Does the enterprise really need a consumer grade UI?

SUMMARY:

In part two of my enterprise UX piece, I share expert views on enterprise UX and then give you my take.

Phone and spring landscapeEditor’s note: This article won the ERP Focus 2014 ERP Writers Award for best ERP article of the year.

When I bailed out on – err, I mean finished – part one of the enterprise UX piece, I didn’t get to my own position on whether the enterprise needs a consumer-grade UI.

I did make some progress in the first piece (Is the enterprise user experience overhyped?), mostly by establishing that enterprise UX has become a defining advantage (or disadvantage) for software vendors. For customers, mobile initiatives are raising the stakes and forcing the user experience issue – otherwise adoption will lag and mobile deployments will either fail or fall short of glowing expectations.

But there are some problems gnawing at me. I put my nagging UX questions to Ray Wang and Holger Mueller of Constellation Research. (To ensure a quality UX, I handled the questions via Gmail). Wang has been pushing UX before it was particularly fashionable to do so; Mueller spends a great deal of time in the HCM space where UX is a mandatory discussion.

Three burning enterprise UX questions and answers

1. Is enterprise UX as important as hyped? Wang’s view? Yes: ‘UX is crucial as it drives adoption, enables productivity, and helps reinforce brand experience. It’s very important.’ Wang went on to say bluntly that ‘deals are lost because of UX.’ Mueller agrees, but he places some caveats on just how good an enterprise UX must be:

It’s important for adoption – but you only have to be the relative best. It’s not like users can leave your app like they can in consumer-grade apps. All the blah blah on ‘consumer-grade UI’ is marketing. As long as employees can only go on vacation if they put their vacation request in, they will do it – no matter how bad the UI is. For voluntary usage, UI more important. But then speed, data,  online/offline matter. There is still an offline sales CRM indudstry.

2. Is consumer-grade level app look and feel necessary? I asked the guys if other factors like speed and offline access are more important than an amazing UI. In other words: do we need to deliver a ‘beautiful and delightful’ experience now?

Wang says the type of user impacts the answer: ‘The use case will dictate what’s required.  Power users may want heavy tables and lots of functions.  But business users will want consumer grade apps or face an uphill adoption battle.’  Mueller is blunt: ‘At the end productivity matters to customers. Being good enough and not losing deals suffices for the vendors.’

3. Do customers have the skills to do this in house, or do they need outside designers? Implicit in this question is another one: do developers need to become designers? The short answer? Outside designers are needed. Wang noted that cloud deployment is another factor: ‘It’s best to work with a design firm to lay out a framework that you can build and extend over time.  UX/UI often is something you need to refresh fairly often.  Hence, it’s easier in SaaS/Cloud models.’  Mueller thinks UX is moving too fast to address it internally:

Outside help is needed. UX moves too fast and becomes stale in-house. Companies should always go outside. Enterprise developers can’t learn UIX – they do not care for it enough and struggle with accepting anyone has a problem with ‘their’ software. UI goes down the drain if you leave it to developers. You need frameworks that they cannot break otherwise you end up in consistency ‘hell’.

Industry and data model nuances can’t be ignored

We already touched on the importance of UI simplicity, but what about an ‘invisible UI?’ In an August 2013 post, Vinnie Mirchandani called out the enterprise UX for being too bulky:

I was reading this Wired quote about Dieter Rams’ philosophy about product design: ‘It doesn’t draw attention to itself; it merely allows users to accomplish their tasks with the maximal amount of efficiency and pleasure. At its best, it is invisible.’ And thinking, in contrast, how loudly ‘visible’ enterprise software UI is. It is a major competitive feature in enterprise deals with plenty of screamns about ‘our UI is prettier.

