Posted by on October 26, 2013

This is about the ‘stretch model’, which is a useful way for verbalizing to people how you intuitively rate their performance. My view is that performance is contextual – it depends as much on the environment as it does the individual person. To work with performance and to be able to improve performance it’s important to be able to articulate and verbalize how you as a leader perceives the performance of an individual. I used this model myself since about 2001.

(This text is from appr 2001, but with a few minor updates)

Every manager seems to know by instinct, which the best and worst performers in the organisation are. The intuitive ranking may be neither objective nor fair, but still I believe that it’s a ranking that we all make.

As a simple example, think of the two questions for anyone in your organization: Would you be sad to see the person leave? If you had the chance, would you hire the person again?

Even if not explicitly verbalized, we tend to use rankings like this implicitly when deciding on work assignments, promotions, rewards, compensation and benefits.

For most purposes the subjective ranking works well.

The challenge I have faced is that it’s sometimes difficult when you need to share your intuitive ranking with others. You need to cross over from a strongly subjective ranking to an open and objective ranking.

The first problem when having to give an objective performance evaluation is to find out what the objective criteria should be. Should it be performance against job description? Targets? Expectations?

In the ideal world, this would never be a problem, because of course (right?) all people working for you have clearly defined, agreed, and documented job descriptions, with clear and measurable targets for their personal performance.

In practice people tend to have quite flexible job descriptions and their roles are drifting over time as organisations are always evolving.

One of the toughest situations you can be faced with is when you realize that a person is not performing and you need to explain why. In some cases it may be easy, because there is a clear violation of a rule or a severe error that has been made. In other cases it can creep on you until you one day realize that everything is not as it should be. But how do you tell it?

The stretch model

I have a small model for being explicit about performance. It’s called the “stretch model”.

As the main problem is to be specific about performance problems, this is what I will focus on here. But, the model can be used equally well to be explicit about top performance.

The model has three layers: Team contribution, personal development and social contribution.

I consider each person as a player on a sports team. Each player has an assigned role on the team and an immediate superior, his coach. I think of the first level, team contribution, as the evaluation of a player during the active game time of the team, when the team is playing on the field. Playing by the field is synonymous with “real work”. The second layer, personal performance, is about how the player works with personal development, to improve in the role on the current team, and to develop as a player to play in even better roles, teams or leagues in the future. The third level, social contribution, is about how the player impacts the overall athmosphere – on the team, during trainings, during social events and everywhere else.

The question I ask for team contribution is “how much does the player stretch to perform in the assigned role and make the team win?” I take the viewpoint of the coach sitting off the field and watching the player in action. Could the player stretch more? Does the player have a good feel for the game and where to give the extra stretch for the team?

Stretch needs to be understood in a balanced way. No player should break during the game. The challenge is to stretch enough for the win, yet be able to complete the whole game and future games waiting.

Stretch is multidimensional. Each player needs to be understood in terms of the elements of the personal skill set. Stretch can be scoped and applied to the relevant elements. An overall evaluation of personal strengths and weaknesses also has an impact on defining a reasonable level of stretch. Except for this one paragraph, this presentation will not be concerned with the specific elements of a persons skill set. The player analogy intuitively accounts for all of this, but when you need to be explicit, the details are important.

Based on the “stretch assessment” you know for your self if a player is to be described as a top performer, what the variation in performance is and if the player has a potential to stretch more.

If the player has a potential to stretch more, or if the stretch is to no benefit for the team, then that covers a potential performance issue. Sometimes you find that a player is not stretching at all. And sometimes you find a player playing a totally different game that the team.

If a player has a problem performing with the team on the field, then that issues needs to be addressed before considering the other levels. The primary contribution of each player to the team is the basis for everything else.

For the players making a solid and durable contribution to the team during games, the next concern is about how to develop to increase their performance.

Also for the personal development each player need to have a stretch. Stretching towards meeting own targets as well as targets for the team, and at a higher level – here the model breaks – for the company.

The final concern is about the social contribution. Does the player have a supportive and contributing role or does the player in some way have a negative impact. How well does the player go with the other players on the team or other teams in the company.

I have been challenged with players who got it totally wrong. One player had the idea that a solid social contribution and stretching for personal development was enough to stay on the team. But I could use the model to explain that without the solid team contribution none of this was relevant. As a consequence I cut his training and required fewer and much shorter coffee breaks, while focusing on “real” work for a while.

Another common problem I have had is players who insist on opportunities for personal development without showing any stretch for their team.

A property of the model is that it nicely includes the concept of “being lazy” as “having a lack of stretch”. Similarly it can be tied to concepts such as focus, attitude, etc.

What I like about the stretch model is that it allows me to express my assessment of a person’s performance without needing to be explicit about job description and target. Intuitively each person understands that he or she is on a team and also intuitively understands the mission (game plan) of that team. Also the notion of stretching to win for the team rings a bell.

This means that you can discuss about performance, “stretch”, in intuitive terms and only explicate personal and team objectives as needed.

I have encouraged each member of my organisation to think for them self about the stretch they make, to better understand their own contribution to the organisation they work in.

In GE (See GE Field Book) an ABC ranking of employees is described. For me the stretch model is useful for analyzing each player before ranking [or even better – not ranking, but giving specific feedback]]. The As are your superstars. The Bs are the ones you won’t sell, because they will be worth more tomorrow. The Cs are the ones on your “transfer” list, for one of various reasons. This way of describing ABC is not fully consistent with the GE way, but this is how I have used it. In GE it’s more a “rank & yank” thing, but that approach is not how we do things in the company cultures that I have been working in.


Related topics

  • The rule of 11. A colleague (Timo H) told this to me. Accordingly Intel is using this rule. A good size of a team for personal management is 11, because this will be big enough for you to clearly see a difference in performance contribution, and small enough for you to manage.
  • Personal performance can vary a factor 1-30. Team performance 1-20. (I guess this can be seen in the CocomoII model, e.g. using the factors Steve used)
  • Rewards and benefit does not always follow performance. GE found that e.g. stock option grants did not match the ABC classification (See GE Field book)
  • Compensation is not just money. People use a multi dimensional system to calculate the payback. For some, the substance of the work is much more important than the pay. For others the social status of an expert role is more important than the pay. Yet for others, money is the only think that counts. People’s performance can vary a factor 1-30 for seemingly no reason. Within the same company, the variation may be more modest, let’s say 1-10. But people’s salaries tend to not vary at the same scale. Why? Because compensation is more ta´han just money.
Posted in: Identity