- Data Strategy
– 18 Feb, 2020
The Human Side of Data Strategy and AI Adoption
An organisation wide data strategy is often comprised of similar components, focused on building a maintaining a technology solution.
Over the past 10 years this has commonly involved migration of data and applications to a Cloud provider. Service providers such as Amazon Web Services and Microsoft Azure have grown rapidly for more than a decade, with growth of 17% expected in 2020 (1). Cloud services now allow organisations to employ sophisticated analytics solutions and Artificial Intelligence at scale.
While a cloud solution forms the base needed to facilitate business functions technology requirements, there is mounting evidence that successful technology implementation is not enough for adoption. McKinsey reports that over 70% of transformation projects fail due to short-sighted planning (2).
With the rise in cloud compute capabilities, Artificial Intelligence is in reach of all businesses, why is it that over 80% of Data Science projects fail to make it into production and generate a return? (3) The following criteria cover 3 commonly overlooked aspects of business functions when building a data strategy.
Criteria 1 – Building the Right Team
Hiring for extremes; the ultra-focused and technical Data Scientist and managers who are overly business focused will lead to a mismatch between skills and communication. Data Scientists need a wealth of soft skills to support communication and understand business requirements. Managers require a base technical knowledge in order to focus the work of their team and remove unnecessary interference and interruptions to their work.
However, even with a balance of hard and soft skills employees should be technically specialised. Distinctions need to be made between Data Scientists, Engineers and Architects, expecting one employee to occupy multiple complex roles at once will greatly reduce their efficacy.
Beyond hiring internal upskilling is key. AI is a fast-moving field, state of the art algorithms are released frequently which requires redeveloping pipelines, updating and retraining models and retesting their effectiveness in production. Investing and developing your employees is key to retaining expensive talent, a concept that is not new.
Criteria 2 - Understanding Organisational Culture and Attitudes
From the top down the application of AI provides an opportunity for business leaders to greatly increase efficiency, revenue and other key metrics.
However, from the bottom up applying these technologies can fundamentally change or replace the responsibilities of many employees. It is common practice to assess an organisations current data landscape to inform planning, the same should be true for organisational culture. HBR found that one of the primary goals of Executives is to use AI to free workers time for more valuable work (4). This approach demands a reshaping of organisational structure and employee roles which will inevitably lead to resentment. Preparing employees for this shift with new skill sets and education will ease the shift.
Criteria 3 – Educating Everyone
Change is often addressed at the project level, which leads to reliance on limited business leaders and roles such as ‘Change Champions’. Factoring these stakeholders into a larger education plan including both executive education and wider general coverage will lead to improved adoption and generation of more use cases.
Harvard Business Review found that 37% of executives struggled to implement AI due to a lack of understanding from their managers (4). Education programmes should therefore be a mandatory part of implementing advanced technologies. Targeted learning and workshops for managers differentiated by business function, and for the wider employee base a combination of talks, online training and other channels should be used.
Find Out More
With the technology and data in place to implement a Data Strategy, the above human criteria represent some of the final steps to realising the benefits of Artificial Intelligence. To understand the full picture Chaucer are hosting an event “AI and Real-World Business” breakfast on March 26th. Attendees will get to see examples of successful AI application and receive guidance on Data Strategy and the journey it requires.
- https://www.techrepublic.com/article/85-of-big-data-projects-fail-but-your-developers-can-help-yours-succeed/ /https://venturebeat.com/2019/07/19/why-do-87-of-data-science-projects-never-make-it-into-production/
Data Strategy & Analytics Expert Chaucer's AI specialist delivering data strategies and capabilities for Fortune 500 organisations. He is passionate about driving data led digital transformation to enable organisations to realise the benefits of machine learning and holds both an MBA and MA in Educational Leadership and Management.