Mirchandani goes on to cite UX innovations in retail, health care, home devices, and automobiles. In each case, designers are rethinking user experience in the context of smart devices and tech advances. In retail, facial recognition can identify VIP customers. Self-checkout spares checkout lines. In the home, smart cookware keeps pans hot, smart thermostats warn of hot spaces with color-coded lights.  Healthcare is pushing out wearable tech of all varieties. The list goes on, but it’s all part of UX now. Mirchandani wraps by challenging enterprise vendors to think beyond their limiting design assumptions, moving past the keyboard and mouse.

Fair points – but enterprise data must still be reckoned with. Diginomica reader Baruch Sachs argues simplicity in enterprise UX means reckoning with the underlying data model:

One critical piece that non one really talks about in the enterprise world is the data design constraint. For a consumer app? You could make it look almost as clean and concise as you want it. In the enterprise software world, people see that and say ‘I want that!’ However, when you have critical business apps that through regulations, or disparate business groups require so much data on screen it becomes very difficult.. There is a huge amount of work and data analysis to be done by UX groups to sift through this data and identify which pieces actually need to be on screen so someone can do their job better.

Enterprise UX – my take

I ran out of space in this post before I could raise questions about what a ‘consumer-grade UX’ actually is. Too often, Facebook is held up as a gold standard, but I could poke so many holes in the Facebook UX it would look like a colander.

For a seamless integration of community and commerce, I’d pick Amazon. And when I chose a newsreader after Google kiboshed Reader, I chose Newsblur because it had the best performance for 1,000+ feeds – even though there were better looking (e.g. consumer grade) newsreading alternatives. But that’s a debate for another time.

Here’s how I see the enterprise UX issue:

  • Distinguish between UI and UX. UI is a smaller part of the overall UX where offline capability and back end integration might trump a sexy screen.
  • The mobile experience should drive UX design towards simplicity, but those lessons should be brought back to the desktop.
  • Speed is (almost) everything, data integration is next in line.
  • Business and casual users have a higher UI standard.
  • UX rarely works without a design process that includes the target user.
  • Apps that are constantly revised and updated are usually better (e.g. ‘start small’).
  • Be very careful about apps that require fast broadband to work effectively (what if your target user is on bad airport wifi?).
  • ‘Consumer-grade UI’ is not a requirement, but it’s a healthy long term aspiration.
  • Great UX means new skills and design processes – outside experts may be needed.
  • Industry nuance and user role are critical (one UX rarely suits all).
  • Process complexity is a poor excuse for ugly design; simplify the data model for the mobile screen if necessary.
  • Failure is less concerning than the status quo, which will result in users going rogue and downloading their own apps.

Final thoughts – what’s next

Compared to where we were three years ago, enterprise UX has come a long way. The reality may leave a lot to be desired, but if vendor marketing urgency is any indication, consumer pressure for a better experience is winning out.

I was curious to find out which vendors have a leg up on UX. I asked Mueller and Wang for their faves. Wang: ‘Some great UI is IFS. Their mobile apps always had great UI. Infor has done a great job with Hook & Loop, and SAS Visual Analytics is pretty amazing.’ Mueller’s picks? He likes the Oracle iPad HCM app, and agrees on Wang’s IFS and Infor examples. At the end of Brian Sommer’s piece on SAP’s UI Makeover, Sommer cited Infor, SumTotal and Sage.

Last year, at the Mobile World Congress in Barcelona, I kicked tires on a number of enterprise apps, and just as many consumer apps. The best of the consumer apps were eye-popping, with a level of design inspiration still sorely lacking on the enterprise side. But they were designed by a different type of developer. That gap is starting to close. More enterprise app developers are integrating iOS and Android rock stars into their practices; more vendors are pursuing Infor’s Hook & Loop approach.

I’ve only scratched the surface on enterprise UX, but my pal Vijay Vijayasankar tells me he is restless to riff on this topic so he can bat cleanup – take it away Vijay.

Image credit: Phone and spring landscape © Givaga – Fotolia.com
Disclosure: Oracle and SAP are diginomica premier partners.

