The Innovation Trap - How Dependence on AI Will Lead to Product Homogeny

You may be scratching your head after reading this title and thinking “AI is the biggest innovation of our lifetimes; how can it be a trap?”

You’d be correct on the assessment of AI being the biggest technological innovation of our lifetimes, but where this goes astray has less to do with the technology and more to do with how it’s being used, particularly by those responsible for evolving existing products.

This is my premise:

AI-powered product management tools that help collect, categorize, and summarize product, customer, and competitive insights will ultimately compress the scope of product strategy and lead to myopic feature-driven and parity delivering roadmaps.

Now allow me to walk you through how I got here, starting with a recognition of human nature.

Humans are wired to take the least path of resistance.

AI offers this up in spades. It takes care of time intensive information collection and summarization work and at computer speeds.

Why spend weeks on something that can be done for you in a minute?

In the product world this translates to collecting and synthesizing large amounts of data from numerous sources, detecting patterns from that data and then recommending paths forward, making decisions, and establishing strategies around the product's evolution and GTM.

Here's the hang up . . .

AI "slaps" at combing all existing and digitally available information and pulling together very logical insights and recommendations from a massive real-world database of knowledge.

Sounds good right? It depends on what you're using it for.

The key word above is "existing." The information that's been captured and accessible already exists, meaning it’s from the past and telling - at best - a story rooted in the expressed reality of today.

Now this can be helpful for certain applications, like competitive intelligence - what competitors are building, who they're targeting, where they're winning, etc.

In this use, holistic and well-rounded information clarifies your competitors' strategies and market positioning and can inform your own GTM strategy, hopefully a differentiated one.

Solutions from Crayon, Klue, and Glean serve up these types of insights and even try to detect early "signals" that inform on competitor strategy ahead of public announcements, but they're still staying in the lane of what exists today and reinforce the notion that you should be taking competitors head on – feature by feature, announcement by announcement.

In established markets, the pressure to ship fast and keep up with, or stay ahead of, competition is real - and often deemed necessary for surviving in fiercely competitive environments. Though these strategies usually have short lifespans, more on that in a minute.

Let's talk about another Product Management use case - customer feedback and feature requests.

Tools from the likes of ProductLift, Pendo, Canny, Savio, and Aha! (and many more) are great for automating the collection, categorization, and prioritization of feature requests and general feedback from customers, including ways to measure urgency and revenue impact.

These tools save a tremendous amount of time by doing the arduous task of data collection and initial synthesis. Data sources can be varied as well, spanning from direct customer input or requests to support ticket information and customer success conversations - all helpful sources to paint a more complete view of the customer's perception of what's needed from your product.

Product management and roadmap tools like the ones described in this article are an important part of the toolkit for those teams responsible for evolving their products and portfolios to the needs of the market – helpful in keeping tabs on competition, building out roadmaps, and prioritizing backlogs.

This is where Product teams miss the real market opportunity and get locked instead into a feature battle.

Too many teams are overly focused on feature parity in competitive assessments. To innovate, they're overly focused on customer feature requests and building onto and into the existing product flow.

They prioritize feature gaps on one side of the roadmap and on the other "innovative" new features that one-up competitors and that are loved by Sales and Marketing teams when trying to control market narrative on why your product is superior to the competition.

The problem is that only the biggest players with the largest access to resources win this game. They win with speed of new feature release. They win with marketing budget. They win with larger sales teams and partner networks.

Because of these advantages and the fact that they have the biggest roster of customers, they get to dictate the pace and narrative of what innovation looks like in your market.

The market leader drives the themes that are most advantageous to their business which protects or grows their market share. If you're not the dominant player in your market, you won't win by playing their game.

Today’s AI-powered tools only reinforce product innovation from the narrow lens of incremental development.

This is what I mean when I say AI will lead to more product homogeny.

Back to that thought on human nature and the wired instinct to travel the path of least resistance – AI is an enabler.

