A couple of weeks after the Google exit comes the announcement that the Thomson Reuters Elite Enterprise business management application used very widely by law firms and law departments is also being closed down.  In both cases the organisations concerned will be faced with difficult migration exercises, hopefully not both at the same time! One of the commonalities is that both companies are ones that you would have thought would be in the business for ever, no matter how unrealistic that might be in reality. Certainly you were hoping that the issues of migration would be for your successor to deal with.

Another commonality is that the organisations will never have carried out a migration of this complexity before unless they have been through a SharePoint upgrade. This makes migration planning very difficult. I have seen many migration plans created in great detail in Microsoft Project before any serious analysis has been undertaken. The chances of achieving the promised deadline are probably zero because there are just too many unknowns and dependencies.

Of course database and application migration has always been an element of IT department skills but things are much more complicated with unstructured content and applications which support only part of an end-to-end process. This is not a routine Extract Transform Load (ETL) project, and these projects do not usually have to support the sort of complex visual user interfaces that are common on web-front end applications.

Among the challenges are

• How long will it be necessary to run the two systems concurrently while all the testing is carried out?
• Do you have the technical, content, process and project skills available in house? If not where are you going to get them from and how long will it take to train them?
• How dependent are you on the few key staff who understand both the current application and the possible alternatives you will be assessing?
• Will it be a like-for-like or will you take the opportunity to enhance functionality? If it is going to be an enhancement then arguably it is an implementation rather than a migration
• If you are going to transfer content items will this be the entire archive or just a selection? If a selection, who is going to make the decision and on what basis?
• What will be the impact on the user interfaces of both users and content providers? Content providers always get forgotten.
• What level of support will you get from vendors and system integrators especially if you are not the only client of theirs undertaking a migration?
• What other IT applications link into the about-to-vanish application? Will they need to be changed, or will proposed changes in them now need to be reconsidered?
• If the completion date forecast cannot be met what are the implications for clients and customers, not just for staff?

There is very little information out there about how to manage migrations.  David Hobbs seems to have written most of it and I’ve just launched a Search Migration Assessment service.  It’s not easy to go and talk to organisations who have done it before, as in the case of examples like Google and Elite no one will have done it before. I’m running a migration planning workshop at IntraTeam on 1 March that will cover content and search migrations. I’ve learned a lot of lessons the hard way over the last few years which I will share with you. Names of clients will not be revealed though!

This may all sound quite negative but I have seen very smart migrations in organisations that do not take their own expertise for granted and spend at least 30% of the anticipated migration schedule just planning around all sorts of scenarios.

Martin White