All the evidence points to enterprise search teams still being significantly under-resourced. Often only a single part-time manager is responsible for an application widely used throughout the organization every day. One solution is to set up a virtual Search Center of Excellence (SCE). This provides the appropriate level of support by having a number of people around the organization contribute their skills and experience.
Over the last few months I have been working with a global high-tech company with a European HQ but substantial business interests in the US and Asia that has decided to set up an SCE to support the change from the current enterprise search application to SharePoint 2013.
Roles and Responsibilities
Before I was engaged on the project the search team already existed as more of an informal community of practice than a formally-constituted team. The five team members came from records management, quality management, IT and two of the business units, one in Europe and the other in the US. Individually they were allocating roughly 30 percent of their time to search management. Two of the team members had a background in library and information science. Through the network of one of the team members a sixth person, based in Korea, was identified who could allocate only 10 percent of his time, but could act as a local contact for the business units in Asia.
The business case for the team was built around the seamless transition from the current search application to SP2013. The company makes extensive use of search in research and in business development so any interruption in service would be very damaging to ongoing business operations. The main responsibility for the team was to ensure that users throughout the company were fully aware of the switch in applications and the benefits that SP2013 would bring to searching a wide range of applications throughout the company.
We established a list of over 40 tasks that needed to be accomplished in the six months prior to launch. The team had the skills to complete the tasks but probably not enough time. It was decided to ask the company to set up a Steering Group for the search team which would be broadly representative of both IT and business departments. The primary role of the Steering Group would be to help the search team prioritize their time and if needed, take action to provide more resources on a short term basis to ensure that the launch was successful.
The Steering Group had five members to start, which was increased to six to bring in a senior manager from the US. Only one Group member worked in IT as the company recognized that the project had to be business-led and not just be a technology swop. Getting the time of senior managers for something that is perhaps tangential to their main business interests was eased by suggesting that there would only need to be a quarterly meeting of the Group unless there was an urgent need to make a resources decision. One of the benefits of setting up the Group was that it showed to staff and senior management that search was seen as an important application for the business.
In the project meetings we spent quite a lot of time discussing what the Key Performance Indicators would be for a successful launch. As part of the launch planning, the team had identified a range of stakeholders across the company in order to develop a communications strategy. It quickly became clear that working out what performance metrics to provide to each group of stakeholders was a core element of the communications strategy. A balance needed to be made between technical performance, query outcomes and overall satisfaction with the search application. The company had not previously assessed search satisfaction so the team decided to run a survey on the satisfaction with the current application to set a benchmark for the SP2013 implementation -- even though they knew the results would show a low level of satisfaction.
Despite the experience of the team, or perhaps because of it, there was a concern about the time required to complete some of the tasks. We decided to build in quite a considerable amount of "float" time to allow for the unknown unknowns rather than look smart with a timetable for launch down to the nearest day. It is much easier to reallocate spare time than find additional resources at short notice.
The launch of the SP2013 application will not take place until towards the end of this year so I will report back on progress in due course. For a case study on setting up an SCE take a look at a brief report from Accenture.