With most of the discourse on AI highlighting its strategic value, this article outlines where to invest in AI to utilise its full business potential.
PwC’s AI performance study found that just 20% of companies capture 74% of AI-driven returns, with the most “AI-fit” organisations delivering 7.2x higher AI-driven financial performance than the rest.
Despite these clear benefits, most organisations are stalled by isolated pilot projects, with little results to show for it. While AI can drive both growth and efficiency, value is only created where investment is disciplined.
Crucially, the leaders in this space split their AI investments between innovative and efficiency-based aspects. Instead of only focussing on how fast a system can cut company time and costs, they prioritise fundamentally evolving the nature and makeup of the business.
The study shows that leaders using AI as a “reinvention engine” are:
- 2.6x more likely to say AI has helped reinvent their business model
- 1.8x more likely to use AI to spot new “value” pools as industries converge
- 2-3x more likely to collaborate across sectors and build ecosystem plays for growth.1
To hit these dual targets, we suggest managing AI as its own portfolio, balancing varying risk and return profiles.
This is how we divide the portfolio:
Proven performance work (40-50%):
- Efficiency automation: Reduces cost, speeds up workflows and improves consistency. It provides a low-risk baseline, though its application is inherently capped as there is a limited volume of operational friction that can be optimised within any business.
- Performance enhancement: Sharpens decision-making quality and elevates strategic execution across core teams.
Innovation (60-50%):
- Product service innovation: Creates unique, bespoke tools tailored to specific customer or business requirements.
- Business model reinvention: Reassessing how value is created, delivered, and monetised.
- R&D acceleration: Increases the speed and quality of discovery and development, specifically in science-based and engineering-heavy sectors.
While proven performance work strengthens foundations, the innovation side maintains ingenuity to generate long-term success. To strike the right balance, performance work must pay for itself and reinforces the underpinning systems, while innovation must prove it can bend growth or R&D curves.
Sustainability of the portfolio
Like any other long-term investment, these processes must be mapped and checked through a comprehensive plan. It should use data to measure results, determine the utility of each category, and guide ongoing balance adjustments. Importantly, this data doesn’t need to range across a broad spectrum, but be high-quality, re-usable and centrally catalogued across priority applications.
PwC’s research shows that AI leaders win by aligning AI use with critical business outcomes and embedding them into daily workflows.
We recommend centring planning around these critical business outcomes:
Performance:
- Establishing a baseline and target metric.
- Designing a clear workflow redesign.
- Formulating a plan to scale across functions once proven.
Innovation:
- Defining a clear intended outcome.
- Confirming a defined proof point.
- Setting a time before you decide to scale or stop.
Moving forward
Overall, sector leaders don’t choose between efficiency and innovation, they fund both, holding each to a relative standard. By treating AI as a disciplined, split investment, we can move past the tired question of ‘is AI worth it?’, and ask the better one: ‘Which AI bets deserve more of our capital in this quarter?’
At 7DOTS, we help teams make these choices with clarity: define the outcomes, focus the spend, and scale what works. This leads to AI that delivers measurable value and sustainable momentum.
References
1* PwC, ‘AI performance study: Want ROI from AI? Go for growth,’ https://www.pwc.com/bm/en/press-releases/ai-performance-study.html <Accessed 05/10/2026.
