#7 Leading AI change: Step 7. Building upon the change
7/.?Building upon the change
The seventh process step of leading change from?Dr Kotter’s model ?finds that, whilst short term wins are crucial to transformational change, declaring these wins as if the change effort is complete can lead to regression in leading the change. Change leaders can be too quick to declare victory, whilst the change resistors can be quick to spot opportunities to resist the change.?Instead of declaring victory, leaders of successful efforts use the credibility afforded by short-term wins to tackle even bigger problems.
According to?research ?into global enterprises successfully transforming to AI-enabled organizations (see also previous articles below):
IMPLEMENTING & SUSTAINING THE CHANGE:
"Building upon the change"
1.??On the question of how innovation leaders build upon the change during the AI transformation they’re leading, some of the key findings were:
?? Aligning knowledge and learnings into roles & skills training – Within the organization
?? Creating a catalogue of AI capabilities – Made available for existing and new colleagues, initiatives and projects.?Documenting what technology is suitable to what problem, along with which experiments were not successful, and why.?
?? Taking proof of concepts and applying them to different use cases across the business - Showcasing how proof of concepts can work across departments, functions and business units, whilst also then adapting the approach in different projects i.e.?taking the best performing experiments and adapting.
?? Improving upon internal communications – By regularly reviewing communication assets, channels & materials, to support maintaining the momentum of AI transformation.
2.??On the question of?what worked, what didn’t & why?when?building upon the change during the AI transformation, key findings were:
What worked:
?? Continual AI experiments - Improving algorithms, accuracy & precision of?applications to support product development and way of working.
?? Building hybrid AI solutions - Gaining subject matter expert acceptance and building on top of that to drive model performance.
?? Consolidating knowledge across AI project team members - Accelerates learning and knowledge sharing across the enterprise.
What hasn’t worked so well:
?? Operationalizing applications across businesses – Where there is a scattered architecture; thus also diminishing knowledge sharing.
?? Process of documentation – Not indicated as one the most energizing tasks for many employees, and also not always clear on who should perform this task in the transformation process.
?? Standard communication material – AI developments and processes can evolve so quickly, therefore i.e. blogs, videos and white papers can quickly become outdated, and require optimizing.
? ?? Aligning data and knowledge sharing– Where developments are not sufficiently acknowledged or understood, a “perception gap” between “actual and realized” challenges can develop.
3.??On the question of what was different about?building upon the change during AI transformation compared to previous non-AI change initiatives, some of the key findings were:
?? Misinformation, false expectations and fear - Are more common in AI transformation, given the uncertainty around technological developments; and that AI concepts are not as easy to grasp.
领英推荐
?? Re-usability and scalability - Previous transformations may have focused on more departmental, functional, local approaches. Whereas there is a greater necessity and requirement to drive economies of scale and efficiencies with AI transformation investments across the enterprise.
?? Timing on when to institutionalize change - ?AI transformation is bigger than previous transformations, in that it impacts most departments and functions.?Therefore, deciding when to bring AI specific project teams into the organization is challenging i.e. where business goals and KPIs are not aligned.
?? Path is longer, so building trust takes time - The need to stay in touch, aligned and continually communicating is more crucial; although important to avoid information overload.
?? Speed of change in AI technology – The field is developing very quickly, therefore there are improvements in computing capabilities and new algorithms seemingly by the day. Keeping abreast of the latest developments can become a challenge.
?? Effort to build on change is substantially bigger –The difference in effort required now to build upon the change, given the complexity in capabilities, resource, and required stakeholder mix, is often misunderstood. Therefore, the need to have agility trained across the business is even more crucial.
?? Maintaining and preserving knowledge building & transparency – Colleagues need to understand, via their different functions, what is developing and how it differs from the past. Critical, given that applying the old models of business in many organizations is not sufficient for AI transformation.?
Up next...
Previous articles in this series: