There’s a theme that’s been infiltrating my inbox – and my conversations – for a little while now. It’s around the notion of “productive dialogue” in the sphere of Artificial Intelligence transformation. The narrative goes like this: early AI adopters are stuck in holding patterns of prototyping and proof-of-concepts, unable to drive measurable change throughout their business, due to a lack of conversations rooted in business value, meaning, and tangible next steps for the stakeholders involved.
Isn’t it fascinating to think that in the realm of “high tech” AI, what’s inhibiting success across the board is the productivity of our conversations?
With that said, in my three decades of experience working with clients across the globe in a broad range of technology transformation scenarios, I’ve learned one key secret about productive dialogue: impactful leaders successfully facilitate growth and change when they speak in the here and now, rather than pointing to lofty goals for the future.
This doesn’t mean you take your vision out of the equation — you always need your north star. It just means that you need to think about the implications of becoming an “AI powered enterprise” from different perspectives — and approach your company-wide interactions accordingly.
Tangibly, we’re talking about replacing “we plan to” with “we are.” It’s also about having a crystal-clear line of sight for every milestone on your ascent to AI growth, and clearly understanding how your current team will be impacted, and more specifically, what this might mean for their day-to-day. In short, your people need to feel like they’re the drivers of change for implementing AI across the organization — rather than individuals simply impacted by change.
Productive dialogue starts with a real business problem, followed by a logical framework for change that involves all the key players.
AI might be relatively new – but business transformation isn’t. I always recommend applying the same frameworks used for successful digital transformation and growth to implementing AI tools and software. The key piece here is, however, that the more people can see themselves as active contributors to the broader goal through specific, measurable activities — the more motivated they’ll be.
Here’s a quick framework to start out with, containing questions to ask yourself at each stage:
Step 1: Clearly outline your business problem
- What are your basic needs and what’s the desired outcome?
- Who are the key stakeholders?
- Why is now the right time to tackle this problem?
Step 2: Determine how and why should Machine Learning be applied
- Why is ML the best way to solve this problem?
- What are the necessary precursors to implement the ML solution?
- What’s the specific ML use case?
Step 3: Outline your metrics and expectations
- How will you know you are achieving success at each stage?
- How will you assess the level of sophistication needed for this ML solution?
Step 4: Build the operational plan
- What are the key projects that need to be done?
- What are the core milestones you should hit and the associated timelines for each?
Step 5: Learn and adjust as you go
- What new insights have we gleaned that should influence our original plan?
- Is the level of positive impact on my team accurately reflecting our original intentions?
- Is our ML model meeting accuracy benchmarks? If not, why are we falling behind?
The operationalization of AI to bring more models to market continues to be a challenge for leaders: much of this is because of the newness of the tools, data and skills required to make AI a success. As a result, it’s often regarded as a different project in its entirety, removed from the day-to-day goings-on of the organization. What would happen if we start to make it feel like any other project or change to tackle — that’s in-reach thanks to the team you have today?
Don’t paralyze your growth due to talent scarcity. Instead, identify upskill opportunities.
A recent report in McKinsey echoes the concern I’ve been hearing from our clients and individuals within our network: how do we begin to address the potential of Artificial Intelligence, if there’s simply not enough talent on the market to meet demands? In the report, a startling 87% of companies will experience challenges related to AI projects due to skills gaps within the next three years.
The good news? Platform providers are working hard to address the skills gap to empower businesses with sophisticated machine learning, without requiring a couple dozen research scientists and data scientists for one project. Stradigi AI’s platform, Kepler, for example, allows users with zero Machine Learning experience to leverage ML, including deep learning, to solve high-impact use cases. My team also conducted a study that included 20 interviews with business leaders throughout Canada and the United States, which found that one of the foremost needs of leaders — across a variety of industries — was encouraging growth for their teams.
Up until now, bridging innovation goals together with upskilling hasn’t been an obvious connection. As upskilling is centered around your current team, it’s also a way to reinforce productive dialogue: your AI transformation and innovation goals are achievable thanks to the experts already on your team. Because no one understands your business quite like the people who work in the trenches every day.
Confront change-making head-on, and be transparent about what it’s going to mean.
Implementing new software that has the potential to significantly influence the decisions your business makes isn’t a small change, even if you start with a small project. You can lay out all of the statistics and facts: it’s going to increase productivity and save you a couple hours a day; it’s going to allow you to use 70% more of your data; it’s going to give you better results. All of these statements are wrapped up into one concept: it’s going to create change. Your teams have a way of working, so be transparent about what it will encompass and the challenges they might confront on the way.
In short, instead of veering away from the reality that change is occurring, be real about it. And go back to the above framework for reference as often as possible, so your team can witness the progress they’ve made and the milestones they’ve already accomplished.
Examine the obvious gaps, and find the right partnerships to fill them.
Your team is your greatest asset – but It’s important to be realistic about upskill opportunities. If you don’t have essential experts, you’ll need to bring them in, either through a strategic partnership or through professional services.
A recent article by Andreessen Horowitz argues that, in the new realm of AI, the lines between pure service companies and SaaS providers are blurring: “AI companies appear, increasingly, to combine both software and services.” This allows organizations to lean on their software providers for advice, questions, coaching and guidance through immense change. And if your software provider doesn’t provide any services, it’s unlikely that they won’t be able to at least point you in the right direction for your needs.
After you build your coveted roadmap thanks to the framework above, take inventory: what do we have right now that we can leverage? This question is necessary on the human and technology sides. AI-transformation might impact every facet of your organization, but it doesn’t mean you need to overhaul everything in the process.
When it comes to implementing AI — the learnings along the way are as important as the outcome. Learning fast, adapting quickly, and maintaining the pivotal productive dialogue with your entire team along the way are the key ingredients for the next big AI success story.
Interested to hear more from Per? Follow him on LinkedIn.
Edited by Colleen McNamara