Lost in Data Translation

Solving the disconnect between marketing analysts and decision-makers with data translators

Imagine you are a marketing analyst who just spent days wrangling, analyzing and summarizing data to help your company answer critical questions. You present your statistical results and, instead of the “aha” moment you expected, your audience seems unimpressed. You feel frustrated that your insights are not having the impact you expected.

Conversely, imagine you are an executive who continually hears that you need to use data to make marketing decisions. Your marketing analyst is presenting the information you need to move forward with a big investment, but it is a blur of statistical jargon. You feel frustrated and resort to making decisions based on your intuition.

Data analysts and business executives often speak different languages. Their roles and expertise may differ widely, making it difficult to interpret what the other role is saying. But as companies continue to embrace big data to drive marketing decisions, the different parties need to understand each other. 

Data science evangelist Hugo Bowne-Anderson interviewed 35 data scientists to find out what they do and how their job is evolving as demand is rapidly increasing for their skills. He found that the ability to communicate—to answer business questions and explain complex results to a non-technical audience—is becoming more important than the ability to build and use deep-learning infrastructures. Enter the data translator, the person who understands both the technical jargon and the intricacies of marketing strategy and can bridge the gap.

This article explores the importance of data translators and their key skills, comparing the job posts for data analysts relative to the desired traits of a data translator. Although experts say data translators play a critical role in organizations embracing data, my analysis of job postings shows that companies hire for both data scientists and data translators in one role. I argue that given the disparate skill sets of translators compared to data scientists, companies have unrealistic expectations for the role that one person can reasonably be expected to successfully perform.

What Is a Data Translator?

As more companies embrace a digital transformation, they hire more technically skilled employees such as software engineers, data scientists and analysts. These roles are often characterized as the “quants,” or people who are more number and data-focused. Data analysts play a pivotal role in providing the necessary data for marketing executives to make informed decisions. However, even the highest quality data will be ignored if the decision-makers can’t understand the data. Research has found that “leaders in senior management do not speak the same language as the analysts,” according to a 2017 article published in the MIT Sloan Management Review. The disconnect between data analysts and decision-makers leads to frustrated data analysts who feel their data is ignored and that their talents and insights are underutilized. In addition, the disconnect can lead to confused decision-makers who want to make data-driven decisions, but the data they receive is not presented in a language they understand.

Data translators require a unique set of skills in order to understand both the technical aspects of data analyst roles and the nuances of the business and its strategies to correctly interpret and communicate data for decision-makers and data users. The article’s authors Chris Brady, Mike Forde and Simon Chadwick created a checklist of seven skills that good data translators will have:

Overall, data translators need domain expertise, sufficient technical skills, good communication skills and leadership experience. Seeking data analysts who possess all of these skills, in addition to highly technical skills, is like searching for a unicorn.

If a unicorn cannot be found, one solution is to build out a data science team based on skills needed for the overall team, and then to hire and train employees to fill these roles collectively. “Rather than assign people to roles, define the talents you need to be successful,” writes Scott Berinato in Harvard Business Review. “A talent is not a person; it’s a skill that one or more people possess. One person may have several talents; three people may be able to handle five talents.” This approach allows individuals to highlight their existing skills and talents while relying on the skills of the team as a whole to meet the overall needs of the company. In other words, rather than expecting one individual to possess all the characteristics of a data translator, these skills are instead dispersed across and throughout the data team collectively.

To create a shared understanding and appreciation for other talents and skills, Berinato suggests that team members take specific actions to understand other roles. For example, data scientists can learn basic design principles to improve data visualizations and designers could learn basic statistics to understand analytical results. This exposure to other skills builds empathy within a team, an important attribute for team effectiveness.

Job Requirements Data

Given the emphasis experts place on the need for data translators for companies to be successful leveraging data, I was curious to explore how prevalently such a position might be listed on job postings. However, looking at Indeed.com, I found no job postings with the title of data translator and only two job descriptions including the phrases “technical translator” and “analytics translator.”

Perhaps companies fill this role internally with someone who steps into the shoes of performing the role of data translator. For example, an existing employee is likely to have domain expertise and could create this role for themselves. Therefore, this role would not be found on job posting websites as it is created organically.

A second explanation for why job postings for data translators are not more prevalent is that perhaps the roles for translators are included in other data scientist positions. For example, a study in the Journal of Information Systems Education analyzed job descriptions for more than 9,000 job postings for entry-level data analytics jobs from 2014 through 2018. Over this five-year span, the authors found significant increases in job postings requiring general statistics, modeling, model development, data management, database systems, business intelligence, programming languages and enterprise systems, as well as for specific software skills or languages, SQL server, Tableau, statistical packages, SAS, R, and Python. The increased requirements are all highly technical, except for Tableau, which can be highly technical but also could indicate an increased demand for creating quality data visualizations. As previously noted, creating compelling data visualizations is a key role of data translators as powerful data visualizations make data digestible to the intended audience.

Table 1: The percentage of required and preferred skills for data analyst and operations analyst roles from Indeed.com job descriptions, compared to each of the seven skills of data translators.

Specific Data Translator Skill	Data Analyst	Operations Analyst
Sufficient analytics knowledge to communicate effectively with the organization’s data scientists.	100%	100%
A willingness to search for deeper knowledge about everything.	86%	90%
Sufficient knowledge of the business to pass the “street cred” test with executive decision-makers; domain expertise.	80%	60%
The confidence to speak the truth to executives, peers and subordinates.	71%	80%
The ability to engage at team or organizational meetings without being asked for input.	86%	40%
An extremely high sense of quality standards and attention to detail.	57%	80%
The drive to create both questions and answers in a form others find accessible.	57%	50%

To discover if companies are hiring for roles similar to a data translator and if communication skills are becoming more important as experts suggest, I looked at job descriptions from Indeed.com for data analysts and various iterations which included domain expert roles such as sales operations analyst, marketing analyst and messaging analyst (for an online messaging company). For each title, I pulled 10 job descriptions. The job descriptions were all similar and, even with just 10 descriptions per title, I found strong themes and similarities across the descriptions.

I then compared the job descriptions to the characteristics of data translators. Table 1 summarizes the percentage of how many listings specifically identified data translator skills as either required or preferred for the roles. As expected, 100% of the posts required “sufficient analytical knowledge to communicate effectively” with the organization’s technical scientists.

Additional specific skills that related to the role of a data translator and are frequently mentioned in the job descriptions include collaboration, written and verbal communication skills, and leadership experience, which are mentioned in over 50% of job descriptions.

Overall, these job postings show that both the analyst job descriptions require and prefer the skills of a data translator with a strong emphasis on communication and leadership. Hence, these findings suggest that while the title of “data translator” and concept of “data translation” are not used in job postings, the specific skills of a data translator are key components of job postings for data analysts.

Conclusion

Companies need someone in the organization to perform the role of a data translator, an individual who possesses specific skills to bridge the gap between the technical and marketing strategy worlds. However, such individuals can be difficult to find. Companies can take a holistic approach to cultivating these skills across the analytics team as a whole. Alternatively, they might create a new position, the data translator, who can perform that role. 

Now imagine your company has built out its analytics team based on the overall skills needed to cover both complicated analysis and communication to executives. The analyst discovers critical insights to your marketing executives’ questions, the data translator understands the results and communicates them with effective visualizations and familiar lexicon. The executives now implement change based on these results, reinforcing the value of the data science team to the company.