Data-driven Strategy in Data and AI

Fostering data-driven culture and mindset around Data Science and AI

Photo by JESHOOTS.COM on Unsplash

The data-driven culture and mindset is becoming more apt in present scenario with regards to what we do in our day to day actions. While data-driven strategy is extremely critical in realizing business value and impact, it is equally important to execute it for accomplishing the target set by any firm. There are many factors that can be considered.

  • Understanding “why data is critical” and more importantly “the right data is critical and not just everything” is becoming normal for most firms
  • From Data Science and AI standpoint, both “data defense” and “data offense” strategies balance the pendulum for driving success, hence an amalgamation of these two strategies are critical
  • Sponsorship has seen a maturity curve towards north considering the business impact and outcome realization by CXO communities
  • Adoption rate has been increasing

At the same time, there are potential challenges and it is important to follow some defined steps or methodology to make sure we are dealing with it in a structured manner. Enterprises across different sizes are following in different magnitude and approaches which are also different with regards to their industry, goal, hunger for success, adoption rate and many other parameters driving to respective business success rates.

Workforce-Culture Alignment in Data Science:

If we refer to the importance of how data-driven culture is defined as per Gartner’s reference of Workforce-Culture Alignment which is a multiplicative factor of knowledge, mindset and behaviour; the same reflects to perform in data science as well.

  • Knowledge in data science = employees understand some of the cultural attributes that senior leadership think and act for their firms to be successful
  • Mindset in data science = employees have the belief in them that the cultural attributes will make their firms more successful and they personally contribute towards that and are committed to upholding some of those actions that are laid down
  • Behaviour in data science = employees inculcate the cultural attributes into the way they perform their work and depend on it to navigate them through challenging situations

Data Science is a team sport and it is significantly dependent on this intervention of “Workforce-Culture Alignment in Data Science” in order to excel and accelerate the journey in Data and AI.

Balance between Data Offense and Data Défense Strategies:

“Data offense” typically focuses on growing revenue, improving profitability, increasing customer experience, increasing customer satisfaction etc. At the same time, “Data defense” usually focuses on reducing risk, lowering total cost of ownership, lowering maintenance costs, solving data problems internally within the organization and standardizing it, making data compliant with regulation etc. The CIO (Chief Information Officer) and CDO (Chief Data Officer) usually have tasks around data defense strategies while CAO (Chief Analytics Officer) will have the focus around data offense strategies. Ideally, CEO (Chief Executive Officers) are neutral to both. For a data-driven passion and culture within the organization, it is very important to form a hybrid strategy that encompasses both and forms an amalgamation of the two. Few very high level use case themes are captured in below diagrammatic representation as a point of view. This is not exhaustive and not accurate to the coordinates from relative importance perspective, however it aims at providing a direction to our thinking process.

The “D A T A – S T R A T E G Y” Framework can be described as per below:

Data-driven culture and strategy traverses through the following step of actions:

  1. DATA STRATEGY: Define goals, objectives

Defining data driven goals/objectives from top leadership is required. It is a good practice to have clear idea whether it is short term, medium term or long term and actions will follow accordingly. Entire organization gets aligned henceforth with a clear objective in mind.

2. DATA STRATEGY: Appropriate data identification

Identification of appropriate / right data is key to success. It reduces cost down the line if we know we don’t need all data or more data, instead very “specific” data that is required for downstream analysis and insight generation process. Additionally, “right” tools and techniques are also needed. It helps to build a robust design, architecture from scalability, maintainability standpoint.

3. DATA STRATEGY: Traverse through business KPIs

Determining business KPIs (key performance indicators) for success criteria is to be performed at an early stage. It could be by industry, by various functions such as Sales, HR, Marketing, Customer Success, Management / Leadership etc. This helps in narrowing objectives for better goal accomplishment. More precisely, right metrics / indicators should also be chosen. Because if we are not defining and capturing right metrics, then outcome will be away of what is expected and will impact customer satisfaction.

4. DATA STRATEGY: Understanding Analytics maturity

Analytics maturity assessment should be conducted prior to starting on a journey or any transformation initiative. This helps assess current AS-IS maturity level and that gives a clear picture about what to focus in future. In Data Science CRISP-DM end to end lifecycle, this helps tremendously in order to focus whether energy needs to be spent on Data analysis, data visualization, integrating “right” data into one place for better analysis, feature engineering, feature selection, model development, model evaluation, model deployment, model management and monitoring, version management and managing data and model drifts etc etc.

