Closing the Knowing-Doing Gap

I have a client with a common problem: The firm has implemented a sophisticated quality management system but many people aren't following it. Sound familiar?

Substitute any number of corporate activities and directives--from making sales calls to implementing strategy to filling out time sheets--and in all likelihood your firm has the same problem. Employees aren't doing the things they're expected to do. They know what to do, they're capable of doing it, they're even motivated to do it in some cases. But it's still not happening.

Stanford professors Jeff Pfeffer and Bob Sutton call this predicament "the knowing-doing gap" in their book by the same title. They conclude--and I would concur--that the biggest difference between companies is not what they know, but how well they're able to put what they know into action. Best practice insights are a commodity these days (just read this blog!), but implementation acumen is a rarity.

In recent posts, I've shared several strategies from the field of performance management. Let's apply that wisdom now to the challenge of getting things done. Most firms take a familiar path in trying to solve problems like my client has. They step up the pressure, tweak the process, reassign responsibilities, do more training, modify goals.

If you read my last post on quality, you might recognize that these steps are all antecedents, things that come before and set the stage for action (or behavior). Antecedents are important, but they're not effective in sustaining behaviors over time. Unfortunately, most managers rely almost exclusively on antecedents in trying to change behavior. There's a better way. Let me outline some key steps in closing the knowing-doing gap in your firm:

Define the specific desired results. Sometimes firms launch initiatives without clear objectives. For example, if you implement a new quality process, what do you hope to accomplish? Improve quality? That's not a very helpful goal (by the way, most quality programs fail to significantly improve quality). How much improvement do you expect? In what specific areas? How will you measure it?

I don't think I need to review here the qualities of SMART goals. You're undoubtedly familiar with the concept. Yet I'm surprised how many firms I've witnessed investing substantial time and money in various strategic efforts that lack explicit performance goals. That makes it much harder to change behaviors. Which do you think works better: Ask an employee to work harder or tell her specifically what more needs to be done? Review your goals and see if they meet the SMART criteria.

Seek to understand why. If people aren't doing what they should, start by exploring the reasons for this. Certain antecedents may be a factor, but you need to consider the consequences of behaviors as well. If someone isn't doing what is desired, that behavior is undoubtedly being reinforced in some way.

For example, failing to do a quality review saves time, is easier, may fit in with one's peers, could earn a compliment for finishing the work on schedule. The individual may not be aware that these consequences are influencing his actions, but you can uncover likely sources of influence through some inductive reasoning and asking good questions. Once you have a better understanding of why people do what they do, you're able to take more effective steps to support behavior change.

Address antecedent shortcomings, but don't stop there. We're all accustomed to the usual fixes--new or revised programs, policies, procedures, action plans, tools, reorganizations, trainings, etc. These can all be part of the solution, but usually are insufficient in closing the knowing-doing gap. Sometimes the "fixes" even exacerbate the problem of inaction.

The important question is always: How will these steps help people do what needs to be done? Be persistent in pursuing the answer to that question. Most "structural" solutions to organizational problems are incomplete. New processes, of course, can only be effective when followed. Technology investments require a corresponding change in how people do their work. Training rarely is effective unless reinforced over time.

Changing behaviors is almost always part of the solution. And it's usually the hardest part. So let's talk about that next...

Identify specific behaviors needed to achieve your desired results. This requires a step called pinpointing, determining those few behaviors that are most critical to achieving your desired results. Don't get overly ambitious. This is the problem with most corporate initiatives--like implementing a quality management system--where firms try to tackle too much behavior change at one time.

A better approach is to phase in change, guided by staged objectives. Don't attempt full compliance to your quality procedures at first, for example. Instead, pick perhaps five pinpointed behaviors that will have the biggest impact on quality. Once those behaviors become commonplace, then add a few more and so on.

Use effective metrics. There are two types of measurement associated with pinpointing solutions: (1) leading indicators that typically measure behavior and (2) trailing indicators that measure results. Obviously, the later is far more common in business. But measuring behavior enables you to better evaluate your progress and to target course corrections where needed.

How do you measure behavior? There are two primary ways--counting and judging. Counting, of course, is more objective and should be preferred where possible. Having identified pinpointed behaviors, say completing a discipline-specific technical review for all multidisciplinary projects, you can then count how often that occurs when it is called for.

For behaviors that don't lend themselves to counting, you should consider judging. Because it is subjective, this kind of measurement should come from more than one person. For example, you could have project managers anonymously grade how well department heads support them in getting quality contributions from technical staff. The composite grades could then be tracked over time to look for improvement.

Metrics work best when oriented towards providing positive reinforcement. Unfortunately, many firms use metrics for negative reinforcement. Whenever you have the choice, choose to measure desired behaviors versus problem ones.

Provide regular feedback and reinforcement. Imagine your favorite college football coach giving instructions to his team on how to conduct the prescribed practice drills. In this case, however, he sends them off to practice on their own while the coach goes to his office. "You can come to my office if you have questions," he tells them, "But I'll wait to give you feedback until after the game on Saturday."

Obviously he wouldn't last long in the coaching profession. That approach clearly would not succeed in getting top performance from the team. But did you notice the familiar ring to that illustration? It's how most business managers direct their teams. "Here's what you need to do. Let me know if questions come up. I'll give you feedback after you've finished."

If you're going to provide effective feedback and reinforcement, you need to periodically observe the pinpointed behaviors. The more frequently, the better. Too busy for that? Maybe you should reassess how you allocate your time if you serve the role of manager. The manager's first priority, in my opinion, is helping the team succeed. (For more on this, check out my earlier post on "The Time Investment Principle.")

What's the difference between feedback and reinforcement? Feedback is sharing information that enables one to adjust their performance. Measurement can be an effective tool for providing feedback. Reinforcement involves creating or leveraging consequences that cause behavior (in this case, the desired behaviors) to increase. I dealt with this aspect of behavior change in an earlier post on positive reinforcement.

Any solution that involves changing behaviors should include these steps. To review, these are the key questions you should address in closing the knowing-doing gap:

• What are the desired results?
• What's motivating people not to act as expected?
• What are the vital behaviors needed to produce those results?
• How will we measure progress toward both the pinpointed results and behaviors?
• How will we provide performance feedback?
• How will we reinforce the pinpointed behaviors?