The 10 most common mistakes in implementing a LIMS (PDF 595 KB)

10 Most Common Mistakes in Implementing LIMS - and how to avoid them Quality Systems International web : www.qsiuk.com email : sales@qsiuk.com Introduction How do you prevent your LIMS project from being disdainfully referred to as “Failed Project”? How do you stop yourself from falling into the unsuccessful project trap, wasting time and money and incurring bad performance reviews from above? Surely the ultimate goal of a LIMS project is to successfully “go live” on time and in budget, whilst achieving user acceptance at all levels of the business. By avoiding the following 10 most common mistakes made in implementing a LIMS you will be well on your way to gaining the success you are looking for from your LIMS project. Mistake Number One – Unrealistic Goals The biggest mistake made when implementing a LIMS is to set the scope of the project to wide and then to compound this by setting tight or unachievable timescales. So how should you set the goals? When your company decides to purchase a LIMS, there will be a number of business reasons that justify that decision. Such as; faster release of product, reduction of paperwork, adherence to regulatory controls, laboratory automation, systems integration, etc. These justifications will often apply to only certain parts of the operation, so a review is required of the reasons, to assess the business benefit and the impact on the areas where the LIMS is to be used. The review should focus on the areas which will achieve the most impact in terms of overall benefit to the business and operational benefit to the users in those areas. The decision on the scope of the initial project should be made on the results of the review, in conjunction with your expected timescales. It is crucial to limit the scope to those areas which will achieve the highest return on investment and user satisfaction. The benefit of this approach is that when the LIMS has successfully gone live in these initial areas, top level management see immediate benefits and subsequently will endorse further rollout of the LIMS into other areas of the business. A final word about your timescales, a LIMS on average takes 6 months to implement, anything less than that in most circumstances is unrealistic. Also it is not only your resources involved in the project, your chosen vendor has to provide resource, the quicker you communicate your timescales to the vendor, it is more likely they can allocate the required resource or advise on availability. Quality Systems International web : www.qsiuk.com email : sales@qsiuk.com Mistake Number Two – Bad Project Team Selection It is very common for project teams to be too large or ill chosen, which leads to every decision being debated at length until agreement is made. The ultimate result is waste of valuable project time, leading to project overrun and where there are wrongly skilled project members involved, badly defined requirements and or lack of project control. So how should you select the project team? It is vitally important to firstly select the project leader, this person should display strong leadership qualities, ability to delegate, and above all be a good listener. It is no good having a project leader that is a control freak, who dominates the meetings without allowing proper debate. The other members should comprise of the following people. A person who has an overall knowledge of all areas: involved in the project. Where there is a strategic key area, a person who has broad understanding of that area. A person who has full knowledge of all the process level details. A member of the IT department. It is good practice to limit the size of the project team to six or less Quality Systems International web : www.qsiuk.com email : sales@qsiuk.com Mistake Number Three – Inadequate Human Resource One of the most significant mistakes made that causes overrun on project timelines or even complete project failure is companies not appreciating how much human resource is required throughout the project lifecycle. How do you assess your human resource needs? Each companies resource needs are different, however if you follow these guidelines, you will then be able to understand and plan your human resource. There are a number of areas and items that will require resource whilst implementing your LIMS, these are as follows. The project team. Short term access to key employees during site survey. Site survey pre-planning. Site survey. Additional hardware install and test. System integration. Instrument interfacing. System testing. Staff Training. Data load and validation. System validation ( a requirement for pharmaceutical companies). Parallel running. Once you have identified the areas that apply to your LIMS project, you will then need to plan the resource for each area. Be realistic, it is better to over estimate than under resource, if in doubt about any areas, talk to your LIMS vendor, they will have a great deal of experience in implementing LIMS and can provide assistance in planning your resource. Quality Systems International web : www.qsiuk.com email : sales@qsiuk.com Mistake Number Four– Lack of IT Resource IT resource is nearly always required at sometime during your LIMS project, a lack of planning often leads to that resource not being available at the time required. How do you plan for your IT resource ? Always get your available IT resource involved or at least notified at the beginning of your project, if you get buy in from your IT department, they will be more likely to provide the resource when required. Always remember there is a huge demand from many parts of your business for IT resource, so do not delay, notification. It may be the case that your IT is outsourced, as you will be dealing with a third party company, be even more diligent in your communications with them, so as to avoid any confusion about timelines. If you have no IT resource within your company, you could hire a third party company to provide the required resource. However a better solution would be to pick a LIMS vendor who would provide the IT technical experience required. Quality Systems International web : www.qsiuk.com email : sales@qsiuk.com Mistake Number Five – Inadequate Pre-Survey Planning Many companies recognize the need to perform a site survey, however many fail to prepare in advance for the survey. This leads to extended timelines and even worse inaccurate survey results. How do you prepare for the site survey? Firstly analyze your current reports to assess which are important to the business and are essential to duplicate in your new LIMS. Also establish if there any reporting shortfalls, at all times be critical, question the need for reports, most companies have reports that are not really used (inherited reports from older systems). The essential reports help to define what data is required within your LIMS, this is very important Next review your sample and or batch workflow, a walk through your organization starting at where the samples originate through to final completion of the samples is a highly enlightening process. Create a workflow chart, then add the business logic to each relevant point. This would be items such as what happens if a sample fails, do you, retest, resample, etc. This workflow chart is supposed to be a skeleton not a highly detailed document, the purpose of the site survey is to put flesh on to the skeleton. Finally, do you need any interfaces