UCISA Support Services Conference 2013: Day Three

Day 3: Thursday 4th July

Thursday morning started with a pair of shorter plenaries looking at two recent service desk initiatives.

The first was a Bomgar deployment at UCA. Being distributed over five sites led UCA to feel that a remote support solution was an absolute must – this would also help with the Senior Officer who uses a Mac at home and needed assistance in setting it up with his printer – a familiar story! They had introduced Bomgar as a tool to fit this need, in order to drive up first-line-fix rates and reduce load on second-level / technical teams. Their previous tools were tied to particular platforms, and therefore the quality / level of service available depended on whether a user was running Windows, OS X, Linux, or something else. Bomgar were also part of the supplier exhibition, and were promoting their recently developed direct integration with the Cherwell service desk tool.

This was followed by a review of the recent Service Desk Initiative (SDI) certification undertaken at the University of Leeds. They are a 14 FTE service desk, with distributed ITSS, handling around 80,000 calls each year. They had been trying to move from “catch and dispatch” or “break/fix” to a customer focussed model of working and were wanting an independent assessment of their operation. SDI was chosen for this. An initial 2-day assessment unpicked everything they did, and took a deep look at their documentation, processes, training, etc. The score wasn’t where Leeds had hoped, but this initial assessment is followed by a series of recommendations to help you acheive your target rating. After 6 months of hard work, Leeds made it through their certification assessment at their desired target level. This is the kind of work that a Service Desk Manager does, and it was hard. Leeds are now hoping to build on their score and go back for recertification at a higher level in future.

During the break I met with the head of the desktop PC team at Royal Holloway. They manage around 900 student PCs and a further 1200 staff machines – so a similar size of user base to Oxford and picking up similar deployment scenarios. Their organisation comprises 5 service desk staff and 5 desktop technicians, using WSUS/SCCM for management. Interestingly, student calls go to the service desk, and all staff calls go straight through to the technicians. The current push in this team is for collaboration tools – providing collaboration from the desktop using Microsoft Link and Yammer. This will be a move away from a plethora of self-selected tools currently in use.

The final session of the conference was a look at Social Media as a communication channel, and brought together many ideas that were raised in other talks. Six main types of SM were identified: blogs, networks (groups based on people’s affinities with each other), content communities (groups based on content type such as YouTube), bookmarking services, micro-blogging like Twitter, and wikis. Internal SM platforms – used for employees to communicate with each other – tend to get low take-up because these don’t build on existing networks / profiles / communities but require people to re-invest in a less established arena, often with little idea of the likely benefits. Working in SM requires us to think about “the 4 C’s”:

  1. Creation (of content) – what is relevant, existing, interesting?
  2. Context – how do people find our content (search engines, sharing, discovery)?
  3. Conversation – listening, interpreting, and responding to other SM publishers, and the reaction to our own content;
  4. Conversion – why are we doing this at all? We must have a goal, which will often be based on someone doing something (buy our product, solve a problem for themselves, find and use a new service, …)

There are lots of stories about SM faux-pas. Customer Service often features in SM scare stories. A video showing how to open a Kryptonite cycle lock using the cap of a bic biro stirred up a lot of chatter, and Kryptonite responded after about 10 days (with a refund/upgrade for customers) – however this incident is estimated to have cost them around $18m.

What is also important though is that the preferred SM platforms change rapidly. Even now, as businesses are getting to grips with blogging, Twitter, and Facebook, students coming up from school are using SnapChat and the Vine. In a few years these two will also be outmoded. Our approach needs to be to develop an overarching strategy that remains applicable as the platforms change, and recognise that flexibility can be powerful here.

The conference finished with a final chance to meet colleagues, and the starting of a journey round UK HEIs of a conference trophy, to be passed from person to person on a tour around the UK, and arriving back at next year’s conference with a series of photos of interesting things learned at each institution.


UCISA Support Services Conference 2013: Day Two

Day 2: Wednesday 3rd July

Breakfast, and time for the first meeting of the day, with colleagues from Plymouth and Liverpool John Moores. Do we have security training for all our staff (IT or otherwise)? At Oxford we have policies, a toolkit, lots of guidance, and some systems, but the only routine training with periodic “refreshers” is on DSE assessment and recruitment (although some security training is on the way as I type). The pattern seems good – initial training, followed by periodic updates (frequency as appropriate), via a self-service interface with some means of checking that the user has taken on board key aspects of the learning. At LJM all IT staff go through a similar programme for DPA, anti-bribery, and diversity training. Sounds like it has potential for Oxford, and the quid pro quo is that our ISBP toolkit is of interest to Plymouth and LJM.

