Equipping Citizen Data Scientists For Successful Exploration

Brilliant business ideas can come from anywhere. More realistically, they can come from a little bit of everywhere — combining pieces from several talented people or teams. As more good minds work together on a project, the results only get better. So why is data analysis handled by such a small team at most companies? 

In a standard large organization, the process of translating data into insights and business practices is an exclusive one that has very few interactions. A director or executive poses a question to a data team, which analyzes all available information and puts together a report and some recommendations for another team to execute. Alternatively, the business leaders in marketing, finance, or sales are left to create their own reports — that runs the risk of being inaccurate. This arrangement perpetuates ineffective work silos and results in suggestions that are misunderstood, incomplete, or even rejected. 

An analysis process like the one described earlier removes the business team’s contextual knowledge from the process and limits the data team’s ability to explain how they reached their conclusion. Essentially, both teams involved are forfeiting their biggest strength. 

A simple solution to this broken workflow is to create a data environment with more diversity. That means including a broader range of information in the dataset. More visibility into different areas of a company means the analysis process can connect dots that otherwise would have been missed. Input diversity is key, but if all of that information is analyzed by one person or one team with a limited range of expertise, the value of that range is lost. The best way to generate insights from a diverse range of data is with a diverse range of analysts and business stakeholders, bringing experience from several different business teams to collaborate in tandem. 

To address this need for diverse analysis, and to maximize internal resources, we’ve seen an increase in citizen data scientists — professionals from outside the formal data organization who are data literate and can add important knowledge to data analysis. These positions are crucial to the spread of a data-centric culture and their business expertise can bring a lot of additional value to the decision-making process. With the right culture, a company can identify these data champions and train them to make the most of their data. 

Creating a Data-Centric Culture

The emergence of these citizen data scientists is not a trend of its own, it’s just one piece of a bigger data-centric movement at large organizations. In order for citizen data scientists to be effective, companies need to first set the stage for their success by establishing a culture of using data to make decisions. These new analysts are simply the last piece in the process.

A good way to build a productive data-centric culture is to start with a data platform that connects a company’s separate input sources and establishes a single source of truth for the entire company. From this connected platform, the data professionals can clean and model the data and share it broadly across the organization to generate insights. With a unified data platform, data teams can share their work and increase internal access to the datasets without worrying about inaccurate information affecting the analysis. This high degree of transparency and trust is what allows non-data professionals to confidently explore and find new insights. 

The concept of utilizing non-technical personnel to analyze data is not new, but until recently, it was very difficult to manage access to information. This increase in citizen data scientists is a result of the evolution of BI tools that simplify data processes. Traditionally, limited data access has made exploration complicated or impossible for business users to perform. These new tools allow data to be curated by a small team of data scientists and shared broadly for exploration that won’t impact the original source. In the last few years, we’ve seen an increase in the amount of data created. 

In the last two years alone, total data collection has doubled, and in the two years to come, it will double again. The new limiting factor in data analysis will be human capital. By making a deep pool of cleaned and modeled data available to more professionals inside an organization, a company can set the stage for the success of citizen data scientists and expand their internal analytical resources.

Identifying Citizen Data Scientists

Once a company has decided to establish a culture that will allow citizen data scientists to flourish, the natural next step is determining who they will be. There might be a few candidates who jump out initially — people who are always making requests for data or posing questions for the existing data to answer — but there are certainly more who can add value. Citizen data scientists don’t necessarily need to come from a math background or know any special skills, they just need to think in a unique way that will make a difference. 

One way to select citizen data scientists is to allow the data team to choose trusted members of business teams and train them to manage their own data requests. Another option is to provide the right tools to a wide base of employees and let them identify themselves. Beyond the obvious candidates, there might be a surprising number of business users who are eager to assist with analysis. With increased access to structured information, these people will naturally begin exploring and connecting dots. The new data explorers will proactively find insights assisted by their contextual experiences and knowledge. Since the culture of sharing has already been established, these insights can be checked by the data team and presented to the rest of the organization. 

