I’ve just finished my term as president of the Potomac Chapter of ISPI. I’ll remain on the board for the coming year as immediate past president. I’ve been working with our vice-president of finance to bring some order to our chapter’s finances, in part by recreating our books using QuickBooks Online.

Yesterday and today, I spent time developing several job aids for new members of our chapter’s board, and for working with the QuickBooks system. As Joe Harless used to say, there are only two places to store knowledge: inside someone’s head, or outside of it.

Inside is usually way more expensive.

Early in my career, I was lucky enough to attend Joe’s Job Aid Work Shop. JAWS, as he called it, opened my eyes to a practical, evidence-based way to improve performance on the job.

For many tasks, and practically all procedural tasks, you can develop effective job aids in far less time than you’d need to train someone to perform the task based only on memory — in other words, to learn the task. That’s because a lot of tasks don’t need to be learned.

How can you tell? All things being equal, you should consider job aids based on:

• How infrequently you do a task. The less often, the stronger the argument for a job aid. Dialing a telephone: teach it. Programming your cell phone: job aid.
• How large the task is. The more steps there are, the likelier you should job aid.
• How complex the steps are. The more complex the steps, the likelier you should job aid. Complexity might involve:
  • Precise measurements
  • Difficult discriminations (when it’s hard to tell X from Y)
  • Decisions within decisions (if it’s a purchase order, and if it’s from Division 6, and if it’s for equipment, and if it’s over $1,000…)
• How likely the task is to change. When you’re pretty sure the steps will change, it makes sense to develop a job aid — it’ll take less time to revise and update than training-to-memory.
• How serious the consequences are. The more serious, the stronger the case for creating a job aid. Burn a CD: teach it. Install a new operating system: job aid.

There’s also the question of your deadline for development.  Creating a job aid always takes less time than creating training-to-memory for the same task.  If you build job aids whenever you can, you avoid spending time on unnecessary training.

So what’s the link with my Potomac Chapter activities? We’re a small volunteer organization. Not everyone remains on the board from one year to the next. Institutional memory in this setting is fragile (not to say feeble).

On the one hand, we need strong tools to help us manage — like an online bookkeeping system that doesn’t rely on handoffs of paper records from one year to the next. On the other hand, we need a way to help new volunteers contribute productively (and with a minimum of personal frustration).

I’ve also found that the analysis leading to a job aid helps me tremendously in understanding procedures, and uncovering places where I don’t understand them. Or, once in a blue moon, where the procedures aren’t clear or don’t cover all the situations they need to.