DNA thought developing


Hello everyone

Just quickly picking up on the DNA thread. If you remember I was building quite a basic picture of what happens in a ‘typical week’ on the average IT service desk. It was really an exercise to see how people cope with the balance of planned/unplanned activities and how cloud computing may (will ) have an impact to such activities.

You may recall my wrap up picture of the Monday to Friday DNA week.

As depicted above I tried to ring fence low level tasks that may be replaced/amended/affected by a cloud strategy. It was desperately basic and lacked any real methodology or plan. However, I knew that and wanted to start right at the ground level on the KISS premise. For sure we can all make things complicated quickly!

Pushing this forward I stumbled upon ( or someone suggested ) an excellent paper from IBM entitled The Future of the IT Department.  Visit http://www.ibm.com/cloud-computing/us/en/

The argument IBM makes is that at some point the advent of cloud will demand a change to how the overall IT Department will function and deliver service. All this is music to my ears of course because it has got my juices flowing again on the risk of standing still in the IT Service aspect of IT delivery. Of course it is a tough one to know what to do because there are few experts out there and all the traditional models we had before are potentially redundant. I therefore like the IBM patented Component Business Model for the Business of IT. Yes IT service is a business and not just a function residing in the dungeon of a back office cost overhead.

So originally i had started looking at just the IT service fulfillment aspect from the lens of the traditional service desk but the IBM report has made me look a lot wider to how the cloud journey is going to change the IT department landscape.

Graphically they have a framework to establish the IT department structures and functions both BEFORE and AFTER

I wasn’t planning to throw any thesis at you in this blog because IBM have done a pretty dam good job IMHO but I would like to see if anyone has any views on the approach to break down an IT department’s functions and services into components as suggested by IBM. I suspect for many they do not and have a much more one dimensional view of the IT department i.e. Development, Support, Operations, Security, Architects. I think that to really get into the cloud impact assessment a much more three dimensional and deeper analysis is needed and the IBM Model is a great resource to look at how you can get there.

As always any comments for or against would be helpful.

Brummieruss.

Advertisements

14 thoughts on “DNA thought developing

  1. lars says:

    you make an interesting comment here . I am a consultant and i see lots of small companies jumping into the cloud without really preparing for how to deal with service issues to their end users. The local IT teams are left struggling and no one listens to them. It is good for me though. They ring me and I go in help them 🙂

    Lars.

  2. cloudism says:

    im not sure – do people really want to go this level. Cloud is cloud. We are all using it now without all this DNA, component stuff. Skeptical.

  3. babara says:

    I disagree. We are a UK retailer where service is vital to our network and we studied hard the impact of moving some systems to the cloud and the change in how we operate and support the systems and users. We swallowed the ITIL mantra about 8 years ago and have quite stringent controls in place. I like to think this has helped us with our own cloud journey because we have good processes in place and a team working well together

    Babara

  4. tom says:

    We are somewhere between your diagram and the IBM one. When we moved to SugarCRM and GoogleDocs we noticed a few tasks changed. We figured some of them but admit we hadn’t really tested the support model with the cloud companies. I would recommend that is a must. Users don’t care where the problem is they just want it fixed. Though we find end users now post tickets direct and we never find out what’s going on.

    TOM

  5. brummieruss says:

    What is interesting with the IBM report is the idea of certain functions no longer needed or a much lower scope. Take for example the suggestion that Service and Solution Architecture is no longer needed. For most the challenge is actually understanding both the current state architecture prior to a cloud decision and then the vision of what the architecture needs to look like after a cloud decision. Unlike the very large enterprises with dedicated enterprise architects and focused CIO office resources, most organisations have a more fluid approach to architecture planning. One dimensional architecture topologies are the norm with very limited alignment between the physical nuts and bolts of storage, networking and security with the impact on service delivery and business continuity. Take the example of an organization that ends up with a cloud strategy that looks like this;

    1. Business ‘influencing apps’ like CRM, collaboration come from a ‘high street’ public cloud provider
    2. Back office ‘supporting apps’ like email, payroll. HR come from a ‘dedicated’ private cloud provider
    3. Traditional ‘heavy duty engine room’ apps like finance, ERP, business process continue to be serviced from existing legacy internal private clouds

    So three very different infrastructure models all with their own architectural challenges, and an overworked internal IT department struggling with stitching together a service delivery model that copes with business change at a pace and a whole list of influences coming their way.

    For me the demand for a conscious and focused Service and Solution Architecture is massively important. In fact I would argue it is more important now than when all an IT department had to deal with was the internal private cloud.

    I find all this ‘stuff’ hugely interesting because the rules of the game are changing and whilst a lot of people look for signs or tipping points, the reality is that the move to the ‘new service architectures’ happens in a fragmented style. Organizations will fall into making cloud decisions as they experiment and test ways of improving the value they extract from IT. Cloud is the new news for many and learning from mistakes does not go away.

    In summary, approaching cloud without a Service and Solution Architecture running concurrently will be a big mistake, and if I could recommend anything, it would be the allocation of responsibility for this to someone to take overall control of the before, during and after. Keeping it simple and adhering to common sense Plan, Do, Check, Action just makes sense – doesnt it?

