Doug Turnbull
Category: AI
It’s not technology that’s holding back AI. Technology comes and goes. Yesterday’s AI model goes in tomorrow’s dustbin.
No it’s not how it’s who. A great team can pivot, try new things, with an inexhaustible spirit of exploration. What’s missing? Search skills. You’re missing the ‘search’ part of the ‘AI team’. The search skills gap is a hidden, under-resourced bottleneck leading to project delays and poor execution.
Why? Let’s dig into what’s happening.
Two evergreen lessons continue to appear in market research and our work with clients:
- AI Skills Gap: Teams cite the lack of skills for AI Project failure.
- AI Quality Depends on Search: RAG is cited by 93% of practitioners as required to get good answers out of AI. In addition, the FRAMES 2025 study quantified that agents with search dramatically outperform agents without.
The implication: Building AI products requires search skills. It’s about developing the who in your team: giving them the tools they need and letting them execute. The AI skills gap turns out to be as much a retrieval skills gap. Build a good search team, then you’ll thrive in AI. Search industry reports deepen our understanding of what’s happening. In their State of Search 2026, Elastic reports:
- 48% of search teams cite lack of skills for why they don’t implement Agentic AI.
- 38% avoid vector and hybrid search due to a lack of skills.
Algolia’s own State of Search Teams in 2025 shows how stressed search teams have become:
- Only 6% of teams have dedicated search developers.
- Of search developers, 71% of developers spend significant time explaining search concepts to their peers.
Organizations want to push to higher and higher levels of AI sophistication. Yet the organization’s own search people find themselves overstretched and overwhelmed.
Even scarier, what if you can’t retain that great search developer? A report from the AI workforce consortium points to RAG in the top tier of in-demand AI skills. In 2026, RAG was listed in Futurense’s most in-demand skills as well. Search talent isn’t quite treated like the professional athletes the way AI researchers hop between Anthropic / OpenAI. Nevertheless, we see the consequences of in-demand search talent in day-to-day jobs: lost talent, leading to less and less effective teams.
So search skills matter; they drive AI quality. What can you do about it? The kneejerk would be to reach for training. Yes, search training courses abound. Training can help but it’s important to treat it as a trigger, not a cure-all. Training only gives technical skills. If you want to retain talent, grow your capabilities, and help your overstretched team, training needs to be the trigger to something bigger. You need to think of this problem about organizational competencies. Not just technical skills.
Without the organization to support them, technical talent turns into instant flight risk. They might look around and say “this place isn’t for me.” They might not see the impact of their work. They don’t see how they’ll grow. Would you work on a team where you so quickly hit a ceiling? Reorienting to solve the search skills gap — indeed the search talent retention gap — means rethinking the organization around AI and search.
Consultants help reorient firms to overcome the AI and search skills gap and become an organization that breathes models, data, and evaluation. One that runs dozens of experiments a day and puts science above egos. One that puts data-informed innovation before endless arguments about the best path forward.
The light version of this would be to roadmap together with an expert to plan the product. But it can also mean digging even deeper: to rethink the organization and its abilities. Consultants have seen the breadth of search at many companies and can help you find the unknown unknowns you weren’t even thinking of. OpenSource Connections, for example, has seen its fair share including hundreds of search teams from e-commerce to RAG.
A longer-term relationship with consultants helps keep things on track. This means not just throwing bodies at the problem but bootstrapping mentorship in the organization. That gives your talent the mentor and growth opportunities they need. You also have a partner helping you steer the organizational ship towards one with search and DNA in your bones. This ensures stakeholders get what they need, and that AI and search talent feels nurtured.
This is where OpenSource Connections can help. We want to empower your team. To help you build an organization that can overcome the AI + search skills gap. It’s not easy out there. AI can feel like a rat race. The good news: You do have options including partners that can empower you and take you to the next level.
If that’s you, get in touch! We’d love to have a conversation.