    Comments are closed.

    1. vijayasankarv says:

      jonerp done..the ball is in play again http://wp.me/pKkXk-D0

    2. ultan says:

      holgermu glad you liked that tablet. When are you coming into Oracle HQ? wait till you see apps on touchscreen Chromebook!

    3. thearttrooper says:

      holgermu jonerp ConstellationRG rwang0 it matters. Users form an emotional bond (which lasts) much quicker with a UI designed for them.

    4. says:

      re: Newsblur: I signed up just out of interest and saw this: There are 228 people in front of you, all patiently waiting on their free accounts on NewsBlur. By going premium you can get full and immediate access to NewsBlur. < Now that’s what I call a hook to buy 😉

    5. greg_not_so says:

      gscott16 jonerp MikeGolz applebyj vijayasankarv twailgum < or maybe it doesn’t?

    6. dealarchitect says:

      Thanks for the mention. Enterprise UI is certainly getting prettier – interestingly from the backwaters of HR. In the last month I have seen gorgeous mobile interfaces from Workday and Oracle. But ent sw is still too timid. Maybe they need to spend some time in the auto industry which is trying out all  kinds of voice, haptic, eye-tracking even kick gesture design. I think every ent sw developer should see the movie Her. Not one keyboard or mouse in sight in the futuristic LA in the movie – all voice, hologram interfaces. But even bolder would be for ent sw to say for many process flows we don;t need UI – many data elements can be scanned or otherwise captured at source and should not need human manipulation. Efficiency is also pretty!

    7. kumarrk21 says:

      You raise good points. I may be stating the obvious here, but ultimately the importance given to UX/UI comes down to cost as well. Consumer apps deal directly with customers and impacts a company’s ability to sell. Enterprise UX/UI (loosely assuming back office type apps here) on the other hand, impacts the internal users and many times, IT budgets for such apps are minimal at the best. Performance/Integration is important and so is also security, encryption, controls and satisfying regulatory requirements, which don’t tend to be the major concerns for consumer apps. For e.g., one of the apps I designed would have provided a seamless experience for a user if I had used a file sharing tool like ‘Dropbox’ to sync data, but the security policies wouldn’t let me use such a service, thereby forcing me to use a not so optimal service and ultimately impacting the UX
      Thanks for this blog – KK

    8. MikeGolz says:

      “greg_not_so: gscott16 jonerp MikeGolz applebyj vijayasankarv twailgum < or maybe it doesn’t?” > “my takes” are a balanced summary

    9. greg_not_so says:

      MikeGolz gscott16 jonerp applebyj vijayasankarv twailgum < agreed!

    10. holgermu says:

      ultan looking forward to see it. Always excited when I see great usability.

    11. applebyj says:

      MikeGolz greg_not_so gscott16 jonerp vijayasankarv twailgum want to record some video on this? In NSQ week after next?

    12. MikeGolz says:

      applebyj greg_not_so gscott16 jonerp vijayasankarv twailgum > unfortunately, not in town week after next. Other time ?

    13. martin_english says:

      This may be a long bow, but here goes ….

      Your enterpise has an ERP (Salesforce, Oracle, SAP…). How many reports are generated by spreadsheets or other tools drawing on data from the the ERP ? How much data supplied to the ERP is generated by tools external to the ERP ? How many spreadsheets are maintained in some dark corner of the enterprise, that don’t even use the ERP ?
       At best, the data is present in the ERP, but it needs to be converted to and from the ‘SAP Standard’. In many cases, the reports / spreadsheets / etc are not even using the ERP.

      Each and every one of these is a failure of UX (if not UI). If that was the only problem, it would be bad enough; the extra cost in maintaining these extras is very rarely captured properly.  The real issue is tha tthere is usually no formal control over these extras and therefore the accuracy of the results can not be guaranteed  In other words, poor UX leads to poor quality data and poor quality data leads to poor decisions.

