In our latest Arkaro Insights Podcast, we explore a question on every innovator’s mind: where can artificial intelligence genuinely enhance Jobs to Be Done research, and where does it fall short? Scott Burleson, Chief Product Officer at the AIM Institute and author of The Jobs to Be Done Pyramid, provides a practitioner’s perspective that should temper both the hype and the dismissiveness surrounding AI in innovation.
Watch the full discussion:
The Current Reality: AI as Research Accelerator
Scott’s assessment of AI’s current state in Jobs to Be Done work is refreshingly pragmatic. For practitioners who traditionally spent hours reading medical journals before interviewing surgeons, or trawling through customer forums to understand market contexts, AI has become an invaluable tool for secondary research.
Rather than manually processing dozens of technical papers or customer discussions, AI can synthesise that information rapidly. This acceleration matters not because it replaces thinking, but because it frees practitioners to focus their expertise where it genuinely adds value.
Furthermore, the same capability extends to the divergent phase of innovation research. When you need a comprehensive list of potential customer needs before conducting primary research, AI excels. It can generate that complete set of possibilities far more quickly than traditional methods—providing a foundation for more rigorous investigation.
The Critical Limitation: Prioritisation Requires Real Customers
Here is where Scott draws a clear line. Whilst AI can suggest what customer needs might exist, it cannot yet reliably tell you what matters most, or to whom.
This is not a temporary limitation that will disappear next quarter. It reflects a fundamental truth about Jobs to Be Done methodology: prioritisation demands understanding not just needs, but their importance and current satisfaction levels within specific customer segments.
As Scott notes, “When you also go to prioritise, you have got to be really clear about who you are targeting.” During the divergent phase of gathering needs, you cast the net wider. However, when it comes to understanding what truly drives customer decisions and investment, you must focus sharply on real customers through proper research methodology.
This aligns with what experienced practitioners understand: the discipline of well-defined job statements exists for good reason. AI can generate plausible-sounding needs statements, but it lacks the contextual understanding to distinguish between what sounds right and what reflects actual customer priorities. That requires human judgement informed by rigorous research.
Moreover, even in the divergent phase—where AI shows the most promise—a certain amount of abductive reasoning remains essential for completeness.
Consider the emotional jobs that Scott’s pyramid framework addresses: how does washing powder help someone be a responsible parent? How does specifying redundant machinery help an engineer maintain their identity as the person who keeps the plant running?
These connections require the inferential leap from observed facts to best explanation—something AI struggles with, particularly for emotional and identity-level jobs.
Human Expertise in Jobs to Be Done: Enhanced Not Replaced
The future Scott envisions is not one where AI takes over innovation research, but where expertise combines with powerful tools.
The twenty-year statistics professional who masters AI as part of their toolkit will outperform the novice using the same technology. Domain knowledge, research rigour, and critical thinking remain essential—AI simply makes these experts more efficient.
This perspective should resonate with anyone implementing AI in B2B contexts. The technology does not eliminate the need for expertise. Instead, it amplifies the impact of skilled practitioners whilst exposing those who attempt to substitute tools for understanding.
The Case for Specialist AI Tools
There is clearly a niche for specialist AI software designed specifically to support Jobs to Be Done innovation—but only if developed with insights from experienced JTBD practitioners.
A walled garden approach, similar to NotebookLM’s capability to analyse large amounts of curated content, could prove valuable. Such systems could be populated with industry-specific knowledge, validated job statements, and research from previous projects.
This creates a foundation that respects the discipline of proper JTBD methodology rather than treating job statements as merely plausible phrases.
Scott’s example of understanding jobs and potential desired outcomes in surgical procedures illustrates this potential. AI can synthesise extensive medical literature and procedure documentation far more rapidly than human researchers.
Yet even here, the tool supports rather than replaces the practitioner’s expertise. The JTBD expert still brings the abductive reasoning necessary to connect technical procedures with functional, emotional, and identity jobs.
Industry-Specific Applications
Consider what this means for innovation in agriculture, food, and chemicals industries.
Understanding why farmers hire equipment, why food manufacturers select ingredients, or why chemical processors choose specific solutions requires deep contextual knowledge. AI can accelerate data gathering and pattern recognition, but it cannot replace the practitioner who understands the identity and emotional dimensions Scott describes in his pyramid framework.
The Collaborative Future of AI and JTBD
Scott speculates that eventually AI systems might be capable of synthetic user prioritisation if educated sufficiently well. But even if that capability arrives, the fundamental requirement remains: someone must do the educating.
