issue. I'd like to be able to let end users redesign screens to suit the way they work, but I also need to be able to support all 800+ users and I can't do that if their screens are all different. 

      Author's profile photo DJ Adams
      DJ Adams

      It's easy for me to sit here and say this (and I'm painfully aware of the reality) but I will anyway. Why should we even be considering the OS in a UI rollout? I thought those days were (almost) behind us.

      Author's profile photo Mark Teichmann
      Mark Teichmann

      These days will be behind us when all 30000 screens are offered as a responsive SAPUI5 screen 😀

      Author's profile photo Former Member
      Former Member

      Indeed, reality certainly bites here.

      Let me play devil's advocate: Why do we always have to assume that future access to SAP should mainly be via a browser based UI, which -in comparison to native OS desktop applications- is likely to be slower?

      Author's profile photo Steve Rumsby
      Steve Rumsby

      Because browser-based apps take less effort to make work, and keep working, on multiple platforms? (He says, as someone who has never had to deal with browser differences, and may be completely wrong:-) And for many users, the speed difference is unimportant.

      For some users, though - think AP clerk spending all day, every day typing in invoices - the speed difference is critically important. A browser-based MIRO may not be that popular!

      Author's profile photo DJ Adams
      DJ Adams

      I'd like to challenge the notion that browser based apps are slower than native. Or at least question that notion.

      • Witness the work that is going into making the DOM renderer and JS engine faster and faster (effectively what is now becoming know as the "new VM" -- and I don't see anyone questioning the Java VM much these days)
      • Look at the recent post about Native iOS vs Web with Sencha (albeit mobile based)
      • Also think about what fast means to some people. If I take the horrendously clunky mail client Outlook (native), which gets in the way of you performing your mail tasks, and compare that with Gmail (web-based), which allows me to complete my mail tasks in a lot shorter time, I ask myself why? Gmail is a lot faster not because it's on the web (but I don't notice any speed difference app-wise anyway) but because it's designed right for the task and fit for purpose.


      Author's profile photo Steve Rumsby
      Steve Rumsby

      Right. Just like "real time" is less important than "right time", so "fast" is less important than "fast enough".

      It is interesting that you compare gmail and Outlook, a product specifically designed to do one thing well, compared to a monolith designed to do many things through a single interface. Seen that dichotomy elsewhere, perchance?

      Author's profile photo DJ Adams
      DJ Adams

      Yes, agree with the differences and the weighted importances.

      On Outlook vs Gmail, I had a bit of a rant about one-tool-to-rule-them-all a while ago: - I do believe in the right tool for the job. Interestingly the browser has moved from being a specific tool (web browser) to a general tool runtime (i.e. VM for HTML5 apps).

      (ps I'm sure there are people watching this thread that will point out that I'm not a big fan of Microsoft software 🙂

      Author's profile photo Robert Shiels
      Robert Shiels

      You are not a fan of MS Software!

      We're already using SAPGui at my place, and new ESS/MSS functionality in the Portal (using WDA in fact). Old and new, two more different interfaces are hardly possible. People adapt, they don't need everything to be the same. They can cope with Facebook on their smartphones, MS Word on laptops, Google Mail everywhere etc. To the user, it really doesn't matter if the application is native, or in the browser.

      Author's profile photo Steve Rumsby
      Steve Rumsby

      Because those days are still not behind us? And even when they are technically behind us, in practice large organisations are typically so far behind the curve, for good reasons (usually), that OS considerations will stay with us for a long time yet.

      Frustrating for some of us, but a reality nonetheless.

      Author's profile photo DJ Adams
      DJ Adams

      Agreed, and hence my bracketed "painfully aware of the reality". A client I recently worked at still have a massive install based of Vista + IE7 which has been around for a while and will probably continue to be around for a while.

      Author's profile photo Steve Rumsby
      Steve Rumsby

      As I said above, there are still sections of my organisation running Windows XP, although that will hopefully be finally eradicated in three months or so. But then we'll be on Windows 7 for a long time I suspect. Well into the age of Windows 9, and possibly 10 - if such things ever come to be.

      That is a really big problem for developing new UI technologies, but it is a reality that isn't going to go away. I guess for many, browsers get updated more often than OSs, and so browser-based UIs have more chance of making use of modern technologies. But even then, as your example shows, some organisations tie down the browser too.

      Author's profile photo Thomas Wailgum
      Thomas Wailgum

      Great topic, Graham, and great discussion in the comments. I recall talking to Sam at TechEd and hearing both the "need for speed" imperative with this but also the stark reality: it's going to take (at least) 5 to 7 years to make significant change. What will the business + technology world look like then? What devices will be using? My colleague at ASUGNews, Courtney Bjorlin, interviewed Sam for an article recently, in case you're looking for more information on the topic.