The first plenary was from Druva, who have provided a “dropbox alternative that meets corporate data storage and security requirements” to the Economics Dept and Business School at Warwick. I spoke to Druva the day before, and the solution sounded interesting – compliance (policy) driven storage with refresh and commit to/from a local copy from multiple clients. The talk also had some interesting points – our users are shouting for “style”, “freedom”, “productivity”, and “convenience”, whilst our employers are demanding “managed risk”, “privacy”, “security” and “protection” – the ever present availability vs. security tension. Unfortunately the rest of the talk didn’t reassure me. “Druva is NOT a backup company” we were told, but then “…and we can deliver all of these core features because, at its heart, Druva is a backup company”. Later on “We configured it to use port 22 as we already had a hole in the firewall and didn’t want to open any more holes up for obvious security reasons”. As configured at Warwick, users can restore to any 12hr snapshot in the last two weeks, from any device (and there is a web interface for retrieval too). One other note, from Warwick’s Economics Dept, was that they had abandoned whole disk encryption due to (a) direct cost, (b) support and productivity cost through lost passwords, and (c) impact on imaging and desktop management.

Plenary 2, and Steven Beavis (Cherwell) talked about measuring customer satisfaction. His key point was that at the end of the day, that’s what it is all about. He suggested that customer satisfaction follows primarily from two areas – perception of value, and ease of access to desired services. Metrics should be balanced across four main areas: Productivity, Efficiency, Satisfaction, and (as a result) Satisfaction. Deriving metrics is a cycle – you set Goals, define Critical Success Factors that reflect acheivement of the Goal, set Key Performance Indicators to tell you when the CSFs are hit, record Metrics which can be used to derive KPIs, and generate feedback as a result, which then helps to set next year’s Goals. Users are increasingly requesting a focus on satisfaction – “I want it fixed, now” – rather than operational targets (we had 10,000 calls last month). Further, reports should be actionable – there is no point in reporting a figure unless it leads to actions for improvement.

This led into a lightning talk about surveying. Lots of people had anecdotal evidence of high levels of apathy – low response rates, mid-range scores, etc. A few success stories were related though. Noel Bruton’s “Random 5” – 5 callbacks per week to users who had tickets closed to gauge their satisfaction in 3 questions or so – this had worked well for one person. Another got a 30% response rate by sending all callers a “Less-than-one-minute” survey. The idea of an email with (two or) three HTML buttons – Poor, OK, Good – to all callers won some favour. One University IT department had sent staff out with bright t-shirts and a list of 12 (yes or no style) questions; these staff then took up post near doors to campus cafe’s etc and asked just one question of anyone walking through – of course most people gave up the answer before they realised that they were being surveyed, and a very high rate of data collection was acheived.

The morning closed with a talk from Edinburgh University about UniDesk. Edinburgh, St Andrews, and the University of Abertay had all been looking to buy new ITSM tools, and thought that doing this together would get better leverage. This led to thinking about sharing the cost of developing ITSM processes, and then the whole hog – a shared service desk. TOPdesk came in as a fourth partner, bringing their ITSM tool and commercial knowledge. From kick-off in June 2008 the whole thing was done in c. 30 months (Nov 2010).

The initial shared service had Incident Management and Request Fulfilment, and extension to incorporate Problem Management, Change & Release Management, and Configuration Management (or at least a CMDB) followed. Sheffield Hallam University have taken this up for their service desk.

UniDesk’s life had not been without lessons to be learned – and many of these seemed to be good advice for our own service desk project. Five things came though:

  1. Keep it simple – no customisations for the initial partnerships makes it easy to maintain, avoids politics, keeps costs down, and makes it much simpler when potential new partners come along. You need to decide not to overengineer it from the outset!
  2. This isn’t going to save the bank – although efficiencies may lead to small savings, these are not generally substantial, especially for the initial group who set it up.
  3. A simple business model is critical. UniDesk agreed a cost per annum, based on JISC bandings of instituation size – but then usage was unlimited (i.e. cost is absolutely defined for the service period ahead – “fill y’r boots!”).
  4. Trust is very important. Each insitution had to make compromises, adopt aspects of the others, give up customisations of their own. This can only be done successfully if you trust that everyone is in this for mutual benefit.
  5. You need to be rigorous in accounting for costs (e.g. staff time) in the shared service – as resources come from the partner organisations, but people will want to question how much the are being charged (the transparency aids building of trust).

