Following on from Neil Beagrie’s session on cost and benefit analysis of RDM we moved on to a Thematic session on the business case for RDM, led by presentations from June Finch (University of Manchester, MaDAM and MiSS Projects) and James Wilson (University of Oxford, Sudamih, VIDaaS and DaMaRO Projects).
June gave us an account of strategies in the MaDAM project which were used to prepare a business case for development to service, and which are being pursued in MiSS; whilst James focused on the effort to identify efficiency benefits and cost savings in Sudamih, necessary to the launch of VIDaaS).
From the point of view of this recently appointed MRD phase two programme manager I thought it was great to hear the experiences of two phase one project managers who have successfully steered their projects from start up (which is where most of the rest of the delegates were at) to follow on work on preparation for rollout and full service delivery.
Both presentations will be available online shortly so I don’t propose to recount them blow by blow. I think it is more useful pick out and paraphrase some of the key issues arising, beginning with MaDAM/MiSS:
– MaDAM was able to establish a Steering Group and a half: they had sufficient number, commitment and influence to give RDM real momentum at Manchester;
– it was difficult to establish the cost of existing RDM; they concentrated on quantifying what it would cost to do the job properly and establish a ‘service’ from which every researcher could expect a certain level of resource; the financial model was a driver in itself;
– there was a tension between their Steering Group’s desire to move away from indirect costing for RDM (presumably, towards including an RDM element as a direct cost in all funding applications) and the advice from their Transparent Approach to Costing (TRAC) team which was concerned about ‘double charging’;
– June showed an intriguing slide of a workbench-like RDM management system from MaDAM, but it wasn’t clear to me if this was to be rolled out as a part of MiSS, because the thrust there seem to be about a simpler storage offer;
– whilst the cost of new storage facilities was provided for in MiSS, the ongoing maintence and support was to be ‘absorbed’ into central IT services;
– RDM costs can be itemised in the new service, including support and storage. A lot of secondary work was done to establish a reliable per TB cost for storage at Manchester. Sadly, this was not revealed.
– James’ presentation began with some context: despite JISC’s previous efforts to model ICT evaluation, a strong grasp on costing remains elusive;
– nevertheless, it is important to establish total cost of ownership of new services and break these down into capital, recurrent and stepped costs;
– there will be startup costs to do with moving from ad-hoc practice to centralised services – inefficiency due to new learning new practices;
– the unaccounted for hidden costs of ad-hoc practice will be included in centralised services and may give the impression of increasing expense;
– advice: brainstorm with as many stakeholders as possible on benefit and don’t get too hung up on how to measure it: ‘soft’ or intangible benefits are known to yield hard results;
– concentrate on what can be measured: in Sudamih this mostly involved staff time savings (there was an interesting sequence of slides showing how investment in training was paid off in eight years);
– virtualising infrastructure (going to the cloud), first locally, then with a HE targeted provider such as Eduserve gives quantifiable savings, though to realise ‘per user’ benefit a significant critical mass is required
Near the end of her slides June noted ‘there is no silver bullet‘ and James began with quotes from John Maynard Keynes and Dan Remenyi reminding us of how difficult is to achieve a failsafe roadmap to success. Notwithstanding this caution, both presentations gave us plenty to reflect on, and, I think, a lot of encouragement that the RDM challenge can be addressed.
It may or may not be a surprise that contrasting presentations (institution wide service vs. specific database infrastructure) produced echoing themes: costing RDM is difficult; do what is practical; cost the future, not the past. They certainly stimulated a lot of debate afterward. The aforementioned tension between direct/indirect costs was explored at length, taking in the ‘hidden costs’ factor, and referring back to issues from the earlier session about what and how to evaluate.
@JISC_DataFlow also tweeted this event.
Footnote: I got the job of blogging this session because I wasn’t listening when June asked for volunteers and put my hand up to ask what the question was. (Tee-shirts with a mug on to my business address please). From this inauspicious start, factor in that I have the keyboard skills of only one of Douglas Adam’s typing monkeys and a stereotypical ability to do only one of, listen, speak and type, and you arrive at this account. If ever there were a need for clarification from the floor it applies to this post, so I hope the other participants will help me out with some comments.