4 Strategies for Success: Deepening Relationships & Building Trust between Data Science and Business Teams

The gap between data scientists & business users when it comes to understanding each other’s language and value is real. Find out how your org can overcome it.

Mike Skypala
Mike Skypala
21 novembre 2022 4 min de lecture
Deepening Relationships, Trust and Confidence between Data Science and Business Teams

Business teams increasingly rely on the insights and models developed by data science or broader data teams. But there is still a gap between data scientists and business users when it comes to understanding each other’s language and value. This gap is created by the discrepancy between data scientists’ understanding of their worth on one side and business teams and how they perceive the involvement of the data science teams on the other side.

Those who work in business functions have become used to technology enablement across all aspects of their role. They rely on office productivity tools, systems, or apps for most of the processes they manage or support. Additionally, Data Science provides further insights and intelligence. Artificial Intelligence (AI) and Machine Learning (ML) capabilities offer extra decisioning automation.

This development is both exciting and concerning. Business teams are now making decisions using highly trained models rather than a mixture of judgement and insight. In some cases, those decisions are even being made for them due to complete automation.

Business users, who revel in the technology enabled world, have a high degree of trust invested in the models and other outputs of the data science team. This can lead to complacency and lack of curiosity when working on business questions. If the model fails to work, business users can feel threatened. Due to the lack of dialogue and understanding about the inputs and process for creating these models, business teams and data science teams struggle to achieve their goals.

You also find business teams who are less tech savvy, especially when it comes to working with AI and ML tools. Sometimes there is mistrust in the technology and concern of how to use it. Some business users might fear that their expertise is being downgraded. These business teams might wonder how decisions can be better based on data alone. To some people AI and ML still has an Orwellian feel to it.

Much of the Chief Data Officer’s and Chief Data & Analytics Officer’s time is spent on democratising data, meaning how do we enable everybody in an organization, no matter of their technical background to work with data confidently and make data-informed decisions. They also ensure that business value and outcomes are delivered from data engineering and data science to the business.

The goal is to have data science teams work with internal product teams or vendor solutions to equip the business user with the means to make decisions and act by creating their own models, alerts, and dashboards. For example, the marketing team being able to run and improve their own customer churn predictive models and better personalize messages, offers and content accordingly.

But this is not the reality in most organizations. Most data science teams still spend much of their time data wrangling and trying to reduce this time by changing the data architecture.

Data science is still embedded deeply in data quality and completeness issues, facing increasing legislation on responsible AI. They also face business users who often struggle to understand the work they do and are suspicious of the value and credibility that data science brings to the table. At the same time business teams expect a lot of the involvement of data science.

Here are four strategies to improve the relationship between data science and the business. 

Firstly, ensure you have a data analytics platform which enables data scientists to use their favoured tools and code, run and train models at speed and have a model governance and management environment which provides full transparency and lineage. This ensures that the data science team is free to innovate and can develop models using best practice. Once trained, the models can be deployed at speed and scale, and the business can start measuring the value, which builds trust and confidence.

Secondly, build a continuous dialogue and knowledge transfer programme between the data science team and the rest of the business. Glossaries are always useful. A library of data products and models and the business outcomes they have delivered is even better! The most advanced organizations invest in initiatives to define responsible AI & ML means, how it will be executed, and what that means for each business user. As legislation on responsible AI is rolled out, business users will want to know not only if their AI and ML models are compliant, but also how they are synchronized to the values and purpose of the business.

Thirdly, transparently share a roadmap of what is being built and delivered around AI and ML. This enables the business users to challenge and build on the importance, relevance, and impact of planned initiatives. Also, data products require an internal launch just as much as those emanating from the product teams. The roadmap should always contain the internal comms and expected business outcome.

Fourthly, ensure your data science teams prioritize efforts to involve business users deeply in model creation, business rules and the key principles and methodologies which drive the models. Co-collaboration breaks down barriers and improves shared understanding, as well as improved models!

As data science becomes more pervasive in organizations, the actions above, plus the Chief Data Officer and Chief Data & Analytics Officer having seats in the executive leadership team, will be critical for ensuring effective, trusted, and durable AI and ML deployment.


À propos de Mike Skypala

Mike Skypala is an Industry Expert specialising in Retail and Consumer Packaged Goods (CPG) for Teradata. He has over 30 years experience working in commercial and analytical roles, having started out at Procter & Gamble before moving into consultancy and data analytics solution providers. He was instrumental in expanding Tesco’s customer data to establish the insights and personalised media and promotion programme with CPG suppliers and has worked extensively in Europe, South Africa and US in advising how retailers should collaboratively and responsibly share data with suppliers to optimise customer loyalty, sales and profit.

Voir tous les articles par Mike Skypala

Restez au courant

Abonnez-vous au blog de Teradata pour recevoir des informations hebdomadaires

J'accepte que Teradata Corporation, hébergeur de ce site, m'envoie occasionnellement des communications marketing Teradata par e-mail sur lesquelles figurent des informations relatives à ses produits, des analyses de données et des invitations à des événements et webinaires. J'ai pris connaissance du fait que je peux me désabonner à tout moment en suivant le lien de désabonnement présent au bas des e-mails que je reçois.

Votre confidentialité est importante. Vos informations personnelles seront collectées, stockées et traitées conformément à la politique de confidentialité globale de Teradata.