It enables team productivity and can surface areas for much needed development, but when everyone is using it the same way, and the context is the known competitive landscape, that gain ceases to be an advantage.

On the flip side, it's an avoidance enabler – avoiding the discomfort of doing the hard, longer-term customer discovery and research work.

AI feeds our human nature to jump straight to the solution and build by doing all the grunt work and placing all the insights and data justification at our disposal within seconds. But this is exactly the work that needs to be done to uncover real customer understanding and discover underserved problems or desires.

Don't take the BAIT.

Thanks to our hardwiring, AI has the potential to be the mother of all innovation traps . . . If I might be as bold to posit a new acronym to the strategy world: Big-Ass Innovation Trap (BAIT).

Using AI in the fashion described is fine if you understand the limitations and keeps its use to understanding the current state of your market, high-level trends, and general customer needs and relative feature gaps. It’s tremendously useful for initial research and surface-level understanding.

However, it's missing all the rich context and unspoken issues your customers are experiencing - and they may not even be aware of it or associate their challenges with your solution.

This is where the gold lies and AI is not going to be able surface that deep of an insight . . . or creatively design a new solution that doesn't exist today.

Most startups understand this. They believe they have found an underserved need in the market and have a unique, better solution to address it.

AI-first startups hold the same charter, but they're also leveraging emerging technology to completely rethink how the solution is architected, how the business ecosystem changes, and how users interact (or not) with the solution in novel ways.

And they build their business with intent and in a way that not only compounds value, but incumbents cannot replicate within their existing infrastructure, systems, and processes - though they'll try anyway and at a great expense.

In this sense, the term "AI-first startup" is becoming increasingly redundant.

Disruption also comes for companies using AI in the wrong way.

Bringing my point full circle, Product teams that lean on AI as their only research and ideation tool will execute on incremental product strategies and will be ripe for disruption – eventually losing customers and market relevancy.

The pressure to ship fast and "adopt" AI in all facets of work and product development will lead to less curiosity and more "me too" releases.

This is the interesting paradox brought on by AI:

The very tool that enables teams to build faster is also the tool that will potentially pull them farther away from their customers.

The difference between ok product teams and great product teams will come down to the ability to leverage AI - not to decide what to build, but support and execute on first principles design work that completely rearchitects how value is delivered to the customer.

Finding the friction to build a better product.

I think we can all agree that the ultimate goal is to grow your product's market share and user base and avoid competitive races to the bottom.

In order to effectively do this, you must do what good product teams have always done . . . step out of the “office” and talk to your customers with an open mind and insatiable curiosity.

Better yet, watch them interact with your product. Look for the friction. This is often easier to do when taking your biased-self out of the driver’s seat and let the customer take the wheel.

I’ll leave you with this one classic, probably overused, example of pain point-oriented product discovery: the bagless vacuum invented by Dyson.

Dyson is the story of the engineer who really didn’t like the experience of emptying vacuum bags and the issues they caused to performance of the vacuum itself, so he invented a new vacuum removing the bag component altogether.

He identified the friction, removed it, and replaced it with a better system. In doing so, he disrupted an established industry.

Engineers at the other vacuum brands were likely working on solutions to unclog the bag, or perhaps developing a bigger bag that required less replacement - all things that addressed the product deficiencies, but didn't completely remove them.

I like this case study because it demonstrates that you don’t have to completely reinvent the product to deliver a differentiated solution and experience.

Advice for staying ahead of competition.

Managing existing products and customers will always require some level of feature enhancements and competitive parity, but those activities have minimal impact to long-term market relevancy.

To truly innovate and carve out your own space in the market . . . Find your industry’s biggest point of friction for the end user and remove it.

AI has a role in this effort (and, who knows, it may eventually get better at spotting points of friction), but it’s not the product or the design goal itself. It’s an enabling technology that can help remove traditional product constraints and inefficiencies.

This is your goal – deliver a better experience by removing traditional constraints or barriers to use to differentiate your product and side-step the feature war . . . and use AI wherever it can support that.