In the first post of this series, we looked at why AI disruption affects startups, giants, and companies with product-market fit differently. We saw that structural forces—like scale economies, network effects, and capability stacks—shape who adapts and who stalls.
This post turns from why to how. The real challenge for product leaders is not predicting disruption but navigating it. While AI is reshaping every industry, companies that apply structured playbooks are better positioned to adapt and thrive.
Three frameworks offer practical guidance: the Wedge → Expansion Playbook, Jobs-to-be-Done, and Dual Transformation. Each gives product managers a lens to make deliberate choices about where to focus and how to scale.
Wedge → Expansion Playbook
Bessemer Venture Partners has long argued that successful SaaS companies grow by starting with a wedge and then expanding into adjacencies. This strategy is even more critical in the AI era, when competition is fierce and hype is high.
The steps look simple:
- Start with a narrow wedge that solves a specific job.
- Deliver a 10x better experience than existing alternatives.
- Integrate deeply into customer workflows so switching becomes painful.
- Expand into adjacent jobs once the wedge is secure.
Case study: Canva
Canva began as a lightweight design tool for non-designers. Its wedge was clear: make basic design accessible without needing Photoshop expertise. By proving its value in that niche, Canva gained adoption and trust. From there, it expanded into presentations, video editing, and now AI-powered design features. Each expansion felt natural because the wedge was strong.
AI example: Perplexity AI
Perplexity AI started as a focused “ask and answer” search engine. Its wedge was clarity and transparency in answers, unlike black-box search results. With traction among professionals and researchers, it is now expanding into Pro subscriptions and workflow integrations.
PM takeaway:
The temptation in AI is to build broad “copilot for everything” products. That almost always fails. The wedge → expansion strategy reminds us to validate value in one focused domain first. For product managers, this means identifying the job where AI creates an undeniable step-change value, then using it as the anchor for scale.
Jobs-to-be-Done (JTBD)
The Jobs-to-be-Done framework asks: what “job” is the customer hiring your product to do? AI changes both the nature and scope of those jobs.
Functional jobs
AI collapses multi-step workflows into one. Drafting, editing, and formatting tasks that once took hours can now be automated in minutes. This matters because functional jobs are where efficiency gains are easiest to prove.
Emotional jobs
AI products often deliver a sense of confidence or empowerment. When a marketer uses an AI assistant, the real value may not just be faster copy—it’s the feeling of being equipped with “superpowers” in their role.
Social jobs
Using AI tools can also send a signal to peers or managers: “We’re keeping up with innovation.” In fast-moving industries, social credibility can be as important as functional value.
Case study: Duolingo Max
Duolingo launched a premium tier called Duolingo Max, powered by GPT-4. Features like “Roleplay” and “Explain My Answer” give learners instant tutoring. Functionally, this helps learners correct mistakes in real time. Emotionally, it provides confidence that someone is guiding them. Socially, it reinforces Duolingo’s brand as an innovator.
Case study: HubSpot AI
HubSpot has integrated AI assistants across marketing and sales workflows. Functionally, it saves teams time on tasks like email writing or report generation. Emotionally, it reduces the stress of creative blocks. Socially, it signals to customers that HubSpot users are equipped with cutting-edge tools.
PM takeaway:
AI features should not be framed as “cool tech” but as ways to get customer jobs done better. When prioritizing roadmap investments, ask: what functional, emotional, and social jobs does this AI feature solve? The best AI products succeed because they align with all three.
Dual Transformation
Michael Tushman and Charles O’Reilly introduced the concept of dual transformation: organizations must reinvent themselves along two tracks simultaneously.
- Transformation A: Improve the core business using AI—automation, personalization, and efficiency.
- Transformation B: Build new AI-native businesses that may disrupt even your own core.
- Linking capability: A culture and leadership model that tolerates short-term cannibalization in pursuit of long-term survival.
Case study: Microsoft
Microsoft is a prime example. Transformation A: embedding Copilot into Microsoft 365, GitHub, and Power Platform. These AI upgrades improve productivity for millions of existing users. Transformation B: building Azure OpenAI Service, which allows developers and enterprises to build entirely new applications with foundation models.
Case study: JPMorgan
JPMorgan applies AI to optimize fraud detection and compliance—classic Transformation A. At the same time, it is experimenting with AI-driven investment advisory products that could open new lines of business.
PM takeaway:
For large firms, the trap is choosing only one path—either efficiency or disruption. In reality, survival requires both. Product leaders must guide Transformation A projects that improve the core while also championing Transformation B experiments that could eventually redefine the business.
Closing
AI disruption is here, but chaos does not mean randomness. Product leaders have playbooks they can apply.
- The Wedge → Expansion Playbook reminds us to focus on a narrow entry point before scaling.
- Jobs-to-be-Done ensures AI features solve real customer needs, not just showcase technology.
- Dual Transformation highlights how large organizations can balance incremental and disruptive change.
The winners in this era will not be those chasing every new model release. They will be those applying structured frameworks with focus and intent. For product managers, that means combining discipline with experimentation, and always keeping customer jobs at the center.
The future of AI is uncertain. But product leaders who adapt using these playbooks will be positioned not just to survive but to thrive.