Saturday, March 22, 2014

A rose by any other name - or, is it HR or IT?

While reading the news of the day yesterday morning, I happened by this post by my trusted friend Prashanth Padmanabhan. 

The debate started well before we started speaking about "CLOUD" or SaaS, discussed many times, at meetings, global user groups and conventions, in private offices and conference halls.
Credit: - isn't it a magnificent art installation?

Indeed, should HR or IT take the lead when it come to HRIS projects? 
When the Cloud started being commercialized, the common understanding was that FINALLY the function could be solely in charge, with very limited or no input required by the IT/IS group - so there would be no need for expensive long term consultants, nor specialized IT headcount. Is that true? More specifically, can we hold it true when we are talking about critical systems as HR Core?

There are multiple moments in any project, each with different requirements in terms of skills, and different expected results. I would like to have a pragmatic look at each of these moments. 

  • Procurement Often at this stage, HR takes the lead (and it is a welcome change from the ERP times, when IT often imposed unwanted or too complex systems). This is the stage Prashanth mentions in his blog; and it should be looked as a partnership. While obviously HR has a huge vested interest in making the right choice, today HRIS is so central to a corporation's activities, authorizations and accesses, that IT must provide strong advice on evaluating and sizing "many aspects of business software such as total cost, security, integration, data migration and analytics", as stated in Prashanth's blog.
  • Stakeholder management Here is where sufficient IT agreement and support not only can save money, but can also help secure the right resources and budget to support the whole project.
  • Scoping, blueprinting and development Ahhh, here we are. Well, in my books, this is the core of the project - everything else is critical, but without this phase there simply would not be a project. The role of the function here is really important; in a Cloud project, there is quite little IT involvement - with input required on some critical pieces, such as ensuring that the interfaces will properly connect all the pieces (consultants can help, but internal IT has also the understanding of what ELSE the corporation is using) and much more. Most importantly, each delay in decision-making will for sure impact the rest of the project and the classical triad time-cost-quality; changes in scope (and scope creep) can feel easy to handle, but in this respect there is really no difference between traditional on premise projects and a cloud approach. Take your decisions, and stick to them, or your project will take much longer than planned. 
  • Testing and Go-live IMHO, here HR can dominate. HR knows best how the processes should be, what the legacy captured, what the new system should capture, what kind of data sets can be used, when data need to be scrambled for security/data privacy reasons, and how to evaluate the results. Of course, IT can support the planning and the set up. 
  • Post go-live support With development done under HR's responsability, with the assistance of consultants/service providers, there is a phase of knowledge transfer that will enable HR to take over this task. Even so, it is difficult to guess what kind of problems will bubble up (hopefully none...), and given the authorizations and interfaces criticality, IT readiness can be a hugely important factor for bottom line success. 
  • Long term support and governance HRIS requires large investments, and a new tool usually changes the way of working of all the corporation's employees, HR function, managers and employees through ESS/MSS and Self Service. In the years, IT has got used to set up extensive change management and governance approach to support a live system, sometimes in a very rigid way. The time required for a change to be done is often more on the approvals and workflows needed by change management, rather than actual development time (even in an On Premise SAP system, any consultant will confirm that changing a table's text takes minutes; but the transportation across the system architecture, the testing and the multiple approvals can transform these few minutes in weeks or months). In a Cloud tool, the development (or the text change, as in the above example) is also short - or comparatively even shorter; and it can be done directly by the HR admin. But does it means that we can bypass change management? how will the system be a stable tool if so? users (and administrators) in all regions and countries must be coordinated, and changes must go through an "approval path" (doesn't it sound like change management?) by all stakeholders. So in the end, IT again can support the set up and provide tools that will allow documenting all this process. 
I will echo again Prashanth in my conclusion. Yes, as "business critical applications move to the cloud, procurement of cloud software is certainly starting to look a bit more like procurement of on-premise software". Actually, I'd go beyond this. Not only procurement, but each phase of an HRIS project (and CRM would fall in the same situation) will not only benefit, but often be made possible by IT buy-in. 

Please tell me what I have forgotten or any additional comment :)