I have what might be some silly questions so my apologies for that! It has to do with system migration and while I’ve been part of planning teams to work on this before, I had a different job title and responsibilities then so I’ve only worked with the data cleanup and staff training portions without exposure to the aspects below.
When a library is migrating to a different ILS/vendor, must the library contract with both the company they’re moving from, and the company they’re moving to, for data extraction and loading from the old ILS to the new one? Or have some libraries extracted the data themselves from the old ILS? Was this done via a module like data exchange (using a Sierra example)? If a library does these processes in-house, how do they deal with data mapping? Anything else you’d like to share?
We recently went through this, as we had a smaller, separate library system that was piggybacked onto our Polaris ILS leave and join the statewide consortium.
In our case, the library initially contracted with the statewide consortium. That contract defined all the data they would need extracted and how their new ILS manager wanted the data mapped (via spreadsheets). Since that smaller library system relied on our library’s ILS manager for all their ILS needs, they passed all of these spreadsheets onto us. We filled out the spreadsheets on their behalf to show what fields we use that would map to the fields that their new ILS would use, then we contacted with Innovative to do the data extraction. We then billed the library for the time spent doing their mapping spreadsheets as well as the cost that we paid Innovative to do the data extraction.
Hope this helps – please feel free to reach out if you would like further details!
Hello Ann,
Some of this might depend on which system you are moving to and from. I have heard some stories of Sierra libraries being charged large amounts of money to extract their own data out of the system if they were moving to a non Ex Libris solution for example. In that case depending on which vendor you are migrating to, that vendor may have ideas about how best to get that data out of Sierra using a combination of SQL, Data Exchange and APIs rather than pay a hefty fee.
I’ve just started working with my fourth library in less than a year to bring them from a stand alone ILS to part of our Polaris consortium. The “new” company will define what they need in terms of data. The “old” company will usually only provide data in one format and, if you ask for their help, will charge you and be jerks about it (technically you are breaking up with them, but it seems a lot of ILS companies don’t have any sense of “staying friends”).
Depending on the level of tech savvy in your library, you might be able to pull the data yourself and transform it into the appropriate format for the new company. With TLC and Sirsi, i’m being provided data files and I am the one that tranforms them into the format requested by Innovative according to their data migration guides. With Destiny and Apollo, we pulled the data ourselves and transformed it ourselves.
The one consistent format for all ILSes is the MRC file. So your bib records (and authority if you do those) will at least be pretty straightfoward. Everything else is kind of a toss up. I know Innovative WOULD do a lot more of the data massaging if I asked them to, I just prefer to be in control of the data (plus it helps when the library has a complaint that I can zero in on the possible issues and fixes).
Like Trevor, I’ve handled many different systems in the past and done data migrations from most of them. I also prefer to handle any data extracts myself when that is possible (it isn’t always). Even when I can’t handle the extracts myself, I’ve always provided the data mapping.
I’ve never contracted with the ILS vendor that I’m leaving but I have worked with outside contractors to deal with data from out-going systems in the past - both the extract itself and any necessary cleanup. Things I’ve contract help for:
DRA data extract and reformulation (simply because DRA did things differently)
going from non-MARC bibs and authorities to MARC
going from non-MARC serials holdings data to MARC for serials and prediction pattern
extraction outreach module reading history into a format that the new system could import (and working with the new system to actually import that data as they’d never done that before)
If you’re able to disclose which systems you’re migrating from and to, we may be able to provide more targeted advice. But it’s understandable if you can’t give those details publicly.