5. DATA STRATEGY: Formulate Strategy fostering innovation

Formulating strategy by fostering innovation helps. Novel methods for handling data, curating data, getting insights out of data helps a lot and for that a strategy has to be defined for value realization. If firms can potentially work on deriving tangible value from data, that would be a great step.

Some of the questions that comes to mind in the process are in the following areas, e.g. finding new revenue streams, improving and optimizing existing business models, creating internal efficiencies and monitoring the same for better control, maintaining regulatory compliance, increasing data literacy rate, building new products and insights, strategy for “X Analytics” (Customer Analytics, Marketing Analytics, Operation Analytics, CRM Analytics, Retail Analytics, Healthcare Analytics etc.), strategy for AI with Cloud, AI with Blockchain, AI with IoT, AI with AR/VR etc.

6. DATA STRATEGY: Hire the “right” Talent
Hire right data experts with appropriate “talents” in the team to perform every stage of operations in a successful manner.

7. DATA STRATEGY: Define Roles such as Data Stewards, Data Owners and Reward Talents correctly and timely

Defining roles such as data stewards and data owners are extremely important and critical. Based on that and other suitable roles, RACI matrix can be formulated. This helps in Data Governance.

Reward talents/teams for using data diligently, bring up the data motivation value chain within organization.

8. DATA STRATEGY: Analyze data and measure value realization

Use appropriate tools and techniques to analyze data. Explore data to understand patterns, correlations, meaningful insights better. Once we understand and prepare data better, then it becomes increasingly simpler to uncover value and realize tangible insights out of it.

9. DATA STRATEGY: Train your staff / employees / talents

Train your staff leveraging right methods, right models and solutions. Design thinking approach, understanding appropriate models for specific use cases, architecture to formulate different models and solutions can be trained. Upskilling, cross-skilling and reskilling are important aspects, but should be executed carefully based on need, outcome, usage, interest etc.

10. DATA STRATEGY: Enable continuous adoption

Adoption from top to bottom must flow and flow seamlessly. Provisioning right access on right data to right people (this is extremely critical as it can only be possible if appropriate SMEs, appropriate people/teams are aligned to perform this very important activity) helps build a robust ecosystem.

Making proof of concepts, proof of values simple and robust helps, instead of thinking very fancy, complex from the beginning. The whole purpose to adopt a culture where these are quick wins and provides motivation towards building a solid foundation and then momentum.

11. DATA STRATEGY: Generate data for business

Define how data is collected, what are the input data sources, which team is involved in collecting data and so on. If survey is conducted to generate datasets, then that has to be planned effectively.

12. DATA STRATEGY: Yield impact and value

Analyze the business outcome and value realization and measure it for success metrics. Document lessons learnt and keep improving on processes on where to improve and at what stage etc.

To summarize, data-driven decision making depends on several dimensions for success and enterprises must follow a structured approach to tackle challenges they face. Data Science value management follow some critical points mentioned above (12 points for each letter in the “DATA STRATEGY” from journey to success perspective). It is important to define strategy with innovation at every stage and execute it with efforts and actions around it.

Disclaimer: The postings here are personal point of views from my experiences, thoughts, readings from various sources and don’t necessarily represent any firm’s positions, strategies or opinions.

Data-driven Strategy in Data and AI was originally published in The Startup on Medium, where people are continuing the conversation by highlighting and responding to this story.

Related Articles

Building a high-performance data and AI organization

CxOs and boards recognize that their organization’s ability to generate actionable insights from data, often in real-time, is of the highest strategic importance. If there were any doubts on this score, consumers’ accelerated flight to digital in this past crisis year have dispelled them. To help them become data driven, companies are deploying increasingly advanced…

A beginner’s guide to data visualization with Python and Seaborn

Data visualization is a technique that allows data scientists to convert raw data into charts and plots that generate valuable insights. Charts reduce the complexity of the data and make it easier to understand for any user. There are many tools to perform data visualization, such as Tableau, Power BI, ChartBlocks, and more, which are no-code tools. They are very powerful tools, and they have their audience. However, when working with raw data that requires transformation and a good playground for data, Python is an excellent choice. Though more complicated as it requires programming knowledge, Python allows you to perform… This story continues at The Next Web


Your email address will not be published. Required fields are marked *

Receive the latest news

Subscribe To Our Weekly Newsletter

Get notified about chronicles from TreatMyBrand directly in your inbox