Someone must bring the domain expertise, customer understanding, and methodological rigour that makes the system useful rather than merely plausible.
This is where collaboration becomes essential. The most effective approach combines AI’s processing power with human expertise in customer research, the rigour of proper Jobs to Be Done methodology, and the contextual understanding that comes from years in specific industries.
As one LinkedIn commentator observed in response to similar content: “AI can widen perspective, but it is human judgement that decides what matters. The future of innovation is collaborative, not competitive.”
The Arkaro Perspective on AI and Innovation
This collaborative philosophy aligns precisely with Arkaro’s approach to innovation and change. We do not believe in handing over slide decks and leaving organisations to implement alone.
Whether you are implementing Jobs to Be Done methodology, deploying AI tools for innovation research, or transforming commercial capabilities, sustainable success requires working alongside your teams.
Through our Understand, Co-create, Enable, and Sustain approach, we ensure that powerful methodologies and technologies become embedded capabilities. This matters particularly when implementing frameworks like Jobs to Be Done that require both technical rigour and cultural change.
Consider Scott’s warning about AI hallucinations (occurring 10 to 15 per cent of the time, according to recent research). Teams need not just the tools, but the critical thinking capability to recognise when outputs are plausible rather than accurate.
This is not something you can outsource to technology. It requires building organisational capability through hands-on coaching and implementation support.
Building Sustainable Innovation Capabilities
The organisations that will succeed in combining AI with methodologies like Jobs to Be Done are not those racing to adopt the latest technology.
They are those building the collaborative capabilities, critical thinking skills, and methodological rigour that enable them to use AI effectively whilst avoiding its pitfalls.
This aligns with our broader perspective on innovation strategy and process. Success comes not from technology adoption, but from building the organisational capabilities that allow you to extract genuine value from that technology.
We do not just coach—we get on the pitch with you.
Related Reading from Arkaro
Understanding Jobs to Be Done Fundamentals:
Jobs to Be Done: The Missing Link in B2B Innovation – Explores Scott Burleson’s pyramid framework revealing why B2B purchasing decisions involve identity and emotional considerations as much as functional needs, and why even the most loyal customers maintain transactional relationships with brands.
Implementing Change Alongside Technology:
AI in the Workplace: Why Most Change Programmes Still Fail – Examines why AI implementation requires people-centric change leadership using the PEOPLE framework, addressing the reality that most technology transformations fail through lack of buy-in rather than technical limitations.
The Human Factor in AI Implementation: Lessons from Aviation for Industrial Leaders – Aviation expert Barry Eustance reveals why AI implementation requires human oversight and critical thinking, offering crucial lessons for industrial leaders navigating AI adoption safely.
Building Innovation Capabilities:
Collaborative Culture: How Silos Destroy Product Launches – Examines how cross-functional collaboration enables teams to understand the complete picture of customer jobs across technical and commercial perspectives.
Value Proposition Mistakes: Why 80% of B2B Products Fail – Demonstrates how starting with customer Jobs to Be Done prevents critical failure modes including misunderstanding customer needs and insufficient differentiation.
Applying Jobs to Be Done in Practice:
B2B Win-Loss Analysis for Cross-Functional Growth – Shows how Jobs to Be Done methodology transforms win-loss analysis from routine data collection into revealing the fundamental progress customers seek rather than just stated requirements.
Beyond Volume and Revenue: The Power of Needs-Based Customer Segmentation in B2B – Demonstrates how understanding distinct customer jobs drives more effective segmentation than traditional volume-based approaches.
Understanding Customer Needs: Essential Input for Change and Innovation – Explores why understanding “what’s going on” with customer needs is critical input for any successful change and innovation project, and how to secure these insights effectively.
This blog post draws from the Arkaro Insights Podcast conversation between Mark Blackwell and Scott Burleson. To hear the full discussion, watch the complete video. To learn more about Scott’s work and The Jobs to Be Done Pyramid, visit thejtbdpyramid.com or connect with him on LinkedIn.
Ready to implement Jobs to Be Done methodology whilst navigating AI’s opportunities and limitations? At Arkaro, we help organisations in agriculture, food, and chemicals industries build innovation capabilities that combine proven frameworks with emerging technologies. Our collaborative ‘do it with you’ approach ensures these methodologies become embedded in your organisation’s DNA, not just this month’s initiative.
Contact us for a conversation about how to build sustainable innovation capabilities, or connect on LinkedIn to continue the discussion.
