Learning is provided in a variety of ways to reflect each colleague’s specific circumstances, from short, bite-size self-learning for those new to data, through 13-week programmes up to an 18-month apprenticeship meaning learners could choice a solution that meet their specific circumstances. To increase gender balance, a partnership was created with The Women in Tech committee to run Python coding classes aimed at women, bringing in 50 to the first class.
Career paths and appropriate learning resources have been defined for each role, up-skilling and re-skilling as required when colleagues progress or switch roles within data. These opportunities are available to both incumbent practitioners and also others in the bank who want to move into data careers. This is creating a highly-engaged community through the academy, which also supports social learning between peers. An internal social media platform called Workplace helps to bring peer groups together, share knowledge and attend monthly learning sessions. The Bayes Centre in Edinburgh also provides external perspectives.
Communities are central to the programme, agreeing and defining career paths by consensus, as well as ensuring the right learning culture is in place. To date, nearly £2 million has been invested with 598 data practitioners increasing their skills and 4,242 non-data colleagues up-skilling embarking on the programme. Diversity is an important dimension of this activity with 35% of the academy’s students female.
Internal benefits extend beyond the impact on recruitment and retention – consistent ways of working and collaboration between data practitioners and stakeholders are reinforced, ensuring data-driven innovation and impact for customers. The advanced analytics lab is already seeing benefits from the inaugural data science curriculum delivered.
Feedback from one graduate illustrates the impact of the data academy: “I learned two things yesterday and I already used them today on an innovation spike, getting me a better answer faster than I could have done before. Brilliant!”
Multiple examples of new solutions and models have emerged from the programme, including improving the process for closing accounts after bereavement, devising models which improve the detection of money laundering by mules and personalised minimum credit card payment schedules. A machine learning-based loan application model has yielded 17% more predictive power than the current model which could deliver £35 million incremental benefit over five years.
