Coming soon... case study interviews describing real-world information system deployments.

Research organizations looking to deploy an ELN, LIMS or other data management system have maybe 4 or 5 strategic approaches they can follow: 1) They can evaluate, select and purchase a specialized (usually expensive) off-the-shelf commercial solution, designed to meet or approximately meet their needs out of the box. 2) They can opt to save money by deploying one of a relatively small number of free, open-source systems, and either use it it "as is", or attempt to modify it themselves to get it to better fit their needs. 3) They can purchase or use generic, not necessarily "science focussed", commercial software components, or use whatever technologies they have at hand to try to meet their data-management needs. Examples are organizations that use Excel for their sample tracking or inventory management needs, and some combination of tools such as evernote, sharepoint, dropbox, or simply a shared drive as their ELN or content management system. 4) If the organization is feeling brave, or if they feel that their needs are so specific that there simply is no available software solution that adequately matches their workflows, (and assuming they can find and afford the right design and engineering expertise), they can opt to build their own system from scratch. 5) Finally, if they need a comprehensive integrated system, or if their data management systems are so complex that they just can't figure out a solution for themselves, they may employ an integration company to build a system that draws on one or more of approaches 1-4 and puts them together to create a complete solution. This last solution would likely only be considered by large complex organizations with significant resources.

The relative merits of these approaches depend largely on the research organization's needs and priorities. If stringent FDA compliance is important, then approach number three probably isn't going to work. If the research organization is small and / or broke, then number five is not going to be an option, simply because it is going to be expensive.

In my professional interactions with various research labs and commercial science organizations I have seen every one of these strategies used with various degrees of success. I thought it might be useful to use this blog to interview some of the "buyers" and "sellers" of these various approaches in order to learn something about the experiences of those who have selected, deployed or used solutions based on each of these strategies. Hopefully, readers who visit this blog might find that the first-hand descriptions of the data management travails of other scientists and administrators will help them to decide which approach is best for them. Watch this blog for the transcripts of these interviews, which should gradually appear over time. If you are a bioinformatics specialist, ELN / LIMS vendor or Scientist who has used one of the above approaches successfully (or unsuccessfully), feel free to contact me if you think your experiences may be valuable to others.