Edinburgh are also using the shared service desk for their Finance and Registry teams. They are getting on well with using the tool, but don’t have the ITIL knowledge to “get” the processes – especially the two-part closure (resolve first, then close with data cleansing).

Innovative Communications was the topic of conversation before lunch. Most people seemed to be focussing on how to engage with students (“lose the suit, grab a hoodie” was the key to success here). Ideas that broke out of the usual molds included:

  • Poster promenades – IT and Library advertising their wares in departments (cake made these events popular);
  • Open Days – a sort of IT conference for academics and researchers – a tour of the data centre had proven particularly alluring!
  • A roadshow, going round the University to show off new/exciting developments. Short talks were recorded and delivered by video to save key staff from spending loads of time on this;
  • The use of 90-second videos to showcase services and projects had been well received. Several Universities used video as a core part of their communications, and had their own YouTube channel;
  • One University had a robot at their freshers fair – you could ask the robot a question and somehow it would try to provide an answer.

Over lunch I met up with two Relationship Managers from Hull University. This is not part of the student welfare scheme, but the bit of IT that links out across the University to do Business Relationship Management (a bit like Internal Account Management). Their role is to speak to departments about current issues, wishlists, strategies, plans etc, arrange for specialist staff to be available for suitable discussions, and advise on new developments / IT strategy. All engagements / contacts are recorded and reviewed, and the information collected is made widely available across the IT department. `The role is tightly integrated with Communications – although Hull are yet to get staff for this. More than one person is needed for this, although with BRM activities spread over several staff, one person could coordinate several relationships. It was seen as key to success that relationships are trunked – i.e. one person oversees / handles relationships with a group of related customers. This helps to join up the customer needs, and fosters sharing / peer support. In terms of internal links and dependencies needed to deliver in this area, Relationship Management falls cleanly into the Customer Services area of IT, but depends a lot on Application and Web Development as these are the areas where most issues seem to crop up.

The afternoon kicked off with a discussion of Herriot Watts’ reorganisation and merger of IT and Libraries. One of their larger issues was a 10 minute walk between the library and the data centre, and several people who had merged library and IT help desks a few years ago are now separating these functions out again – it seems that students want library help from libraries and IT help from IT.

Next we heard from Dan Batchelor (University of Wolverhampton) who is the outgoing president(?) of the Student Union there. The union had been failing, and he led them to ditch the bar/club/table-footy culture and become a support body for students. Part of this saw them become “the Student Voice” in IT decision making, and they really did seem to have great engagement through formal and informal channels at both individual and organisational levels. This meant that strengths established in one year were not lost when the executive was re-elected for the next. His story was very persuasive and I, amongst others, plan to investigate the potential of our own unions (OUSU in Oxford’s case) to help link with students.

The last formal session of the day was a talk about surviving in changing and challenging times. Paul McGee was certainly an animated and engaging speaker, and many people will doubtless remember some of his sound bites: “The future does not belong to the strongest, but to those most able to adapt (Darwin)”, “Shut Up, Move On”, the receptionist whose name card read “Director of First Impressions”, and the fact that “change makes us uncomfortable”.

Thus ended the second day, giving way to the second evening, the conference dinner, the after dinner talks, and the discussions of service desk and managed desktop issues into the wee hours.

UCISA Support Services Conference 2013: Day One

This is the (now) annual UCISA SSC, attended by around 150 staff working in IT support from across UK HE. This year’s conference ran from Tuesday 2nd July – Thursday 4th July, at the John McIntyre Conference Centre, Edinburgh.

Day 1, Tuesday 2nd July:

Arrivals and registration led onto meeting up with colleagues (new and old) from other institutions. Common themes soon emerged, with discussion around service desk KPIs and metrics, questions around whether AV support is a service desk responsibility or something for a specialist team, and developments (mainly around the use of virtualisation) in desktop provision. Data security concerns also featured, but with the focus varying between confidentiality (e.g. in the case of lost / stolen laptops) and availability (e.g. in the case of device failure when the only copy was stored locally).

I had an interesting chat with Sheffield University’s Service Improvement Coordinator. At Sheffield they felt that ITIL’s Problem Management process only addressed part of a broader picture of (reactive and proactive) feedback on service quality. Their SIC role encompasses a set of responsibilities covering customer satisfaction and feedback, service reviews (including interviews), Incident patterns, and more. This role also overlaps with ITIL’s Business Relationship Management and Availability Management, but doesn’t have any resources or authority to make changes – this happens through influence on service strategy boards. Sheffield has c. 165 services in their Service Catalog.

