Room to Grow: How To Manage Change While Continuously Changing.

In my last post , I outlined that the most efficient way to provide AI with the detailed context information it needs to deliver accurate and valuable results is to collect that information as it is created—whenever and wherever that may be. This simple statement implies significant changes. These changes will impact everyone in an organization, though most will be minor. Enacting this change is the biggest barrier to properly capturing context. Below are three related guidelines to help your organization embrace change at a pace that minimizes pain and maximizes value.

Externalize Descriptions

Managing change effectively involves finding ways to minimize its impact. One efficient method is to contain change. By creating an external database to describe the different elements of an engagement, teams can create, use, and modify a customized data framework without affecting the applications and processes that collect and generate the data. I'll explore other benefits of externalizing descriptions in future articles, but for now, it is the best way to create a risk-free environment for development and innovation.

Think Slow, Act Fast

In their book “How Big Things Get Done ,” authors Prof. Bent Flyvbjerg and Dan Gardner emphasize the importance of “thinking slow and acting fast” to minimize the risk of failure for large, complicated projects. “Thinking slow” means collecting and analyzing as much input from real-world experience as possible before committing to any major action or decision. “Acting fast” involves executing many small tests designed to provide insights related to the larger challenge. By collecting these insights, teams can better identify accurate requirements, develop better plans, and mitigate the fear of the unknown.

When developing a customized data framework, teams should not hesitate to create, delete, and combine descriptions—testing and trying different options as their understanding of requirements and operating processes evolves. By externalizing the descriptions, they can easily create copies and versions to iterate “on paper” as deeply and frequently as desired. This iterative process leads to insights that improve the data framework and inform future data integration requirements.

Search for Small Victories

The best way to test a data framework is to apply it to a real-world case. The quickest way to find the initial cases is to partner with teams that have a pressing need or strong interest in solving a specific problem the data framework can help address. The key term is “partner.” A partner has unique expertise that can provide valuable insights and additional requirements that contribute to the project’s success. By sharing the burden and benefits, partners will be more committed to the implementation, improving the likelihood of success. Moreover, a partner’s positive experience can create enthusiasm, making them champions of future changes.

The trick is not to be picky. The goal is to iterate quickly. Rarely does the perfect data framework use case appear at the outset. Instead, take the opportunities at hand. It may mean solving a smaller problem, partnering with a “less visible” team, or working on a less interesting project. Quickly delivering small wins, regardless of the use case, is a great way to build momentum and demonstrate the framework’s benefits. String together enough wins and change just becomes “how things are done.”

Interested in a specific topic related to preparing organizations for AI, the universal data framework for customer engagement, or tips and approaches to helping your organization change? Let me know in the comments below!