Stepping into the CPO Role at Flomni
Joining Flomni as Chief Product Officer was a big challenge. I was tasked with leading product strategy for a platform that mixes omnichannel communication and AI automation, built for medium-to-large B2B companies. Flomni's core idea was strong: bringing together different channels and using AI to handle many interactions automatically. But turning that idea into a product that worked well, could grow, and was reliable meant I first had to really understand the situation on the ground before setting a clear path forward.
What Kind of Product Work Was Needed?
One of the first things I needed to figure out was what kind of job this actually was. Where could product effort have the biggest impact for Flomni and our B2B clients? Looking at different types of product work, and focusing on delivering real business value, I saw it was a mix:
- Feature Development: We handled many channels and AI topics. But I questioned if the features were deep enough. Were there big gaps in important B2B workflows, like tricky support cases, or better CRM connections? It seemed we needed things like better analytics or more control over the AI.
- Growth: Getting new customers is important, but for us, scaling also meant making the product easier to use. How could we make setting up omnichannel flows simpler? Could we improve self-service tools for AI setup to help clients get value faster?
- Product/Market Fit Expansion: Flomni worked with medium-to-large companies, and some needed on-premise setups. I wondered if this stretched our main SaaS system too thin. Was the AI general enough, or did we need to focus more on specific industries?
- Scaling Work: The core promise depended on smooth integrations and reliable AI. This meant tackling underlying tech issues, making the platform stable under heavy use (especially voice), managing AI costs, and meeting the tough security and compliance needs of big companies.
Pretty quickly, I realized that while building new features or expanding was tempting, the biggest impact would likely come from scaling what we already had and growing its use by improving the core experience. My gut told me we needed to make sure the core product was solid before expanding too much.
Strengths and Strains
As I started digging into how Flomni actually worked, I saw a common picture for fast-growing B2B tech companies. It highlighted areas where our ways of working needed attention:
- Omnichannel Breadth vs. Depth: Connecting many channels was a definite plus for us. But the experience wasn't always consistent across them. It took a lot of ongoing effort just to keep these integrations working and improving.
- AI Potential vs. Practicality: The AI could handle many topics, which led to impressive automation numbers (75-90%). However, setting it up, training it, and making sure it was reliable and safe for specific client needs was proving difficult. In B2B, accuracy and control are everything, needing more than just basic AI capabilities (similar to challenges I'd seen elsewhere, like in legal tech).
- HR Focus: Our HR tools (for recruitment/onboarding) were a solid niche, but it also made me question our focus. Was it a separate product or just a feature set? How did its specific needs impact the core system design?
- Customization & Integration: Being flexible, including offering on-premise setups, helped us land large clients, but it created drag. Each significant customization added technical debt and slowed down core platform updates. Finding the right balance between custom work and a scalable SaaS model was clearly crucial.
- Cost Efficiency Promise: Our sales pitch relied heavily on cutting client costs. This meant our own internal efficiency (how we ran AI, our infrastructure) directly affected our ability to deliver this value profitably. We needed to get smarter about optimization, perhaps using more open-source tools where it made sense.
The standard product path showed steady growth, but I saw that complex enterprise deals often pushed the limits of our customization capabilities and support teams. This really highlighted the need for better feedback loops between Sales, Support, and Product – something I knew we had to improve.
Voices from the Team and the Market
Talking with the internal teams and gathering feedback from sales and customer success was essential for understanding the real picture. Here's some of what I heard:
- Integration Hurdles: People consistently mentioned that adding new channels or connecting deeply with custom client systems was often complex and took too long.
- AI Configuration: There was a clear need from clients for easier ways to guide the AI, set boundaries, and add specific knowledge without needing engineers.
- Scalability Bottlenecks: Under high usage, certain parts of the system or AI services would sometimes strain.
- On-Premise Overheads: Supporting on-premise installations required a lot of dedicated resources, pulling away from the main SaaS model.
- Documentation & Self-Service: We clearly needed better guides and tools so clients (and our own teams) could manage setups more independently, giving teams more autonomy.
Forging a Focused Product Strategy
Putting all this together – the assessment, the operational strains, the team feedback – it became clear we were trying to do too much at once. Expanding with new features, deeper customizations, and maybe even new industries was happening without ensuring our core offering was completely solid and scalable. Like many companies, we were spread too thin.
It felt like we had a choice: keep stretching, or refocus. We chose to prioritize making the existing platform stable, high-quality, and efficient. We landed on a strategy centered on three core ideas: focus, quality, and speed.
- Narrow the Focus: I decided we needed to temporarily slow down on adding new channels or major new AI features. Instead, we'd double down on the most valuable things we already did (like customer support automation) where we had proven success. The thinking was: improve quality for our best clients first, keep them happy, and then find ways to grow revenue later (through upsells and reduced churn).
- Raise the Quality Bar: We needed to invest seriously in platform stability, performance (especially for real-time voice and high message volumes), and security. This meant improving monitoring, automated testing, and how we handled problems, and making sure AI results were consistently reliable and controllable. Quality had to come before rushing things out.
- Reduce Technical & Configuration Debt: To eventually move faster, we first had to clean up. This meant fixing tricky integration points, building better tools for non-technical users to manage AI setup and safety rules, creating standard ways for common customizations to reduce one-off engineering work, and improving tools and processes for our own developers.
This "strengthen the core" approach, similar to the idea of "making the basics great," wasn't about stopping growth. It was about setting us up to grow sustainably. By improving the base platform first, we could make it easier and faster to add features, work with partners, and onboard clients reliably down the road. It meant focusing on making the current product work exceptionally well for our target B2B clients before chasing too many new, shiny ideas – a common trap I've seen companies fall into when they neglect their core potential.
Conclusion
So, my first year was largely about shifting our focus from trying to do everything (breadth) to doing the most important things really well (depth). This meant strengthening the core product and maturing our internal processes.
For a B2B platform like ours, aimed at large companies with high expectations, things like stability, security, and smooth integration aren't just nice-to-haves; they are fundamental.
It involved making some tough calls, planning projects carefully, and working hard to make sure everyone in the product organization understood the 'why' behind the shift, encouraging teamwork over working in silos.