Before You Start an AI Strategy, Make the Case To Stop

Estimated read time 8 min read

Helping clients get ready to integrate generative AI, I’ve noticed something interesting.

After walking through use cases and how the technology could impact the workflow across business functions, I find generative AI doesn’t reduce their need for resources. Instead, AI more likely adds new capabilities to the team – things the team usually doesn’t have time to do because they spend so much effort on content creation.

For example, generative AI can assemble useful topic maps and structured taxonomies for an existing corpus of content. These maps work extraordinarily well for auto-tagging future content and spotting holes in existing content categories. Since the team typically hasn’t made such a map, AI’s integration doesn’t make their existing work more efficient. It just makes a new task possible.

The integration of generative AI really requires a new commitment of time, making the initial challenge different.

Given adding new resources is an exceedingly rare option, the real question to answer is: “What are you going to stop doing to start doing the other thing?” 

Building the ‘stop’ business case

If you ever want to prove to your boss that marketing is the hardest thing to do in business, tell them it’s the only function built on the ability to change into unknown things.

Imagine if the finance group’s operational strategy changed from quarter to quarter. One quarter, they use Generally Accepted Accounting Principles (GAAP) standards to recognize revenue. For the next 90 days, they switch to International Financial Reporting Standards (IFRS). The quarter after that, they must switch to a newly invented standard. How might their ability to change be measured?

Marketing and content are functions built on a foundation of change. The activities have been – and always will be – fluid. Marketing success can’t follow the “right” path. It must come from the new best direction as the old paths end. Those intersections make the most interesting parts of being a marketer.

A funny thing happens at the end of almost every advisory session – after the group has spent the day talking about a new marketing initiative or the impact of the road mapping generative AI’s integration into the workflow. Someone (usually someone who has been quiet, though sometimes it’s me) says something like this: “Um, we should probably align our expectations with reality.”

The room becomes quiet. Heads nod. The realization sets in: This is going to be hard. Real people will have to do all this stuff. As the heads nod, the inevitable objections arise:

  • “We still have to support the sales team with the materials they need.”
  • “We still have to publish those four newsletters every week.”
  • “We still have to update the customer resource website.”
  • “We still have to launch that new product website next quarter.”

Having watched these arcs of realization over and over, I start these sessions – before ever getting into the new initiative – by asking about all the things the content team does. Then, I ask why they do each thing and list their juicy rationalizations.

Near the end of the day, when the heads nod about expectations and reality, I trot out their list and ask, “Which of these things can you stop doing?” People look at each other. Uncomfortable laughter ensues as someone says, “None of it. Senior management will want us to keep doing it all.”

The business case you must build isn’t why you should do the new thing. It’s why you should stop doing the old thing. The ‘stop’ business case is just as important as the ‘start’ business case.

Business professor Michael Porter famously says, “The essence of strategy is choosing what not to do.” Whenever you tackle a new marketing initiative – the integration of AI, a new content hub, a resource center, a customer help publication – you ensure you won’t do it if you don’t address what you’ll stop doing.

You can learn from media companies

When it comes to content-driven marketing strategies, you can learn once again from media companies.

Media companies understand two important things about launching content products: 1. It’s exceedingly difficult to know what will succeed with audiences, and 2. You must be willing to stop something as easily as you start it.

Think about the TV pilot season. Major television networks invest in at least one episode with A-level talent for 12 to 20 shows. The goal is to see if sample audiences respond positively. Only about 20% of pilots ever air on a network.

These days, the competition for audiences is fierce. Something like 599 scripted television series aired in the United States in 2022. Of those, about 150 will be canceled in a year. Another 50 or so might be viable for the long run. Few will run more than a few seasons.

And no one knows which, if any, of the shows will be winners.

Media companies aren’t investing millions of dollars to throw the proverbial spaghetti against the wall. They understand tomorrow’s mega-hit may come from surprising origins. They do the research, dive in with creative talent, and take the time to develop and test the product.

That’s the opposite of how most businesses develop bigger content or marketing properties. Too many companies strategize, come up with ideas, winnow them based on budget, and invest in the story or format available resources can support.

Imagine if a media company developed a TV show based on budget, not the idea. They’d check what talent would fit the budget. Then, they’d figure out what kind of show their cameras and technology would allow them to shoot. Only then could they figure out the story.

That approach doesn’t produce the opportunity for a hit.

More importantly, though, the shows would be designed to start and stop easily.

Compare that to a common scenario in today’s product marketing world.

Recently, a marketing leader at a large technology company told me they decided to add a new content platform. They sought recommendations on the technology to create its content, including the topic and related blog posts. They needed AI to drive it because the content platform would be another responsibility for the already busy team.

I asked him about their 10-year-old blog with declining traffic, search results, and editorial quality. I asked why they would continue that and add this new thing.

The marketing leader explained: “While my team creates the content for that platform, the demand-generation group manages the platform itself, and they want to keep their platform. So we have to support both.”

You can see this story is destined not to end well.

Few companies have the resources or checkbooks to develop multiple blog platforms or produce unlimited content. But brands can develop and decommission content products.

But to do that, you must ask better questions: “Do we really need to implement that new hub or resource center into our enterprise CMS system before we launch a prototype and test it?” Or “What activity is producing results that, if it ended tomorrow, no one would really notice?” “Can we redirect the creative talent to something cool and innovative instead of more infographics for sales?” Or “How can we build a better, more efficient production studio approach for testing stories, formats, and audience engagement?”

It’s no one’s fault when the story, platform, or format stops resonating with audiences (or never gets started). As Captain Jean-Luc Picard once said in Star Trek: The Next Generation, “You can commit no mistakes and still lose. That is not weakness. That is life.”

Why not free yourself and your team to stop doing things that aren’t working?

What will you stop in the new year?

In the new year, you likely will be asked – or take it upon yourself – to figure out how to integrate generative AI into your marketing. I strongly advise you not to look at it as an opportunity to figure out how much more you can do of the stuff you already do.

Rather, look for opportunities to use generative AI to create something new, build a more efficient process, add research capabilities, or bring on a virtual team member to create derivative content. In almost all these scenarios, you’ll need more time, more resources, and different skill sets. This is fantastic. But before you add those things to your existing plate, just figure out what you’re going to STOP doing.

It’s your story. Tell it well.

Subscribe to workday or weekly CMI emails to get Rose-Colored Glasses in your inbox each week. 


Cover image by Joseph Kalinowski/Content Marketing Institute

You May Also Like

More From Author

+ There are no comments

Add yours