Ian MacDonald (Co-operative Group IT at Manchester) presented a the subject of Assessing and Benchmarking to Drive Continuous Service Improvement. Some of the key points of this talk were:

  • Real value is derived when assessment and benchmarking is integrated into a strategic intent to continuously improve capabilities and deliver improved services;
  • It needs to be ongoing.

A Jack Welch quotation (“Face reality as it is, not as it was or how you wish it would be“) captured the idea that we might have the view that we’re doing a good job, but our users might have quite a different perception – we can’t get awat with just believing that we are “good” at what we do – we need to get some external references on this.

An interesting model was discussed, where Value for Money (seen by the user/customer) is generated through a combination of costs and value. Costs arise from the hardware, software, premises, and staff used to deliver services, whilst Value is derived from the products, services, people, and image that we offer. Importantly then, Value is perceived – and can be influenced, whereas Costs are tangible – and cannot be influenced (they are what they are).

In terms of understanding how we are doing then, there are some options. Self-assessment, available through several free structured questionnaires, can give us a score that can be compared against an industry norm. Certification provides independent assurance that a certain standard has been met or exceeded. Benchmarking enables measurement of the time, cost, and quality of our activities, which can then be compared against best practice or peer results (e.g. from the same industry sector) – this provides the best understanding around the costs of delivering IT. Finally, Awards are another way in which excellence can be recognised – and has a strong link back to the image aspect of Value.

The next plenary was a group discussion of support models in different HEIs. Although there was a lot of variation in the details, broadly speaking most people seemed to have a central IT function providing commodity services, and some form of devolved local/specialised IT function in departments / schools / colleges. Those who represented central IT took a general view that they wanted more centralisation, but few had found ways to overcome the objections (or rejection of the idea) by local IT. Oxford’s position of recognising distributed IT as a key asset, and wishing to strengthen the support interactions between local and central IT was fairly unique.

Our first business showcase of the conference was a joint presentation by Cherwell and the University of Wolverhampton. The UoW undertook a project to replace its service desk toolset. The project timeline was something like:

  • Early 2011: desire recognised, project kicked-off
  • Oct 2011: vendor demos
  • Dec 2011: PQQ responses and tender process
  • Mar 2012: 3 suppliers shortlisted, scenario-based proof-of-concept sessions with vendors (13 scenarios used, focussing on previous or predicted situations), vendors varied in approach and preparedness
  • Apr 2012: Order placed with Cherwell
  • Jul 2012: Implementation finalised, dual-systems run in parallel
  • Aug 2012: Go live

The talk looked at how the relationship developed between Cherwell and UoW, and touched on a number of lessons learned during the project. The key messages were that the full extent of project impact needs to be considered – training for all staff and key users, awareness of the change across the broader university, and not expecting perfection after a single run through.

Next we heard from Manchester Metropolitan University, who have been undergoing a programme of cultural change. This had some interesting overlap with Oxford’s central IT reorganisation. They had faced the question of “Are we there yet” – and decided that with organisational change you can rarely say “yes” as the endpoint is really hard to define or measure. They had used Myers-Briggs personality assessments, 360 degree feedback, changes in language (e.g. the area where infrastructure work is now called “the office” rather than “the bunker”), and they have introduced social events, recognitionschemes, and “Make A Difference (MAD) Day” where managers all reported on something acheived by their team in the last few months that had made a difference to someone or some part of the organisation.

The final event of the day was a series of 3 PechaKucha (20×20) talks. In one (John Grannan, Leeds) looked their OneIT Transformation Programme – another organisational change! He described the goal of this using a phrase that would be persuasive anywhere: “Showing users that IT is something they want, not something they have to fight with”. One of their focusses has been on Service Definition – providing a clear list for users so they know what is available, who it is provided by, and how/where to get it. They are using ITIL and CSIP, but have hit the issue that in terms of process maturity models, the business needs to match IT in order that the “more mature” conversations can take place. The final session was given by Peter Tinson (UCISA) on the subject of Leadership. He made several interesting points: (1) everyone working in IT is a leader – which we can only do well if we are respected (for doing a good job), professional, understanding of the business’ needs, and advocates for IT. He noted that decision making is often based only marginally (2%) on fact, and heavily (98%) on opinion, which is based on perceptions, interpretations, beliefs, and aspirations.If we take the time to listen to our stakeholders, talk to them in terms they understand, and are consistent in our message, then we will become valued and trusted partners.