between your LIMS and third party products and or instruments. If instrument interfaces are required make a list of the instruments, detailing make and model, check for availability of the manuals, if not available list manufacturers contact details. If you wish to interface to third party products such as MRP and ERP systems, notate what are the prime functions of each interface and what data is to move or be shared by the applications. Ensure that whoever is the expert for the third party product is available at some time during the site survey. Quality Systems International web : www.qsiuk.com email : sales@qsiuk.com Mistake Number Six – No or Badly Performed Site Survey To not perform a site survey is a guarantee to have a troubled or failed LIMS implementation. A poorly executed survey will result in a specification that does not reflect your LIMS requirements, hence the delivered LIMS will not provide what the business really needs. How to avoid this costly mistake? Always have a site survey performed, you may feel that it is not warranted as the intended LIMS appears to meet your requirements, however you must understand that lack of communication between you and your chosen vendor is the biggest cause of LIMS failures. The site survey and the resultant specification document forms the blueprint for your project and if performed correctly removes all misunderstandings between yourself and the vendor. If your vendor does not recommend that you have a site survey, find a new vendor. The reason for poorly performed surveys is caused by the choice of who should lead the survey and write the specification document. It is rare for a survey and document created internally to actually succeed in defining the requirements, this is due to the fact that internal staff just do not have a wide enough experience in implementing LIMS or in depth knowledge of the chosen LIMS. It is also the case that third party consultants are too general in their IT experience to comprehend the requirements of laboratories, and as such the resultant specification is not worth the cost. Your chosen vendor is the right person to lead the survey and write the specification, in practice the vendor will have implemented LIMS into a wide range of industries beside yours and as such will be able to suggest different ways of utilizing the software for each requirement being discussed. The vendors experience will always exceed yours and any third party consultants, also there is another major benefit in using the vendor, It provides you with the opportunity early on in the project to work with the vendor and to assess their true competence, if unacceptable, you can then find an alternative vendor. Quality Systems International web : www.qsiuk.com email : sales@qsiuk.com Mistake Number Seven – Lack of Data Load Validation It is obvious that data has to be loaded into the chosen LIMS, however many companies overlook validation of that data to ensure the accuracy of the data load. This issue is just as important whether the data load is manual or electronic. Can you put a cost, on product being released to customers out of specification, due to the specifications being entered incorrectly? How to perform data validation? If your data load is manual, then dependent on the capabilities of the LIMS you should be able to set access and security that makes it a requirement that data entered by one person has to be approved by someone else, before that data can be used by the LIMS. Where the data load is electronic from an existing LIMS or other electronic sources then the volume of data will be larger as it will most likely to contain historic samples and or batches. It would still however make sense to manually check each key data record, such as specifications and methods, as these can cause extreme problems if incorrect. The historic samples and batches may contain very large volumes of records; in these cases due to the volume it is best to randomly sample the records for correctness. The more records the higher number of random checks is required. Quality Systems International web : www.qsiuk.com email : sales@qsiuk.com Mistake Number Eight – Insufficient System Test Many companies make the mistake of not allowing enough time to test the LIMS software or in some cases don’t even bother. The problem with this is if the software is released to the users and they start to find issues, then their confidence in the new LIMS diminishes with each new issue. Better to find these in advance How to system test? Data has to be input into the LIMS so as to be able to perform a complete sample and batch life cycle. This initial test data has to be validated to check for accuracy. It is good practice to write simple scripts that are followed when testing; this will guide the testers through the various workflows. Also if any issues are found, once rectified, the scripts can then be used again to provide repeatability. The testing should be carried out until such time as the software conforms to the agreed specification. This testing regime is important even if you have purchased an off the shelf solution, as there can be data related bug or issues. Mistake Number Nine – Inadequate Training Proper user training should be provided to all levels of staff that will be using the LIMS, if this is inadequate the users will experience problems and place blame on the LIMS rather than the lack of training. How to get adequate training? Your vendor is the key to training, you should plan and discuss the training requirements for your installation with the vendor. Quality Systems International web : www.qsiuk.com email : sales@qsiuk.com Mistake Number Ten – User Buy In Not winning the hearts and minds of the users is a sure way of having a disrupted project, or a total project failure. How to ensure user buy in? There is a fine balance in how much involvement all users have during the design part of the project, in practice only key users are involved, however once the specification is completed all users should be given an opportunity to review the document. This is the first part of the buy in process by making the users feel involved. All comments and feedback from the users should be treated in a manner that shows that you are interested in their issues. It is highly likely that issues raised are in fact wishes that are not justifiable or beneficial to the business and so care must be taken to explain to a user why their issue will not be addressed. It is important that no members of the project team act in a high handed manner with users when dealing with users’ perceived issues. It is also imperative that you explain to your LIMS vendor about your user buy in policy and that they follow your guidelines. As the project progresses the vendor will be in regular contact with your users, their attitude to the users can have a huge beneficial effect, or the opposite. Ensure that the LIMS has been fully tested and that a complete sample and batch workflow process can be fully executed from start to finish. Do not let end users near untested software, unless they have been trained and expectations set to assist in the testing process. Make sure the users have been fully trained and are fully competent with the LIMS. Plan your ‘Go Live’ day, so as to have the project team members and staff from the vendor available to handle the minor issues that arise with the users. Finally, the more issues you resolve in a timely manner with your users the greater the buy-in