Your organisation’s Collections Management System (CMS) is a tool; it’s important to keep that tool sharp and useful. A CMS is only as useful as the data that is in it. Bad data can lead to slow and inefficient searches, force duplication of effort, waste time with unnecessary data entry, and lead to unhelpful reports or web catalogues. How can you be sure that you’re not dulling your system with bad data?
First of all, be sure that your organisation has a standard for entering data. Whether it’s taken directly from an industry standard like SPECTRUM or ISAD(G), or is an in-house concoction, a standard allows you to better control what goes in to the system. Once you’ve got that standard, make sure it calls for data that meet these six characteristics.
Good data is …
More than anything else, good data must be consistent. This means using the same fields, the same words, and the same format all the time. Writing your own cataloguing manual or working from a published standard is a great help with this. Keeping data consistent makes searching easier as there are fewer terms to search for and places to search. It also makes it easier to use your CMS’s mass-editing tools to correct or alter the data, if necessary. The more consistent your data, the easier it is to fix any other problems that may exist in it.
Every piece of data in your system takes up digital storage space, must be maintained to keep it up to date, and costs time and energy to enter—time and energy that is desperately needed! Just like you should know why you are collecting an object, you should know why you’re recording a bit of data. Data shouldn’t be kept just because it always has been. Figure out why you are putting data in every field that you use in your CMS and what that field is doing for you. And remember, it’s okay to leave a field empty!
(3) Fit for purpose
In addition to justifying each field, be sure that each field is doing its job well. Fields in a CMS usually fill one of two roles:
Indexing: these fields are usually for keywords, dates, and people’s names. Being consistent is most important here. For example, even if an artist signed their name differently on three or four different paintings, the artist’s name should be always entered in the creator’s name field in the same form—this makes it easier to search on that field
Recording: these fields hold more ‘free’ text, such as a long description, the inscription on an object, or a museum wall label. Accuracy is more important in these fields, so if an artist signed their name differently on three or four different paintings, those different signatures should be recorded on each respective painting’s inscription field.
Who is going to be seeing each field? Remember, data doesn’t necessarily stay in the CMS—it can be sent out for a variety of purposes. A description field only meant to be seen in the CMS by your co-workers can be used much more liberally (‘need to come back and FIX THIS!’) than one you intend to display on your website or export to an exhibition catalogue or finding aid.
Your CMS is designed to reduce the amount of work you have to do. It centralises data to reduce the risk of duplicating effort (or worse, having contradictory records), keeps track of common terms and suggests them to you, checks ID numbers for uniqueness, and much more. But for these features to work, your data has to go ‘with the grain’ of your system, not against it. If you want to use standardised terms, for example, choose a field that enforces the use of approved terms (or customise a field to use them).
If you ever find yourself feeling like you are fighting against the CMS, take a step back. It might be time to check the manual, ask a peer, or book yourself onto a training course to see how that operation is meant to be done.
If you’ve been reading this list and thinking ‘there’s no way I can do all that!’Don’t panic! You need to balance the desire to do it all with achievable goals. You probably don’t have time to fill in 200 fields per object to these standards, but you can surely do five or ten fields. Figure out what are the most important fields and focus on those. You’ll find that standards like SPECTRUM also lay out minimums that you can draw from. It does you no good to set absurdly high standards if that keeps you from inputting a single record.
Good data is a lot of things. But most importantly, it is within your grasp—and along with it, a more useful and efficient CMS. Just take a step back and have a look at what you can do to make your data more consistent, justified, fit for purpose, audience-appropriate, and integrated with the CMS.
If you’d like some help, we offer a course on the tools available in your Axiell system that can aid in this, and more in-depth bespoke support is also available. Get in touch to find out more.