Avoid this Common Six-pack of LIMS Implementation Mistakes
Be proactive, inclusive, and persistent and you will more quickly realize the productivity and quality benefits of your new LIMS.

Laboratory information management systems (LIMS) have been around for decades, and unfortunately so have failed implementations. In this article, we’ll review six common mistakes that lead to implementation failure and give you ways to avoid them. First let’s define implementation failure. The obvious metrics include running over budget, taking too long to implement, and not getting the features you expected. These issues can infect any kind of major software implementation from LIMS to WIMS, and from ERP to CRM. According to a recent survey of executives, more than two thirds of IT projects did not meet budget, schedule, or scope objectives, or were so bad the projects were abandoned.¹ Even a more generous research study targeting project managers indicated that at least one third of IT projects were failures.² If you’re in that bottom one third, you would rightfully be disappointed.

At the end of the day there is an even more important measure of failure or success—did you improve operations with the implementation? If you aren’t more productive or if you didn’t at least improve quality in some demonstrable way, the implementation failed. So how do you assure that your implementation project will be successful? Avoid these six common mistakes and your chances will increase.

1. Doing it on your own
Just because you can doesn’t mean you should. There are a variety of LIMS offered today — some are no doubt easier to implement than others. But it is highly unlikely that any system will perfectly match your business rules and lab practices. There will be configuration changes, tables to set up, nomenclature to modify and myriad other nuances you will want to incorporate.

And let’s not forget there is another constraint, you already have a day job. Implementing even the simplest system requires a fair amount of time dedicated solely to transitioning procedures, data, networks, and other systems.

Remedy: To assure a positive outcome, you need a thorough understanding of three very specialized disciplines, 1) lab operations, 2) LIMS functionality, and 3) IT infrastructure management. Do you have that kind of time? Do you have those kinds of skills? A trained implementer will have at least some experience in all three areas and can provide the quickest path to a successful implementation. They will also provide the necessary project management skills to develop a plan and meet the schedule. If possible, that implementer should come from the LIMS company. There may be good third party implementers for some products, but a company-based implementer is more likely to be intimately familiar with the latest system functionality.

2. Bypassing training
Training is often the first line item cut when you go over budget. It is tempting to try and get by without it, taking a learn-on-the-job approach. But even using the most intuitive system will be a new experience for lab workers. There will be new practices, new terminology and new work flows. You don’t want chemists fumbling around in a manual while they’re filing a beaker with hydrofluoric acid. The subsequent problems caused by under training tend to linger well beyond implementation. An inverse relationship develops between the amount of training and the number of calls to the LIMS Customer Support Help Line. This is not the most productive way to deliver training.

Remedy: Formal training provides step-by-step instructions and exposes people to the overall flow of the product to give context to work practices. Not everyone in the lab needs to be a power user but the team must have at least three levels competency. First, anyone who will use the system needs a basic understanding of core capabilities. Even with the most intuitive system, training on basic functions and a few tips and tricks will accelerate the improvements in productivity and reduce downtime. Second, you do need at least one or two power users who understand the full capability of the system. These folks can serve as in-house trainers, administrators, and troubleshooters, and will help other team members as the lab uses more sophisticated features. And third, separate from LIMS training, you need some form of IT support so that network and communications infrastructures are managed properly. A well trained team is an empowered team — one that will take ownership and responsibility to implement changes.

3. Counting too much on customization
Customization is seductive. As the famous slogan goes, you get to “have it your way.” Unfortunately, a LIMS isn’t a hamburger. When you customize a system you are actually creating a one-of-a-kind piece of art. Every change must go through additional integration testing to make sure you didn’t break something else. Truly custom development — customizing the code to specifically match your workflow — requires an engineering learning curve. You are potentially having code written that’s never been compiled before. This process often results in an open-ended implementation because one change frequently leads to another. That’s when you run smack into the first cousin of customization, named “scope creep.”

Remedy: Plan, plan, plan. Make sure you thoroughly document requirements before the system is implemented. Be deliberate about determining where the system needs to change in order to match lab practices, and be open-minded about when to change lab practices to follow the more efficient flow in a LIMS. And if there isn’t a good match between core functionality and your needs, don’t try to “customize away” the issues. When customization ends up driving implementation, you are essentially designing your own LIMS from scratch. Conversely, a highly configurable system typically has favorable cost and time factors. There is just less to do and it’s generally easier to make changes. And because a configurable system has a well-defined set of features and functions, it is easier to develop achievable schedules and budgets.