      Author's profile photo Former Member
      Former Member

      Thanks Graham.

      With this blog it gives me the feeling I am not alone.

      Even if I do understand where SAP Screen personas can bring some business values I am still concerned that it pushes us in a corner.

      An enhanced User Experience is not only about nicer and simplifies screens (and an end user may even realize it).

      That's why we have seen SAP doing investment for example with Side Panel (and other tools associated to WDA framework).

      So if customers are spending time to redesign with SAP Screen personas and it is well perceived by end users then the adoption of foreseen strategic SAP UIs will be stopped.

      My point of view is that strategic SAP UIs should allow to give the same personalisation capabilities as SAP Screen Personas (assuming this product is well adopted by the customers).

      My quote for Sam (if he is reading this blog) : it is not fast if it is not right.

      I have done my own analysis of SAP Screen Personas potential use and business value ... but I am now under strong sales marketing pressure of SAP (guessing related to one of the line of your blog).  

      Author's profile photo Former Member
      Former Member

      So if customers are spending time to redesign with SAP Screen personas and it is well perceived by end users then the adoption of foreseen strategic SAP UIs will be stopped.

      Is that such a bad thing? Is it just me or are you saying that -despite users liking something- you are putting strategy first?

      I have done my own analysis of SAP Screen Personas potential use and business value ... but I am now under strong sales marketing pressure of SAP (guessing related to one of the line of your blog).           

      That's also of interest to me. Would you like to share the results of your analysis? What did you base it on?


      Author's profile photo Ingo Deck
      Ingo Deck

      I have to admit I really like this blog from Graham but especially the discussion that started below. It shows perfectly - while all of you (like myself) have quite some empathy for the topic usability, user experience or however you want to call it - that the topic is not really simple and everyone has his/her own approach to the topic, sees different priorities, prefers different options/paths... which from my perspective is the tricky thing: UX is not an objective thing that can be achieved... I always liked to say that if you ask 10 different people about the usability of something you get 12 different answers... so this discussion for me is a good reflection of the complexity of this topic.

      For sure there are lots of "generic improvements" that can be made to SAP user interfaces which will be liked by the majority of customers or better users, but the last mile often is a specific path for a customer, even more often for a user. So the flexibility SAP offers is also key for real UX success for customers.

      So if customers are spending time to redesign with SAP Screen personas and it is well perceived by end users then the adoption of foreseen strategic SAP UIs will be stopped.

      My point of view is that strategic SAP UIs should allow to give the same personalisation capabilities as SAP Screen Personas (assuming this product is well adopted by the customers).

      I would also agree to Michael and ask the question if it is such a bad thing if customers use SAP screen personas if that solves some of their UX related issues for their users.

      But I would not agree that this will stop adoption of new strategic SAP user interfaces/applications as also SAP Screen Personas will not be the last and final solution to all UX issues.

      Both are ways for our customers to tackle the UX topic... for one customer the SAP Screen Personas way is the better one in their current situation, for others it is a different option (like adopting new apps based on Gateway/SAP UI5). And I am pretty sure that one or the other aspect, that is really solved very nicely in SAP Screen Personas will be taken into other product areas as well as you can for sure learn a lot from solutions that get quite a good response from the market.

      Again from my perspective there is not the "one way that fits all needs", but with the diversity of SAPs customers SAP also has to offer options.

      Anyway, just my 2 cents, and already looking forward to new posts on this blog.. really interesting to read, so thanks to all of you for sharing all your thoughts, concerns, ideas...



      Author's profile photo Former Member
      Former Member

      As stated by Ingo "User Experience" is a tricky subject.
      I am not sure it is easy to explain everything through a blog.

      I am not expert in that subject but I am learning.
      One of the reference point I have is this website

      I have used it to quickly assess the maturity level of my "customers" : level 2.
      Main characteritics : they rely on Intuition, "Good User Experience is an high priority" which means nice screens.

      By feeling the pain of the SAP end users at work I quickly identify that some "common features" were missing (especially around contextual help, error message handling, value selection, contextual information, spell-checking for free text and so on  ... and of course screen noice, lack of real estate use, ...).

      On top of that the main challenge is the change management.
      Any new User Interface roll out will require a Change Management (= Training).
      Of course when you have a couple of 100 users it is not a big deal ... but when you have a couple of 10K users that's not the same story (so something may not be a technical disruption but could be a business/people disruption requiring associated effort which may not cheap).
      That's why it is so important to think twice before rolling out a new UI.

      One of the factor is that "end user" does not realize that "beautiful, simplified" screens is not a panacea.  Without knowing it they need more.
      On top of the that there is more and more need to run the same transaction on multiple device (including iOS one).