      BTW, if you want to be particularly pedantic, the same thinking could be applied to ANY customization of the ERP.

      hth

    14. says:

      martin_english Martin thx, your comment adds a lot here. In a way I think you are artciulating not only why the definition of UX needs to be expanded/adapted for the enterprise. You’re also speaking to, I believe, a broader groundswell that is expecting UX improvements. Not just from a consumerization angle.

    15. says:

      kumarrk21 Right – no doubt compliance/security/budget constraints impact what is possible for enterprise UX. However I think the point is coming where you have to square off against those things because the risk otherwise is to roll out projects that are simply not used or adopted. In some cases you can force some adoption but in other cases, rogue apps will be adopted by users and/or a disruptive vendor will come in and start spreading a better experience like a locust. 
      But planning for UX success definitely means thinking about the issues you’ve raised, glad you added this.

    16. says:

      dealarchitect Vinnie thx. In my online research yours was the only piece the look at the impact of UX industry by industry and embedded/smart tech etc.. I don’t see how you could approach this topic other than with that lens. While in Las Vegas, I watched an animated card dealer who was scary good at her job. I didn’t do a comprehensive study but her table was always pretty full – it was a brilliant design. And yeah, no mice or keyboards – just the “hit me” type buttons.

    17. says:

      dahowlett One thing I like about Newsblur is their firm ambitions to monetize. Doesn’t mean they will last the test of time but I’d rather pay a pittance for a great product I can count on than have free versions cancelled right and left. In a sense that becomes part of a bad user experience (moving from product to product.)

    18. says:

      But should you be expected to pay extra for the UX that the vendor is pushing? Definitely see that vendors will lose out if they have a poor UI/UX, but it doesn’t seem right to have a key part of the UX as a separate line item in the pricing configuration.

    19. says:

      bryanoak  This is an open debate. Some vendors think it is OK but then they have highly complex price books. Most people think this is a mistake.

    20. dealarchitect says:

      bryanoak a while ago I was discussing with the UX lead for a large vendor about auto industry UI innovations – voice, haptic, eye tracking erc. He was dismissive – only at the high end – Lexus, BMW – he said. I told him Ford, Hyundai etc were rapidly bringing them down market and he still did not agree. The irony was lost that his products are already high end in the software world as far as pricing goes. Probably thinking of a new line item as you say.

    21. says:

      bryanoak It’s a good point and one I alluded to in part one but since we had flogged that extensively in the SAP Fiori piece Den linked to in his reply I decided to put that off for now. I do plan to blog on this topic from another direction in terms of the ‘free enterprise,’.
      But to answer your question briefly no I don’t think vendors should charge customers for UI enhancements for a very simple reason: I think they will lose deals and market share if they do. There are caveats of course, if you read the Fiori debate comments one issue is that when you offer a new UI for older on-premise products, there are often unavoidable technical investments even if that UI is free. 
      From a customer angle, when you consider many enterprises are still running IE9 or even lower as a standard browser, you realize that customers have a technical debt of sorts and they will have to upgrade not only the browser but a range of products to offer better UI experiences. That’s not easy. 
      Many of the best UIs will come from third party developers and partner apps. And those will usually come with a price tag, and rightly so. 
      In sum you raise an important but that must be debated but even if you believe, like I do, that customers shouldn’t be charged for UI enhancements, the answers aren’t simple.

    22. says:

      dealarchitect bryanoak  Vinnie – You already know that most inputs in MRP/ERP are automated via 3rd party systems – no UI change required. As for the rest – can you imagine trying to input this statement: “Show me product profitability via region aggregated per quarter and grouped high to low” via voice when Siri can’t tell the difference between a burrito and burial?
      And please don’t beat me up on ‘old’ ERP. Like Pacolini before, it isn’t going away. Any business situation ends up in a transaction that needs to be subject to governance. Fancy explaining how voice/haptic etc would work to auditors? Or providing test cases?

