In another article, I raised my giant nerd flag and wrote a companion piece to this article the Three Simple Hacks for Quality Assurance, and within it, mentioned the importance of the quality loop as a cornerstone of quality assurance. If you haven’t read it, I urge you to follow the link above after you’ve read this.
Today, I will give you some general tips on how to ‘do’ quality in practical terms in the designing and building of quality systems, whether from scratch or a major review of your existing systems. These same principles apply whether you’re building a complex system, a simple procedure, or performing an audit. The scale of operation makes no difference; the principles don’t change, nor does the process. The secret to excellent but relatively easy quality assurance is a smattering of systems theory, of which several different models exist. I won’t go into detail here today, but these tips all have a direct link to the concepts of systems theory, and no matter the size or scale of the system you are trying to build, they all have the same structure and behaviours.
So, here are my top tips for building robust systems.
Define Your Purpose & Acceptance Criteria
- If you don’t know what you’re doing or why, stop and figure it out. You need to know before you can do anything of use.
- What is the problem you need to solve?
- What criteria do you need to meet?
- How will you know if it has been successful or not?
Designing Your System
- Never underestimate the power of reverse engineering!
- Whether designing a new process or troubleshooting a problem, start at the end. Know the benchmarks that indicate success and failure.
- Working backwards. At each stage, know its purpose, what should be coming in and going out, and the acceptance criteria for each.
- Map it out. It will help identify where links exist to other systems and what information is being transferred into and out of yours at those points.
Know Your Boundaries
- Systems are rarely linear or straightforward; they often contain sub-processes or systems connected to or nested within them.
- Knowing precisely where boundaries and links exist between independent, dependent or nested systems is essential.
- Knowing where it begins and ends is just as crucial in simple processes.
- This demarcation defines the inputs and outputs and is crucial for auditing, troubleshooting and identifying where failures or problems originate. It might not be where you think it is or where a critical failure occurs.
Identify Trouble Spots
- An infinite number of things can go wrong in any process; some are minor and of no consequence, while others will be critical.
- Points where critical failures are more likely to occur are CRITICAL CONTROL POINTS (CCPs).
- You need to look for where information or material transfers from one system to another, or a significant transformation or processing occurs.
- CCPs provide feedback. Where systems are connected, useful information is generated for monitoring, auditing, or verification.
Write it Down!
- Documentation is necessary for operational efficiency, but this is not its only function. It is essential for quality, training, insurance, legal protection, etc.
- Documentation must accurately reflect what is actually done and how it is done, using clear, simple language. I cannot stress this enough!
- A baseline test for your documentation’s usability should be whether it can be easily understood and followed by someone new to the organisation during onboarding and training. If not, break out the red pen!
- Documented procedures must answer the what, how, when, why, and who and cover the possible problems and corrections.
- Some of the key indicators that procedures might need revising are as follows.
- A new employee with experience cannot follow a procedure.
- New or inexperienced staff become confused because the trainer cannot clearly demonstrate it, and it takes longer than it should to learn.
- Experienced employees are ‘adjusting’ how they perform a procedure to make it work for them.
- A procedure results in inconsistent quality or frequent failures with no apparent cause.
Identify Critical Records
- You must know when records get created and ensure they are captured and stored, but not all are created equally.
- Try to avoid duplication of records. Duplicated information that has not been processed, transformed, or transferred to another system is probably unnecessary, but make sure to verify this before deleting it.
- The guiding principle for determining if a record is critical is the consequence of not having it. If it were a disaster, then the record is critical.
- Some critical records are less noticeable.
Capture, Organise & Secure Records
- Processes generating records must be traceable to the associated entity and the appropriate record management system where the record is kept (closing the quality loop!).
- Whether a manual system or a tier one quality management system (QMS), ensure they are reliably and faithfully recorded, retained and recalled when needed, without the risk of corruption or destruction.
- Have multiple backups! Just not in the same place.
Audit Before You Are Audited
- Don’t wait for an external audit. There are three (3) preconditions for conducting internal audits.
- Scheduled internal audits. Yes, you do need to do them!
- External events that create obsolescence within a process (procurement, suppliers, raw materials).
- Corrective actions in response to an incident.
- During external audits, get out in front of any issues and seek advice if necessary. Use the auditors as a resource. They are not the enemy. However, if you aren’t proactive, you might just become your own worst enemy instead. The auditor will also appreciate that you’ve made their job easier. Auditing is demanding and mentally draining for everyone.
Corrective Actions
- Respond. Don’t React. When things go wrong, look at the how, what, when, where, why and even the who. Patterns will emerge, identifying the root cause and finding the most appropriate corrective action for a long-term resolution. Not just a quick fix!
- Whether building a new system or trying to fix an existing problem, knee-jerk reactions to issues without planning won’t help. They will compound the problem and probably create a raft of new ones!
- Actions have consequences, not always immediately, but they happen eventually. Stop and think about those as much as you can – go back and reread the reverse engineering points.
- If you can’t definitively identify the source of the problem, you can’t fix it, and no amount of tinkering where you think the problem is will.
- If you’ve planned and documented, you should be able to anticipate all but the most obscure outcomes.
- Document your findings. It is a valuable resource for reducing reinvestigation and operational downtime the next time a problem arises.
Test. Check. Correct. Monitor. Repeat.
- Initial testing must be rigorous, but monitoring and continuous improvement is forever.
- Sorry, there’s no other way.
The trick is in the thoroughness and extent of planning you do in the lead-up to building your system. Put the quality in at the front end. If you haven’t read it, please check out my other article; Three Simple Hacks for Quality Assurance, which explains this further. Don’t cut corners because you’ll spend more time fixing problems and mistakes, fielding complaints, and working harder than before you put the system in place. A poorly designed system increases wastage, cost, and workplace dissatisfaction. Work smarter, not harder. That’s the whole point of systems, and most definitely applies to their initial design. No system is perfect straight off the bat, but it’s much easier to fix a well-designed system than a poor one.
A final comment: There is one factor you have absolutely no control over: Never underestimate the possibility or impact of a single inadvertent, random, or deliberate action! All systems have operational tolerances or limits, and some people have a gift for finding them! Mistakes and failures will always occur, and not all eventualities can be planned for or anticipated. However, the better the system is initially planned and built, the easier and quicker the identification and resolution of problems, with the bonus of a proper fix and improvement rather than just a band-aid and a prayer.