    I am going to post something soon about how I think classic internal IT departments can learn a lot from the way application developers approach the Agile Way to code development. I hope it will make sense once I have written it!

    have a great weekend

  6. Mike Emenalo says:

    Nice post starts the mind to wander

  7. brummieruss says:

    Mike my mind is always wandering 🙂 and here is another random ‘wander’ I had this morning.

    One thing I often think about with directory services was the traditional reliance on the device. You know the need to have a unique Netbios name for each machine, the effort of adding and removing workstations from directory services ( purging ) and the reliance on the workstation name for day to day management i.e distributing updates, AV updates, security scanning etc. Directory services relies heavily on the workstation as a way to manage policies and restrictions for the user. Organizations need a directory service ( for a considerable cost ) to define access priviledges and policies. With the cloud however and the move from the ‘glued device and user environment’ to a ‘loose decoupled experience’this reliance is becoming diluted. However, the more service procured in the cloud (and from different SaaS providers ) the more complexity is being introduced. Provisioning user access becomes a headache and one which the traditional service desk often does not realise how to cope yet still deliver a better service.

    Look around and you will see lots of people like Okta who provide a IAM as a service, where they act as the broker to all the SaaS apps an organization may consume to facilitate a single sign on experience by extending the internal directory services into cloud. Phew sounds easy doesnt it! It isnt but from the business perspective which surely is your starting point to work back from a user’s access is much more about the individual than the device. Cue a BYOD discussion!

    Snapshot questions to be thinking about in all this Service Desk impact analysis etc could include:

    * How much does managing the identity management aspect really cost? ( reasoning is that before you start moving workloads to the cloud it would be a pretty good idea to put some ‘cost’ on the table )

    * If you are using cloud how is the licensing model working in your favour or are you moving a problem where per seat licensing creates silos of underused licenses? ( so the problem just gets moved?)

    * How quickly can new accounts be setup and pulled down compared to how you do this today? Are the benefits of a cloud IAM service significantly better or actually only marginal? ( measuring how you do it today therefore is a must have KPI i suggest )

    * Is account lockout and password resets a big deal today? ( again is this measured and does it feature in any customer satisfaction index you may run? Solving this is a big big bonus if you can measure and fix with a cloud IAM Service )

    * Is the up take of new accounts so variable ( perhaps seasonal ) that there is no option than to move to a cloud based IAM solution to ensure peaks and troughs are managed cost effectively?

    As always the original debate spins off another thread? What is great news to me personally is that cloud again presents another angle of the ‘perpetual Rubik Cube’ view of IT infrastructure, Cloud and delivering the best IT service. And unless you are a genius your rubik cube ( like mine ) remains work in progress 🙂

    Brummie.

  8. lorenzo says:

    heh man you just love your rubik cubes 🙂

  9. gold price says:

    The typical enterprise today is challenged with modernizing an aging service desk, moving to a more customer-focused approach to end-user services, and leveraging current investments in IT assets. However, the decision to purchase an ITSM tool – even a SaaS-based one – may not account for factors that can lead to additional costs.

  10. Howard says:

    Brummie
    thought you would like to chew on this
    the last 3 months I have been working through a similar DNA type exercise
    here is a snapshot

    Interally we had 75 distinct projects – work packages
    I define a project as a piece of work that takes more than 4 hours to implement
    On average the planned time to deliver was 3.25 days – total 243.75 days – one full time person
    so lots of small BAU type projects – classic IT ops stuff
    My analysis revealed that 60% of these projects overran by 50% so that in reality we thought they would take 243.75 days days to complete but they took about 366 days

    What is interesting ( and relevant ) is that 30% of these small projects had a relationship with a cloud service somewhere – ie. setting up tablets, backing data to cloud services, testing access to new cloud apps and we aren’t being smart at doing them speedily

    I know that in a years time there will be more cloud work and that unless I do something with the processes ( as you are discussing ) the overrun on manpower will only increase

    Do I have Project management? I do but on small BAU projects I have to rely on the Ops guy managing their workloads ( or not )

    Does cloud improve this fulfillment piece? Not on its own.

    Am I suprised? Not really but you know what? The business assume because its cloud the time to deliver should magically accelerate and they would not be happy to be told that we are taking longer to deliver IT service.

    I have a lot to do – keeps me in a job 🙂

    H

  11. shazt says:

    We do Deming Cycle – Plan Do Check Action. It works. Everyone gets it.
    http://en.wikipedia.org/wiki/PDCA
    have a nice day 🙂

  12. francois says:

    I am interested in the DNA discussion Brummie. How do people measure end user experiences when discussing service fulfillment? I ask this because a lot is spoken about processes and streamlining etc but nothing about user experience. I am working as a PM on a project to do just this.

  13. ITHurts says:

    We use Nimsoft for our EUM. Mainly synthetic agents to monitor certain applications. Biggest hurdle for us was all the scripting we had to learn so its not a quick win solution. No way.

  14. Charles McEwan says:

    We do a DNA whiteboard with the managemt group via VC every Mpnday as part of our service improvement campaign
    It’s been running for 4weeks and already making a big difference
    Thanks for free advice Brummie
    I dropped you a private note to say thanks

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Advertisements
%d bloggers like this: