Fresh from digitalNow 2025 in Chicago, one thing is crystal clear: associations have moved past asking "should we use AI?" to grappling with "how do we actually make this work?" For those of us building technology for associations, the conference revealed both the progress made and the persistent challenges that remain.
The 18-Month Implementation Stories
The most valuable sessions weren't about AI's potential—they were implementation case studies from associations sharing what actually happened when they tried to deploy AI at scale. ICBA spent 18 months rebuilding five websites using member behavior data. The Northern Virginia Association of Realtors re-architected their entire content infrastructure to become "AI-ready." The American Academy of Pediatrics focused on solving specific business problems rather than chasing technology trends.
AGU's Thad Lurie presented 18 months of AI deployments, focusing not on the technology itself but on measurable business outcomes. These weren't overnight transformations. They were methodical, sometimes painful journeys of building the right foundations first.
The Infrastructure Reality Check
Multiple sessions hammered home an uncomfortable truth: you can't just layer AI on top of existing systems and expect magic. Success requires proper foundations—content architecture, taxonomy, metadata, and what speakers called "ingestion pipelines" so AI can actually understand organizational data.
Ian Andrews from Groq didn't talk about chatbots or quick wins. He covered inference technologies, energy-efficient AI architectures, and regulatory frameworks. This is infrastructure-level thinking—the unglamorous but essential work that makes AI actually function at scale.
For associations still struggling to get basic insights from their data despite investments in multiple systems, this presents a fundamental challenge. How do you build AI capabilities when getting a simple cross-system report still takes weeks?
Culture: The Unexpected Bottleneck
Perhaps the most surprising theme: organizational culture emerged as the primary barrier to AI adoption, mentioned more frequently than any specific technology challenge.
The Morton Arboretum created "AI Explorers" programs—judgment-free spaces for non-technical staff to experiment. Organizations ran Shark Tank-style innovation contests. The "Beyond the Bot" panel featured leaders from AANA, AACSB, and MDRT discussing how they created environments where people felt safe to try, fail, and learn.
The father-son keynote team of Conor and Finn Grennan put it bluntly: culture, not tools, determines AI success. You can have the best technology in the world, but if your team doesn't feel empowered to use it, you've achieved nothing.
The Trust Crisis No One Wants to Discuss
Marcus Sheridan's opening keynote tackled the question lurking beneath every session: are associations still relevant when members can get instant answers from ChatGPT?
When members can Google anything, ask AI anything, and find YouTube influencers teaching their profession, why would they need their association? The conference didn't offer easy answers, but the consensus was clear: associations can't compete on information access anymore. They must compete on trust, community, and the application of knowledge to specific professional contexts.
The Data Analytics Gap
While sessions focused on content AI and member-facing chatbots, a quieter theme emerged in hallway conversations: associations still can't answer basic questions about their own operations. They have data scattered across dozens of systems—AMS, financial platforms, event management tools, marketing automation—but no way to connect the dots.
One CEO mentioned waiting three weeks for a report on member retention by segment. Another described their "analytics graveyard"—expensive BI tools that no one uses because they're too complex. The irony wasn't lost on attendees: we're implementing AI for members while still making decisions based on gut feel about our own organizations.
This is where solutions that democratize data access become critical. If department heads could explore data through natural conversation rather than submitting IT tickets, if insights took minutes instead of weeks, if analytical capabilities grew smarter over time rather than remaining static—that would represent real transformation.
What "AI Adolescence" Really Means
Speakers described associations as being in "AI adolescence"—past the wonder of discovery but not yet mature. This metaphor perfectly captures the awkward current state:
- We know AI can help, but implementation is harder than expected
- We've run pilots that show promise, but scaling them seems impossible
- We want to be data-driven, but our data remains locked in silos
- We talk about innovation, but our cultures resist change
The path forward isn't about more pilots or proof-of-concepts. It's about building sustainable capabilities that deliver consistent value.
The Practical Path Forward
Based on what we heard at digitalNow, here's what associations actually need:
Start with accessible wins. Not every AI initiative needs to be transformational. Sometimes just getting faster access to your own data represents massive progress.
Focus on adoption, not just implementation. The best technology fails without users. Build solutions that non-technical staff can actually use.
Connect before you transform. Before adding AI capabilities, ensure your data sources can actually talk to each other.
Measure what matters. Stop counting pilot programs. Start measuring time-to-insight, user adoption, and business impact.
Moving Beyond the Hype
digitalNow 2025 revealed an industry ready to move past experimentation but struggling with execution. The challenges aren't primarily technical—they're organizational, cultural, and foundational.
For those of us building association technology, this represents both a responsibility and an opportunity. Associations don't need more complexity. They need solutions that simplify, that democratize access to insights, that grow smarter over time, and that respect the realities of how associations actually operate.
The conference made clear that successful AI adoption isn't about having the most advanced technology. It's about making that technology accessible, trustworthy, and valuable to the people who need it. That's the real work ahead.
Tags:
Nov 7, 2025 5:13:43 PM
Comments