Thursday, 2 April 2009

Engagement surveys - linking satisfaction, engagement and business results

OK, so I'm still posting my thoughts on engagement following on from HR's recent post on this that I referred to under the first stop on my new tour of the blogosphere, (ie the letter 'A').

I've posted about engagement being a bucket, and about engagement vs satisfaction - about it being more about commitment, but also something that is a measure of output rather than activity, and which is also of a higher level of value in the HCM value triangle.

These factors mean that engagement surveys can be really interesting because they provide an opportunity to link satisfaction and engagement, and to draw some conclusions about causal relationships between these elements in the activity and outcome columns of the HCM value chain (part of the general service value chain or strategy map).


Satisfaction (activity) questions should focus on 1) the elements in the employee value proposition ie those practices that the organisation is hoping will engage its staff; and 2) key concerns of staff which can picked up through an initial diagnosis before the survey itself begins.
Engagement (output) questions should focus on the individual people themselves, for example:
  • Alignment
  • Commitment
  • Advocacy
  • Passion
  • Effort (discretionary behaviours)
  • Retention

There's nothing new about this insight, but I'm still often surprised by the number of surveys that don't provide this capacity.
Problems with Engagement Surveys
Some of the problems I encounter include:
  • No attempt to link further forward in the value chain, ie to correlate engagement (outcome) with business impact
  • Not enough engagement questions - you need at least half a dozen measures to make it worth correlating responses for satisfaction and engagement questions
  • Not specific enough engagement questions ie lack of thinking about what needs to go in the bucket (survey firms that want to maintain their benchmark databases are the main cause of this problem)
  • No engagement questions, only satisfaction ones (Gallup is the main culprit here - Q12 consists of satisfaction questions which have been correlated with business impact but it misses out the vital level of outcomes in between.  Ie engagement is defined as something that is caused by satisfaction and itself leads onto business results, but Gallup deliberately avoids specifying exactly what engagement really is!
  • Lack of detailed reporting - no ability to interrogate the data, to be able to report locally on a team's engagement to the team manager etc.

My advice - don't put up with problems like these.  Develop your own, bespoke survey in-house, based upon key engagement drivers that you've identified beforehand (through focus groups etc).  Accept the trade-off that you won't be able to benchmark your results with the fact that you'll end up with something much more meaningful for you.

Previous engagement posts:

My next post will help you extend beyond engagement to survey on human capital itself.

  • Consulting - Research - Speaking  - Training -  Writing
  • Strategy  -  Talent  -  Engagement  -  Change and OD
  • Contact  me to  create more  value for  your business
  • jon  [dot] ingham [at] strategic [dash] hcm [dot] com .


  1. Great post. empirical approaches to causation in so many HR practices. The first problem you state is killer.

    There's a great XKCD cartoon: "I used to think correlation implied causation. But then I took a stats course and now I don't think that","Sounds like the course helped","Maybe".

    At Sonar6 we've been studying recently whether the competencies that businesses measure staff on actually predict performance. This is not hugely complex regression, yielding surprising results... but the most surprising thing is the way this sort of analysis is still considered out of the ordinary.

  2. Thanks for the comment Mike. Agreed. Hope you'll let me know when your study is published too. Even if organisations can't conduct this sort of analysis themselves, they should still know about the research that has been conducted elseshere - it all supports the business case.


Please add your comment here (email me your comments if you have trouble and I will put them up for you)