Ibbaka

View Original

Skill clusters and role archetypes - insights from pricing experts

Steven Forth is a Co-Founder of TeamFit. See his Skill Profile.
David Botta is an artist and data scientist at TeamFit. See his Skill Profile.

Skills come in clusters. Any skill will be associated with other skills and one can measure the strength of this association in several ways. What do these skill clusters mean and what can we learn from them? This is an active area of research at Ibbaka and informs our open competency models - this post dives deep into the skill clusters of pricing experts

Why is skill cluster research important?

We want to understand the patterns of skills associated with specific roles.We want to see if certain skills act as magnets around which other skills cluster.

Let's look at a concrete example. We recently undertook a piece of Skill Insight Research into pricing expertise. Ibbaka is growing its pricing expertise to better support our customers' businesses and wants to understand the skill trends in this profession to better understand who to hire and how to train them. One question that emerged from this work is "Are there different ways in which people approach pricing challenges and is this reflected in their skill profiles?" If the answer to this question is 'yes' then what do these patterns tell us?

We ran the data gathered for this skill insight research through some of our analytical tools. The data set came from (i) a large survey we conducted on pricing skills, (ii) LinkedIn pricing groups, primarily the Professional Pricing Society, (iii) the Ibbaka Talent Skill Graph. We found four groups. More on the methodology below.

The 4 types of pricing experts (according to their skills)

After exploring a number of axes, we settled on two that seem to capture the key difference between the groups.

  • X-Axis - does the person work with well-structured problems or poorly structured problems?

  • Y-Axis - does the person work at the leadership level or the operations level?

Let's look at the skills associated with each pricing expert role.

  1. Analyst - This is the most common role. It represents about 50% (N=274) of the total. Analysis of other data (LinkedIn, Job Postings) suggests it is even more common in the wider world. Associated skills include statistics, data analysis, pricing optimization, pricing segmentation and so on.

  2. Coach - The pricing coach works primarily with sales people and is as likely to be found in sales operations as the pricing, marketing or finance function. It represents 33% of the total. This is a critical role in value-based pricing. Most sales organizations need a lot of help in communicating value and the coach is the key enabler here. Communication skills and industry knowledge are more important than analytical skills (though some level of analytical skill is generally present in the skill profile for this role).

  3. Strategist - This is the role many consultants and pricing executives aspire to. 14% of our dataset show up in this cluster. In the broader world, I suspect the percentage is lower. Our surveys and the members of the Professional Pricing Society skew towards consultants who aspire to be strategists. The most important skills here are the ability to listen (and active listening), to recognize patterns, to structure choices, and to handle ambiguity.

  4. Designer - This is an emergent role. It is relatively uncommon at about 3%. With so few people it is premature to say much about this role archetype or even if it is a stable role archetype at all. The designer sees pricing as part of a larger system, one meant to understand, create, communicate and then capture (in price) the value of products and services. Pricing designers are often consultants brought in from outside to work with product managers to come up with the basic pricing model. The skills associated with this role are not yet stable, but we expect to see things like service design and market segmentation (which requires both qualitative and quantitative data analysis).A different way to think about pricing roles is to ask which role, if any, is best equipped to solve the different types of problems described in the Cynefin framework.

At the risk of oversimplifying (and annoying people in specific roles) I make the following associations.

Analysts (and the current generation of pricing software) are best at solving obvious problems, which are tightly constrained. Most pricing optimizations are treated this way. Designers solve the complicated problem of pricing model design. Coaches have to deal with other humans, often in conditions of stress where there is little time for real analysis. Humans are good at these complex problems. Strategists help executives make choices under conditions of uncertainty. They help people escape from analysis paralysis and deal with change. The problems they are called to solve are often Chaotic.

For any significant role, there are likely to be a set of archetypes that represent different ways of addressing the problems associated with the role. David Botta and I both claim the skill of design thinking, but we bring very different associated skills to this role. Combining our different approaches is one of the things that drives innovation at Ibbaka. (Compare our skill maps on Ibbaka Talent, they look very different.)

Skill cluster methodology

Our current vision for role archetype identification is that we will triangulate three types of data.

  • Skill surveys (the above analysis is based largely on an application of our clustering technology to the data from the Skills of Pricing Expertise survey

  • Job role analysis (we have not decided how to handle this data, which comes from several different sources including job posts and what we can tickle out of LinkedIn data

  • The Skill Graph, where skills are connected to roles and to other skills

We have made good progress on our clustering technologies and on understand how to gather data that leads to meaningful clusters. One critical thing is that one must combine structured and unstructured data. The unstructured data generally comes from open questions in surveys, interviews and various documents. The next step is to combine the Skill Graph and clustering analysis into a pattern recognition package (we are writing our own package in R) that can propose the skill-role archetypes. We want to do this while allowing for the possibility that for some roles there will not be clear archetypes (we don't want to overfit).