Over time, these citizen data scientists will establish themselves as a valuable arm of the analysis process. They won’t serve as a replacement to formal data scientists, still lacking the skills to take unstructured data and model it for wider analysis, but rather as a complement to them, adding contextual knowledge from their field that the data scientist doesn’t have. Close collaboration between data professionals and the citizen data scientists will result in faster, more actionable business insights that will combine the most valuable aspects of both parties.

Getting Started with Citizen Data Science

The first time a citizen data scientist submerges themselves in analysis, they’ll need some guidance. Data analysis involves examining a lot of factors holistically, so it can be difficult to pinpoint one specific place to start. The best thing to do is to allow these new analysts to start with the data that is closest to their business expertise. They’ll want to examine the numbers behind reports or dashboards that they regularly use to make business decisions and study how that information is obtained. From there, they can branch out to related teams and projects to begin exploring relationships. As long as there’s faith in the curated datasets, these explorations will result in accurate findings.

With a model-as-you-go approach, citizen data scientists can comb through data and create visualizations of their findings with no technical knowledge. Modern BI tools allow the new generation of analysts to create drag-and-drop charts and graphs that can be shared instantly.

Once the citizen data scientists are comfortable with the analysis work, it’s time to set them free and let them start adding value. The easiest way for this new class of analysts to approach data analysis is simple: ask questions. The citizen data scientist brings external business knowledge to the analysis, so they should start their analysis by asking a question and then looking for the answer. Depending on the citizen data scientist’s expertise, that question will vary, but ultimately this process of asking questions and looking for answers is a simple way to add instant value. 

In time, the questions will become more advanced, pushing the limits of what that individual analyst can uncover. They will evolve from understanding basic data to recognizing patterns over time and eventually collaborating with the formal data team to build and analyze predictive models. 

Maximizing Value

To get the most out of these citizen data scientists, it’s vital to correctly understand the nature of their work. They shouldn’t be limited to handling reports that they might have previously passed off to the data team. It has to go deeper than that, beyond self-service and into self-exploration. They’re still taking some of the time-intensive, repetitive reporting work away from the data scientists, but their work is more valuable than that — they’re adding new expertise to the analysis and asking questions that data scientists might not have considered. 

The most valuable citizen data scientists are going to be autodidacts. With a little bit of training, they’ll begin expanding their boundaries and looking for new insights related to their specific area of expertise. Given the right structure from a CDO and a properly curated dataset, the most valuable citizen data scientists will pinpoint ideal outcomes of data exploration and ask the questions needed to get there. 

One of the most important lessons for citizen data scientists is that not every question has an answer. A lot of data exploration will result in unclear or insignificant findings. Even if they’re asking the right questions, the data might not be mature or complete enough to show the answer. Part of the growth at this position is realizing that data exploration is a scientific process, requiring several iterations of adaptation and retesting. Sometimes even under the best circumstances, an answer isn’t available. To add the most value to an organization, citizen data scientists need to be able to explore without fear of failure. The goal is to establish a process that will enable valuable data insights, not to force insights from every instance of data analysis. 

It’s important to encourage the concept of data exploration not as its own end, but rather as a means to business insight. While the discovery can be very rewarding, the translation of that insight into a value is where the company actually benefits. The citizen data scientist is in a unique position to maximize business value because they sit closer to the team that will take action on data-based recommendations. A data team usually has to relay their findings to a business team to take action, so an incomplete understanding of the data could result in loss of value. The citizen data scientist cuts this step out of the process, meaning better results for the business.

Empower Citizen Data Scientist Exploration

Modern companies are collecting data at an incredible rate, much faster than they can bring on formally trained data scientists to analyze it. When a company puts the right culture in place and gives this new class of analysts the right tools, the results can be incredible.

The best part about the citizen data scientist movement is that companies don’t need to search for them. They’re an untapped analysis resource inside an organization, with the right structure and the freedom to explore, they can provide insights that otherwise wouldn’t be found. This new resource is an easy way to broaden the overall data strategy and include deep knowledge from every business team in the data strategy. 

If you want to see how your team can use Sisense for Cloud Data Teams to increase the value of citizen data scientists, watch a demo or set up a free trial today.

Watch demo Start a Trial