      Some of the "Usability" constraints explained above are not adressed by SAP Screen Personas.
      Personally I believe WDA/NWBC/Side Panel/FPM/... should help to meet some of the constraints (my customer and myself do not know all the underlying-unknown-undiscovered capabilities).

      Based on what I have seen SAP Screen Personas may have more capabilities for "end user screen personalisation" (to be verified) than the strategic UI.

      SAP has presented SAP Screen Personas as an enabler to revamp some old screens which are not widely used. This approach makes sense.

      What does not make sense for me is to modify transactions 'widely used' (FI0x,VA0x,ME2x,MM0x,..) ... I would expect SAP to revamp these screens with the latest technology (if not already done).

      I would like to avoid multiple change management challenge.

      I foresee the following nightmare.
      I enable the product for some end users without giving constraints (for example checking/knowing SAP Screen Revamp roadmap).
      They redesign screens when we could do more or less the same with the perceived SAP UI strategic directions.

      Once done end-users discover they need or they ask more usability items then I am in a corner.

      I am not saying I have not found "business value" case for SAP Screen Personas.

      I do not feel a SAP consistent message.
      SAP does refer a lot to UI Revamping in their Enhance Pack ....
      Then in some SAP Screen Personas presentation they refer to identify the most common transactions.

      My personal expectation : SAP should deliver the top 30-50 transactions on their strategic UI framework (WDA-NWBC-...).   It is easy to find !!!
      On top of that they should deliver the same level of personalisation as SAP Screen Personas on their Strategic UI Framework (if not already done).

      SAP Screen Personas should be "reserved" to the other transactions they do not plan to revamp.

      I do not want to go in details in my personal intuition as I need to sell the approach to my customer and verify it through UCD.
      I will recommend him to control/to limit the use of this new SAP tool (any how it will be my customer decision ... on top of that I do not know yet the licensing price which can always be a surprize : remember the story on Adobe Interactive Forms .....).

      For most common transactions it could be a tactical solution to "pilot" the business added value of screen simplification .. to justify investment on the SAP strategic UI framework. 

      So my bottom line comment : only UCD approach will allow to identify the right Usability challenge .. do not trust end-users when they try to tell you what is a good user experience (I do not like the word User Friendly for example). It is easy to tell what is a bad user experience but it is not easy to define what is a good user experience.
      My personal recommendation : for now 10 years my customers have not adopted new UIs .. it means we need to take another process approach (as tools is not the main roadblock).

      Having said I am sure I will be able to find a "business added" value case ... but it could be pressure on the expected feature of the SAP Strategic UI framework.

      About nr of options : sometime less is more ! You have seen in other blogs the nr of UI tools .. it becomes difficult for a customer to feel comfortable with this amount of choice.

      My 1 cent ....

      Author's profile photo Tom Cenens
      Tom Cenens

      Hi Graham

      Interesting blog post, thanks for sharing.

      SAP is still evaluating it's partner strategy for SAP Persona's which sounds to me like things are taking too long and slow adoption is right around the corner.

      Best regards


      Author's profile photo Graham Robinson
      Graham Robinson
      Blog Post Author

      Let's hope you are wrong. If you hear any details let me know?

      Author's profile photo Former Member
      Former Member

      Hi Graham,

      I used SAP some 15 years ago and started again working with it recently.  It was quite a shock to notice that the UI has changed very little from the 90´s. And there is a large number of usability issues starting from the login screen. The overall impression of the system based on its UI is sloppy and outdated. Having a reputation of an extremely complex system is not a merit but an indication that there is something wrong. And usability for the size of company that SAP is should not be something mystical. Finding the most talented usability experts of the world should not be difficult either. To my understanding, leaving the system look & feel & function as if ‘your nephew made it in the garage’ is pure ignorance. Take the 10 usability heuristics of Nielsen ( and start working. ps. Starting to use this site consisted of numerous phases  and lot’s of error messages.

      Author's profile photo Graham Robinson
      Graham Robinson
      Blog Post Author

      Hi Marjukka,

      welcome to SCN.

      You are right that the standard SAPGUI user experience has not changed radically since the 1990's. The great thing is that the transactions that were built in the 1990's still work on the latest SAP technology platform.

      There are, however, many other UI technologies now available in the SAP technology suite that can provide a much more contemporary user experience. Hopefully you will get to experience some of them soon.


      Graham Robbo

      Author's profile photo Former Member
      Former Member

      Hi Graham,

      I'm glad there are new UI technologies available.

      Though it does not change the facts that - as also indicated in the comments of this article - the majority of users still use the old clients and there are some serious usability issues that in my opinion should have been fixed a long time ago. Or at least then communicate to companies that are investing in SAP that the old technologies are no longer developed actively and it is their problem if they still are willing to put their money into a traditional windows client solution.

      And being backwards compatible is something you should expect from any system.

      Cheers, Marjukka Tohkala