Processes the wrong way around

I was having a chat with one of the consultants who works in process and service improvement, he was telling me a new organization that he is working with, it’s a medium sized manufacturing enterprise based in the South East of England and it was a great illustration of deferred success from a service delivery standpoint. The conversation is summarized with anything that might highlight the company or my colleague removed, my questions are italic

So what’s going on?

I’m working with a client that’s got their IT the wrong way around, it’s quite weird.

What do you mean the wrong way around?

Well everything is done on the basis of the client knowing what it is that they want. The wrong way around.

I don’t get what you mean, what’s wrong with that?

It makes some certain assumptions around client technical knowledge, transparent communication and encourages behaviour opposite to delivery.  I’ll give you two specific examples. A new joiner requests access to a vendor application, the call goes through to the application team who check and permission the user and then closes the call. The user gets the call complete message and then tries the application and gets access denied, he therefore logs another call for access only to be told he has been granted access to the application code, but he has not been added to the group to access the application because that requires adding him to a group for which he will need to raise a new call. What group he asks, oh we don’t know that, ask application support is the reply.  The second and favourite example, user requests a new SharePoint site for their departmental collaboration, the user has to fill in the request form, the users needing access, the specifics needed to setup the site (name etc), he then gets an email asking how big the database will be, which database server he would like to host the database on, and what backup metrics and level of availability and scalability is requested.

But surely when you request application access that’s defined within the help system, end to end? You don’t want to be raising multiple differentiated calls, besides as a user you might not understand active directory or even proprietary application permission metrics and standards.  On both are their not default parameters? How would a user know database server names, how would they obtain that information indeed from an operational standards perspective do we want end users knowing server names?

This is why I think our service desk system and processes are in reverse, they are infrastructure, component based not service based.  We are systematically addressing the fundamentals then moving on to application or infrastructure as a service in very rudimental ways, firstly you ask for NAS storage tell us how much, we the service desk will handle the rest, we just need user names, the same for a SharePoint site,  a web site. We need to get the service desk, the projects on-boarding process seen as an enabler, with as much automation and simplification as possible. Put another way I want a large green button saying SharePoint site, when that’s clicked it asks development, production or non-standard offering, when you click development/production you enter the user names, press submit and we/an engineer handles the rest, from provisioning the site, requesting and applying groups (all using service desk processes and standard approvals).

Whether we engage internal teams or a cloud service provider for a new something as a service instance, the interface shows one team, one engineer responsible, the underlying activities, the knowledge is not inherent in logging a call. The user does not need to know groups, metrics for availability or specifications – we want them to say we want to do this… and we’ll handle the rest in a pre-defined package either standard or ‘non standard’, standard requests gaining more swift implementation.

Three key objectives:

Increase delivery – through effective task allocation, process optimization to make sure that when we log a call it’s going to the right teams to cover the request from an end to end standpoint, to ensure that what is delivered is a functional service, that has a name to it so that the user can contact them directly to get any questions or issues resolved.

Reduce shadow IT – we discovered people were either doing things ‘off book’ which meant using whatever technology was available from a cloud service, to a file server or desktop to reduce costs, disclosure and having to ‘speak to the helpdesk or IT’. One department had their Windows XP desktop running their SQL databases including SharePoint, their file shares and a few applications as it was easier to request a PC and do it themselves than debate policies and procedures with ‘IT’.

Transform the perception of IT, the way we work with our customers and continually improve our offerings around their requirements, either in house or adapted software and infrastructure as a service based on our standards and compliance issues.

In essence, we have a multi stream approach: ‘I call it User defined IT’

1. Reduce range of possible service desk calls available, from the 1289 combinations to two per ‘activity or action’ one is standard and goes through the normal procedures.  The second is non standard and is handed to a service desk manager or service resource co-ordinator to understand the requirements and manage the request manually through the teams. There should only be one type of application access call, two types of SharePoint request, one development which has no resilience or backups, and production which has both of these features.

2. Migrate common services to more agile infrastructure where the underlying infrastructure is abstracted from the end user so that sharing of resources across departments and services can be leveraged, this then also allows us to examine buying in some ‘non core’ or ‘approved core’ services.

3. Retrospectively tidy up security permissions, access rules and procedures as the systems are migrated to new infrastructure including SharePoint, NAS, Citrix etc, in effect verify that everyone currently with access should have access and ensure that the way to request access is established within the service desk portal, again so it works end to end.

This site uses Akismet to reduce spam. Learn how your comment data is processed.