    23. dealarchitect says:

      dahowlett dealarchitectbryanoakDennis, with that attitude we would have never introduced the POS scanner and would still be entering each item at the grocery aisle manually. As an accountant are you telling me every depreciation journal entry, AP invoice, T&E needs to be manual? Plenty of small and large businesses where still manual and many more examples

      On Siri, we pay for enterprise grade software why do they have to use Siri? For the billions that keep going into mouse and keyboard UI improvement projects the industry could have been enhancing voice interfaces. Again, auto UX designers don’t have the luxury of insisting on keyboards at the wheel so think and invest in other UI. Not saying voice will be the most efficient for every transaction but we have not given users a chance…

      just about 5 years ago the most popular mobile devices had a physical keyboard – look how far that landscape has evolved……

    24. says:

      dealarchitect You talk as though nothing is happening and that’s not true. You ask for consumer grade experience but the technology in many cases isn’t needed or isn’t up to snuff for enterprise purposes or there is a management dimension.  

      On recurring entries, we did a lot of that automated work 20 years ago. Today, we’ve gone further with learning systems capable of automated capturing and allocations for around 90-95% transactions from banking systems as example. 

      In many cases we’ve (finally) integrated customer and AR data so that field service/sales doesn’t sell to the wrong kind of customer and/or problem solving can be accelerated, DSO reduction etc. 

      On enterprise grade voice, the tech is not that much better than it was 15 years ago when I was following Lernout & Hauspie, Dragon and Nuance. 96-97% is not good enough. If it was then the keyboard and mouse would not be required today. 

      In reality, there are many cases where the keyboard  – whether virtual or real and mouse ARE the best alternatives. Fancy dictating a book from start to finish with all the corrections/edits that go with it?

      In enterprise, approvals of all types are routinely undertaken by handheld devices. T&E is being automated but there is a long way to go. Far more complex in enterprise land than people think what with policies, approval methods etc.

      BUT – if you look at broad scale adoption, you quickly find the blocker for those things that can be automated is not the tech, it is a management issue. 

      In short, much as you might wish but as Jon’s article points out, UX is far more complicated than simply applying a piece of tech.

    25. dealarchitect says:

      dahowlett dealarchitectDennis, may be different in Europe but in US plenty of AP invoice processing, T&E, Asset tracking, matching of docs is manual. I will give plenty of time to vendors who describe improvements in OCR, mobile capture of expenses, bar coding of assets, algorithmic matching of docs compared to yet another UI pitch. Maybe just me but that would be far better impact on white collar productivity…

    26. says:

      dealarchitect dahowlett – without wishing to sound rude but you’re not talking about a lack of tech but a lack of adoption. In much the same way that the physical methods of bank/credit card transaction completion are years behind Europe, food labeling is in the Stone Age, fraud detection is poor and the amount of paperwork generated by the simplest transaction makes no sense whatsoever. The tech of which you speak is available to US companies. It is hardly the fault of vendors if customers don’t use it.

    27. dealarchitect says:

      dahowlett dealarchitectthat I will grant you but Den, out of the box show me major vendors which are investing in OCR, mobile T&E, machine learning etc as applied to white collar processes? They all seem to point to partners for such functionality while out bragging each other on prettiness of UI

    28. says:

      dealarchitect dahowlett  and as if right on cue – three uploads in succession to Tripit failed the OCR…so I still have to enter details manually 😉

    29. rwang0 says:

      fwimmer Love the 3d bubble graphs.

    30. fwimmer says:

      rwang0 Thank you. And huge thanks for the kind words re: UX. That made my day! 🙂

    31. RobMiller31 says:

      i4_1 workers are people too! Can think of lots of example where poor UX gets in way of business performance…

    32. i4_1 says:

      RobMiller31 Oh, you don’t need to convince me. Some of those arguments fall into ‘not even wrong’ category.

