Charting Success with David Pardo: The Story Behind SNGULAR's Skills Tree

Recently, we set a goal for ourselves of having quantitative indicators that would allow us to better plan the professional trajectories for our people at SNGULAR. By incorporating this analysis into our professional development, it gives a clearer picture of how to help team members be better prepared for their positions and provide a clearer picture of our scope. David Pardo is the running point on the Skills Tree project, which allows us to map the skills, competencies, and knowledge of each person. By gaining more comprehensive knowledge of each team member's attributes, this ensures that these competencies align with the objective of SNGULAR.

In the following interview, you can learn in much more detail how this work was carried out and the value we are beginning to obtain from it:

How was the process to build SNGULAR's Skill Tree?

We began the process by reflecting on previous initiatives with this same goal, by accessing what could be reused, and combining it with external data. The CSO from The Open University and Manfred's competency data were particularly useful.

Once we identified the need to build our own ontology, we began with a hierarchical list of technical knowledge. We structured it as a tree, with "requires prior knowledge of" as the main relationship. To avoid spending too much time on the technical side, we wrote it in markdown format and stored it in a GitLab repository as plain text. In addition to technical knowledge, we also incorporated soft skills and sector-specific capabilities. We plan to add experience in functional domains according to team specialization but have postponed this since the teams already possess this knowledge.

After creating an initial version with general elements, we prepared pipelines to organize data in a friendlier format. We chose the Neo4j graph-oriented database for this. That was when we began to enrich the skill tree with more details. As soon as we had a more or less complete version, we opened it to SNGULAR's teams. Instead of creating a specific frontend, we integrated with Google Spreadsheets. It provides 80% of the capabilities at less than 10% of the development cost. In order to incorporate more active voice: Each team director has a sheet they can mark skills as strategic or in use. As a result, we reduced the elements on the tree from 1323 to 713. the tree from 1323 to 713 elements. Some obsolete technologies remained, such as COBOL and FORTRAN, but also others that might become relevant in the future, like columnar databases or stream processing systems like Apache Samza. This filtering allows us to focus our business on technologies that have at least one champion committed to keeping that technology alive within their team.

After filtering and importing pre-existing data from scattered documents, we prepared another pipeline that generates a specific document for each colleague. Although it was tedious process, we asked each individual to provide information about their personal relationship regarding their respective skills. Are they interested in the skill? Do they have experience? Would they prefer not to deal with it daily? With this insight, we had everything we needed to start adding value to the company's activity.

What were the main takeaways in this process?

Personally, the most important thing was getting to know inner workings of SNGULAR. I had to sit down with a significant percentage of the organization to understand our strengths and weaknesses. From a technical standpoint, we used known and solid technologies. We preferred to play it safe rather than try new proposals. Boring technology has its advantages.

What stage of development is this project in and what are the next steps?

In recent days, we have added all the information from the Zoho ecosystem, which we use at SNGULAR, using Google Big Query data as the golden source. Additionally, we have expanded the scope of the skills document to now include how people and projects are connected to understand different contexts for the use of each technology along with its relevance and maturity level.

The next step is to include a 360 evaluation of colleagues to detect hidden knowledge and future leaders in the organization. I would also like to include all primary data (capabilities, project relationships, analysis...) in Tableau for integration with financial and analytical data.

What short-term benefits are you observing since its implementation?

The benefits are seen from four different perspectives related to Sngular's organization:

- Commercial: We acquire an immediate and long-term vision of the company's capabilities from a skills standpoint. We can objectively answer questions like, "Can we confidently handle a CQRS and Kafka microservices project in three months?"

- Communication and Marketing: The system allows us to set priorities. We know where to focus our efforts to be more efficient in both customer acquisition and talent hunting.

- Delivery: The objectives here are fourfold. Eliminate knowledge silos, improve job allocation, audit teams' suitability for client needs, and detect potential problems early.

- People: Ubik allows us to formally define roles in the organization. Identify the set of knowledge, experience, and skills needed to advance in our professional careers. Answer the question, "What do I need to reinforce to keep growing, and who can I learn it from?"

In summary: UBIK is a reflection of the organization. We extract distributed knowledge and make it available to everyone.

How does this project fit within the Teams of Teams strategy we are developing at SNGULAR?

An initiative like UBIK is necessary in any organization whose product is based on software development knowledge, whether the strategy is based on a sector-specific organization or focuses on functional domains. It will serve as a ubiquitous language for communication between teams, but also with the rest of the organization's stakeholders. In my opinion, UBIK is compatible with a Team of Teams strategy and also with more common customer-driven structures.

What is the UBIK project and how is it related to the Skills Tree?

UBIK is the set of tools that encompass Sngular's People Analytics and Workforce Planning Optimization framework. The Skill Tree is the ontology of knowledge required for software development. UBIK contains all the initiatives related to the Skill Tree.

How could other companies implement a similar process?

The first step is understanding the organization and its environment. Objectives, strategy, stakeholders... From there, you can begin to analyze the current workforce. What they can do, what they're interested in, where they differentiate. With that knowledge, you can make proposals that align both sides. It's essential to have tools that allow you to measure if you're approaching the goal. If we promote certain skills, invest in training in technology, soft skills, or domain knowledge, we need to know if that investment has a specific return. In consulting firms, the rest is straightforward. Provide data on opportunities, proposals, assignments, and capacities to cross-reference them. Train people in the necessary skills, assign them to vacancies where they can add the most value, and match workers' needs with those of clients.

What benefits do you think would be obtained from it?

The most obvious are the economic benefits. An initiative like this makes it possible to better match people to the needs of projects, thus leveraging more capabilities. It's also essential to systematize career paths within the company. Knowing what someone needs to strengthen to move up the ladder helps retain employees and improves talent acquisition. However, it's crucial to remember that this is a tool, and as such, its benefits depend on the decisions made based on it. The more mature we are as a company, the better we can expect to leverage it.