4. Not engaging the full team
You see your LIMS as an exciting new tool for everyone in the lab. Others in the lab may see it as threatening or just one more thing to learn. People get into a rhythm with their job and disrupting that rhythm disrupts their lives and the flow of work in the lab. Change is difficult. Change that you have no control over is even more difficult. Too often the larger team is shielded from the distraction of the implementation and not engaged in the process. But the success of the project ultimately depends on their acceptance and enthusiasm. You need their expertise, patience, and help. If that larger team is not involved in some way, they may resist the effort and may even undermine your chances for success. They will pick at the system, accentuating any weaknesses rather than looking for ways to create a smooth transition.

Remedy: Inform and involve. For a lot of folks, that may simply mean that you keep them up to date on the business objectives and status of the project. Everyone needs to understand what’s in it for them. For others, you should have them think about their work domains and consider how they can help the implementation. What prep work can they do? What potential issues do they see — and how could they solve those issues. You don’t want twenty people on the implementation committee, but keep the larger team engaged. Get their feedback on functions that will impact them and be sensitive to the fact that you are making changes that will affect their daily life.

5. Doing it in your spare time
Uh, what spare time? You don’t have spare time. That’s the reason you’re implementing a LIMS in the first place. A LIMS is one of the largest and most important investments for the lab and shortchanging the implementation by trying to squeeze it in between other priorities will have long-lasting effects. The project will be delayed every time the team gets busy and it will never be fully implemented. Schedules will be revised, corners will be cut and in the end, a poor implementation will lead to poor results.

Remedy: There isn’t a good or bad time to implement a new information management system. You’re probably never going to have a three month lull in your business. If you do, a LIMS probably isn’t your biggest worry. And the longer you wait, the longer it will take to realize the benefits of improved productivity. The key is to be proactive.

First, you have to make the implementation a priority — from the top down — and you have to allocate adequate resources. This must be seen as an imperative for the company. Second, there needs to be a point-person, someone who will take responsibility and who has decision making authority. Otherwise the project will meander. And third, work closely with your LIMS company project manager to build a realistic implementation plan. Make sure you understand the phases of the plan and agree on the deliverables. This is not just an investment of dollar: it is an investment of time. Like any good investment, you should expect it to payback more than it costs.

6. Losing focus
No LIMS is going to address one hundred percent of your needs right off the shelf. Even with a well-thought-out plan you will still run into the occasional problem that can’t be quickly solved. Maybe a manual process that can’t easily be automated. Or a function that doesn’t quite work exactly as you thought it did. Or a database translation issue. These are all important issues that ultimately must be addressed. But should you shut down the project or scrap the new LIMS because of these issues? Your business is too important to delay progress.

Remedy: Pick your battles. Keep perspective and don’t lose momentum over what could be just small bumps in the road. Yes, you should absolutely hold the LIMS vendor accountable to make sure you get what you paid for. But think big picture and prioritize the issues. Put the issues into four categories: 1) cannot go live without (i.e. there is no workaround), 2) serious issues that have short-term workarounds, 3) minor issues that will be an inconvenience until resolved, and 4) minor issues that will have no material impact. Only the items in the first category should hold up your implementation. Before moving forward however, all the items must have a plan of action. Don’t leave any issues unaddressed.

Conclusion & Keys to Success
Avoiding the six-pack of common mistakes does not guarantee success but falling victim to them will almost certainly ensure failure. Be proactive, inclusive, and persistent and you will more quickly realize the productivity and quality benefits of your new LIMS. The keys to success?
• Provide clear leadership & transparency
• Develop schedule, budget, objectives
• Communicate company-wide priority
• Don’t skimp on user training
• Dedicate resources

Notes
1. The Standish Group, Chaos Summary 2009, (April 2009)
2. Sauer, Gemino, Reich “The impact of size and volatility on IT project performance,” Communications of the ACM, Vol. 50, Number 11, (November 2007), p. 79-84.

John Albert is the Director of Implementation for Promium, LLC, (www.promium.com) the provider of Element DataSystem LIMS. Mr. Albert is a chemist and has been involved in lab operations and LIMS implementations for twenty years. He may be reached at editor@ScientificComputing.com