Enterprises are increasingly operating in a dynamically changing and fluid environment. They are constantly changing gears just to keep pace. CXOs are constantly looking for ways to overcome or create disruptions in a world becoming increasingly complex. Infosys Consulting Blog gathers a community of subject matter experts who are driving pragmatic conversations around that which is changing and that which needs to be rethought, redefined and redesigned for enterprises to achieve market-leading performance roadmaps.

« No Data? No Problem | Main | Re-visiting core competency to stay relevant in business »

What does vanilla taste like?

From order management and invoicing through to recruitment, enterprise resource planning (ERP) systems should open opportunities to create positive customer experiences.  But when businesses install "plain vanilla" versions of these business-critical systems, they risk negative customer experiences and dissatisfaction, leading to manual workarounds. A successful ERP implementation is one that follows a user-centred design process.

sourcing-programs.jpg

From order management and invoicing through to recruitment, enterprise resource planning (ERP) systems should open opportunities to create positive customer experiences.  But when businesses install "plain vanilla" versions of these business-critical systems, they risk negative customer experiences and dissatisfaction, leading to manual workarounds. A successful ERP implementation is one that follows a user-centred design process.   

A plain vanilla out-of-the-box system implementation has certain obvious attractions, starting with the fact that it requires no customization or additional installation. Surely it is cost effective to install a system that is tried, tested, and meets 95% of the business requirements?  With less time spent on requirements gathering, design, specification and user testing, the plain vanilla option ought to be successful for the project and the business.   

Not only does plain vanilla promise business stakeholders an improved system with tried and tested processes, it offers on-time and on-budget delivery; a compelling argument for executives looking to announce the successful transformation of their business. It all sounds too good to be true, and it is. 

Here's a real story of how it can all go wrong.  A successful global company (let's call them Firm X) was implementing an Oracle ERP system to replace its old and disparate platforms to improve efficiency and simplify processes for both its employees and customers. Add the fact that the program represented Firm X's largest IT spend in more than three years. 

The global head of Human Resources, along with his senior business and IT executives, decided to go with a plain vanilla installation. There were convincing arguments for this decision. A renowned management consultancy mapped out the business processes, assigned monetary value to each process, and concluded that the best value would come from a plain vanilla installation. An equally renowned system integrator felt confident that without customisation the programme stood an improved chance of delivering on time and on budget.

So, plain vanilla it was.  And, as promised, the system was indeed delivered on time and on budget.

But then the problems started. Customer complaints flooded in. Job hunters couldn't complete the online application form. Employees couldn't find customer data that was previously available.  Users abandoned the system and resorted to other channels, or back to old systems and workarounds.

The project sponsor called for emergency usability testing* to uncover what was going so wrong.  The testing found that the design of the interface was so unintuitive, complex and ugly that for some tasks users abandoned 100% of the time.

The testing results were far reaching.  The relationship between Firm X and the system integrator broke down over arguments about whether a problem was "defect or a change request".  Lawsuits were threatened. The global head of HR was eventually fired, and  the program was abandoned.

What the survivors of the Firm X catastrophe learned was that in addition to offering "on-time and on-budget, vanilla systems are also characterized by: 

  • Lack of utility. All functionality is delivered.  Even functions with low end-user and business value.  This has the effect of creating clutter on the screen, confusion between functions and errors through unnecessary complexity.
  • Poor user interface design.  Plain vanilla is ugly and difficult to use. Unintuitive field behaviour, validation messages and screen flow can break the customer experience and expose inconsistencies between positive experiences in other areas of a website.
  • Reliance on training.  At a certain point in the development process someone invariably notices how difficult the system is to use, but usually it's too late to change without massive impact on time and budget. The solution is typically to offer training, online demos, a help section or instructional copy.  But training as a remedial strategy to bad design never works. You simply can't "train out" bad design.

Together, these factors hurt the bottom line.  Indeed, recent research shows that says poorly designed software is costing U.S. business $60 billion annually

 

How do you avoid pitfalls of plain vanilla? Improve it.

ERP vendors may argue that companies' needs are so varied that a "one size fits all" approach can't work.  But there are broad areas of commonality that all companies share, such as online recruitment or catalogue design.  Breaking usability best practice is nothing to do with tailoring to meet the needs of clients. Rather, it's a cynical business model that relies on poor design to encourage consultancy revenue.  

Until ERP systems start getting the basics of design right, customers and end users must be included in the development process from the start in order to guide customisation and expose design issues early. 

Take a user-centred approach
In the case of Firm X, at no point (before it was too late) did anyone listen to or observe users using the old or proposed systems.  By rejecting all customisation, the programme lost the chance to simplify the plain-vanilla version based on what's important to users.

Too often, stakeholders and consultancies neglect to involve users from the outset when deciding the scope of what to customise.  Even if the programme is set on a plain vanilla path it is critical to observe and interview users interacting with the current systems and to find out early how they react to the plain vanilla version.

By applying this method a business will get real insights into what's important, uncover critical problems up front, and help guide strategic and tactical decisions about scope. Customisation does not necessarily jeopardise schedules and budget goals if you customise with end-users.  

A vanilla system, by definition, means end user needs are not considered early on.  Test the vanilla system with end-user first, and use the insight to drive customisation priorities.  

The bottom line: Focussing on what's important to customers always simplifies the solution and always lowers project costs.

(* The author conducted the usability labs)

Comments

Hi Rob

Lovely article. In no way trying to take your credit for the same, I would say that the article demonstrates why sometimes when an organization is undergoing an overall of its processes, it may also be a good idea to implement the strategic business rules using a customized BPMS platform rather than a plain vanilla process flow implementation.

Of course, an off-the-shelf implementation has its merits as well in standarization and lower cost of implementation but given the growing realization among companies that they need to fight on their strengths, a BPMS platform might give them that boost in incrementally designing a System that taps the companies strengths and helps to identify and deal with its shortcomings.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Please key in the two words you see in the box to validate your identity as an authentic user and reduce spam.

Subscribe to this blog's feed
 

Recommend on Google

Follow us on

Infosys on Twitter