We talk about data science daily and the immense value it can bring an organization when understanding data assets and improving business processes through prediction and measurement. Despite how often we hear about data science, many organizations are still early on in this journey and have not made the first steps to hire data science talent into their team. The reason for this lack of investment can vary from lack of funding to lack of a leadership team willing to champion the capabilities and value to the organization.
When hiring your first data scientist in an organization, the leadership championing the investment must ensure the conditions for success are set and that the new hire will effectively apply their skills to the most impactful parts of the organization. Simply put, your first data science hire will only be successful if they focus their time on the most impactful projects, bring the proper mix of skills and are backed by platforms, investment and organizational flexibility.
As you look to hire your first data scientist, you must consider and set the conditions that will make them successful. As with any new skill or capability, the organization must make a wide range of investments and adjustments to its operating model to ensure this capability is effectively utilized. Setting conditions for your data scientist include:
- Access to Data & Data Platforms – A data scientist is only as effective as their tools and systems. As you build a data science capability, establishing a basic level of data platform or enterprise data warehouse capabilities will ensure the new hire can hit the ground running with data that has been collected, enriched and validated against business processes.
- Engaged Executives – The operating model for working with a data scientist differs from other roles, and the executive team must embrace the skills and experience this new role brings. This manifests through executives that spend time explaining their organizational structure and goals to the data scientist and engaging in the brainstorming of possible business structures and processes to apply analytical techniques.
- Defined Expectations – The organization must be aligned with the objective of the data scientist. This enables all supporting teams to apply the right investment, time and engagement for success. Expectations can focus on specific business units, specific parts of the customer journey, new innovations the company looks to bring to market or a focus on specific back-office operational functions.
- Supporting Capabilities – Data scientists work best when they have supporting teams around them and tooling to facilitate their data modeling activities. Hiring a data scientist is the first step. To truly ensure they maximize their impact, they will require the support of data engineering teams and a budget to purchase tooling and compute capacity.
- Culture of Experimentation & Change – Many times, a data scientist will have a new analytical model or data finding that must be validated on a subset of users or over a period of time. The organization must build processes and approvals to enable this experimentation quickly and with minimal delay. This is facilitated through processes for a quick approval and a culture that is willing to experiment, measure the outcome, accept success and failure and move on quickly.
After you lay the groundwork for your new hire, you should look at who you will hire as your first data scientist. This key hire will set the tone for engagement with business partners and lay a framework for how the team will eventually grow. Some critical considerations during the hiring process:
- Job Description – Building off of your defined expectations, your job description must capture the anticipated business partners for this role, the organization’s operating model, the types of problems the data scientist will tackle, and the technologies to be utilized. This gives the recruiting team proper clarity for identifying candidates and the candidates the proper context to set expectations for the role and level of learning they will require to succeed.
- What it Means to be First – Setting the tone for a new capability in any organization can be difficult. A question that must be asked of every candidate for this role revolves around their level of understanding – what it means to be the first new data scientist, how they plan to leverage that position and what their steps will be in the role to better educate others about the role, expectations and operating model. The more clarity the data scientist can provide on WHO, HOW and WHY they will engage, the higher their chance of success.
- Someone That Was First – Even better than hiring a data scientist that understands what it means to be first has built this capability at other organizations. Their experience navigating organizations, educating others, creating a technology stack and defining early goals will be highly valuable and accelerate the journey of applying the capability and growing its impact.
- Mentoring – Much of the work for this first data scientist will be mentoring others on the potential for the capability, how to apply data science techniques to different problem types and advise the organization on how to assess potential use cases against a limited capacity to explore and create models for use within the business. This first hire must be effective at mentoring others and educating them along the way. I have found data scientists that also teach highly impactful because of their dual skill of educating and solving complex problems.
- Patience – As with any introduction of new capability to an organization, change and acceptance take time. Your first data scientist must understand this. They must be able to advise others on realistic timelines to complete specific work activities but also work with other teams that are building new muscle memory and operating models.
- Operating Laterally – Most important for this data scientist is their ability to operate outside their management chain and work laterally across the organization. This will maximize their impact and the awareness of their capabilities. This is crucial to facilitate access to data, validation of results, acceptance of approach and execution of process change to apply learnings and analytical models.
Once you hire your first data scientist, their first 6-months will be critical to establishing new operating models that become routine while setting conditions to scale the data science organization. Engagement in impactful projects should be your priority while identifying new investments in people, skills and technology required.
- Engagement & Awareness – The most important thing this new hire can do is meet as many people across the organization as possible, listen to their needs and begin to paint a picture of where data science will be impactful. This should be followed quickly by prioritization of asks by complexity and impact to begin to organize work and to set expectations.
- Inclusion – You should ensure your data scientist is included in all staff meetings that pertain to the expectations you set earlier regarding their focus, including business units, specific parts of the customer journey, new innovations the company looks to bring to market or a focus on specific back-office operational functions. Access to the key decision-makers and regular communication will facilitate a shared understanding of the data scientist’s work and sharing of outcomes.
- Metrics – Today’s enterprises live and die by OKRs. It is a key tool for executive teams to set priorities for evolving organizations. The earlier your data scientist can establish their own OKRs, aligned to corporate priorities, the sooner the business will understand and support their efforts.
- Investment – As your first data scientist begins to understand the areas of engagement and the business teams with the biggest opportunities, you will inevitably identify tooling and technology needs to support their work. You must ensure they have support from leadership to invest in tools that will make them more effective and impactful.
- Accepting Failure – The first model your data scientist builds will most likely fail to meet expectations. They will probably explore datasets of low quality that do not yield actionable results. We must accept these learnings, document them and move on quickly to identify more valuable work and focus areas. This is part of the process and an acceptable learning curve for new capabilities.
Starting this journey will be a learning experience for your organization. You will be learning new terms and technology and applying them to new operating models. Your success comes from communication, collaboration and flexibility. Many organizations will leverage an outside consulting firm to begin their journey and ensure the technology and operating models are in place when your first data scientist joins the team.
We have helped many organizations begin their data science journey. Reach out if you are beginning yours and are creating conditions for your first hire to be successful!
Watch for future postings discussing scaling your data science team after an impactful first hire.