    33. robertgillham says:

      I must admit I find myself slightly confused by your two posts on this subject – I think this is possibly because you start of taking the common technologist perspective that UX=UI, then sort of argue yourself out of it in this second part! As another person who has been involved in UX ‘before it was fashionable’, it’s frustrating to find the language of my trade being used in the dev/tech space to refer to nothing more than jazzy UI components. So it’s encouraging to find a more considered discussion happening in this space.
      One thing I was take you up on is confusing trends in UI design for best practice. Facebook is genuinely not considered a gold standard by serious UX practitioners.

      Nonetheless, I think we are basically in agreement, particularly that process complexity is not an excuse for poor user experience. I wrote something similar recently (from the UX practitioners perspective here):
      http://www.foolproof.co.uk/insight/blog/the-user-experience-of-enterprise-technology/

    34. says:

      robertgillham Robert, first off thanks for the link and the well thought out comment. I”m not sure we disagree much. 
      A couple items of note: this blog became two parts, so I didn’t really take a position until the second part. I started off with UI because there were several significant UI makeovers from enterprise vendors and that was a good way to frame the growing importance of UI. 
      As for Facebook not being the UX gold standard, I couldn’t agree with you more. But in the enterprise, there has been discussion about “Facebook for the enterprise” for several years now, and given the immense popularity of Facebook it is often equated with the consumerization of the enterprise. I felt it needed a couple of pokes, but I can see how a serious UX professional would have moved well beyond such assumptions.

    35. UsmanSheikh says:

      Hi Jon,

      Good blog post and discussion about the importance of consumer grade UI/UX for business use. I would like to draw your attention to a new kind of “enterprise” app – one that does not look inwards towards the data locked within the enterprise but outside the enterprise to find new sets of information and data to enable/empower business use cases. Please check out Sweetspot by XTRA IQ Inc. currently only available for iPad. This app is ready for you use right “out-of-the-box”. it aggregates, curates and analyzes business content from the Internet, Social Networks and Media (a much bigger though unstructured source of business content than any enterprise system) to enable sales and marketing to be able to find sweetspots for their business agendas.

      We keep forgetting that more business content is being created outside the firewalls of enterprise systems than within.

      This app is a good example of a new breed of business apps that:

      1)  is ready to use just like any consumer app (no expensive and time consuming integration required). Just download, sign up and you are up ad running!
      2) is lightweight and good for simple but important business use cases. I have been using this app and have been able to gather tons of excellent business intelligence on companies and industries. It has provided me with a competitive information advantage over others.
      3) has a “consumer grade” UI. It has borrowed from the Flipboard/Zite metaphors, It has Evernote like note taking capabilities. BUT business relevant
      4) has a Freemium model and very low cost to use!
      5) like SalesForce.com addresses the needs of the individual user first and not pay homage to the IT departments. 
      6) is SaaS?cloud based.

      Not every enterprise use case has to be complex and heavy. Check it out I think you will see that “consumerization of IT” is not a pipe dream. Cheers.

    36. says:

      UsmanSheikh some good points there, thx. Most of them jive with what I’ve written. I probably could have added more on the need for apps, not applications, – apps with quick iterations. And I would agree with you in some cases that apps pulling fresh external data can be a factor in the enterprise. As you say, those types of apps can typically be procured very easily, especially for workers at companies where they are free to download their own apps on certain devices. 

      However, having reviewed in excess of a hundred enterprise mobile apps, the real killer apps I have seen combine external data sources with back end enterprise data and connectivity. Being able to update inventory on the fly or add notes to a collaborative sales account and have the update wherever needed – that’s big for many companies. And it adds to the complexity. 

      Consumerization is without doubt a major enterprise trend, and I would argue a force for good in enterprise apps design. But many kinds of apps are needed, from complex and security to lightweight and easy to download. What is non-negotiable is that user experience must be a priority for those apps to succeed.