CIO&Leader spoke with Vivek Mehra, CTO, CSO & Co-Founder of Onlygood.ai, to decode how an AI-native enterprise navigates infrastructure, talent, governance, and measurable success to truly embed AI in its DNA.
AI has moved well beyond the hype cycle. Yet, transitioning successful pilots into full-scale, ROI-driven implementations remains a significant challenge. According to an IBM study, only 25% of AI initiatives have delivered the expected ROI, and just 16% have been scaled enterprise-wide. A parallel report by Informatica reveals that although GenAI investments are rising, two-thirds of enterprises still struggle to operationalize their pilots.
CIO&Leader: How do you define success when transitioning an AI initiative from pilot to production?
Mehra: Success in transitioning an AI initiative from pilot to production should be measured by clearly quantifiable KPIs (Key Performance Indicators). These KPIs should include how much time, effort, and costs are saved from the initiative, and the scale of impact across the enterprise, its partners, suppliers and customers. Another key metric has to be a measurable impact on the product or service quality, in turn impact customer satisfaction and sales growth.
CIO&Leader: What are the core pillars of your current enterprise AI strategy?
Mehra: Onlygood is an AI-native enterprise, and the key pillars for our enterprise AI strategy are:
- AI-enabled design of processes across all our product development, technology execution, sales, marketing and ESG divisions
- Skilling/up-skilling of teams on the latest tools and technologies, including our partners and vendor teams
- AI-led innovation for product features and modules
CIO&Leader: What key AI use cases have successfully moved into production, and what measurable impact have they delivered?
Mehra: The core use of AI in our product is analytical insights – how to help organizations understand their carbon emissions and ESG scores across their supply chains. This helps to create a real decarbonization pathway, design low-carbon products and services, as well as align supplier net zero goals with those of the enterprise.
CIO&Leader: What infrastructure or architectural changes were necessary to scale AI effectively within your organization?
Mehra: To build an AI-led organization, we began with hiring the right team – that is adept at working with new AI-tools and technologies. AI-driven Innovation also requires active collaboration across business, product design, and technology – this has required us to put in place strong collaboration tools. We have also designed a configurable, scalable modular, cloud-agnostic architecture, robust develops processes for quick releases and industry, or client-based customizations.
CIO&Leader: What are the biggest challenges you’ve faced in operationalizing AI, and how have you addressed them?
Mehra: The plethora of choices available requires that much more effort in identifying the differentiating aspects of each. Choosing the right AI-tools that can be adapted and optimized for sustainability use cases and integrated into our roadmap requires consistent research, experimentation and innovation.
CIO&Leader: How are you preparing your workforce for scaled AI adoption, and what organizational shifts have been required?
Mehra: Everyone in our team is actively using AI in their job functions. We provide continuous training and upskilling courses for the teams, and have incorporated use of AI in their KRAs.
CIO&Leader: Looking ahead, what does your AI roadmap over the next 12–18 months look like — especially in terms of GenAI or foundation model deployments?
Mehra: Strategically, we expect to continue to expand our use cases based on GenAI for all our product releases. Additionally, we will be building our own models on top of our chosen LLMs.