We Built a 76-Term AI Glossary and Didn't Put a Form on It
The new /glossary page defines 76 terms across five domains: AI, web, database, identity, and servers. No email required, no signup wall, no form to fill out before you can read the definition of "RAG."
That last part is the decision worth explaining.
What the Glossary Covers
The audience is owners and operators of service companies doing $1M to $10M. The kind of business where the owner is in every vendor call, hears terms like "webhook" and "row-level security" and "MCP server," and has to nod along while mentally noting to look it up later.
Every entry has three parts. First, a plain-English definition with no jargon inside the definition itself. Second, the problem that existed before the thing, because the history is what makes the term stick. If you know that row-level security exists because developers kept writing queries that forgot permission filters, the concept clicks in a way that "a database access control layer" never would. Third, where you'd actually hear it: a line like "a developer saying 'I'll build you an endpoint for that' means one new, narrow capability, usually hours not weeks."
The five domains (AI, web, database, logins, servers) let someone filter to just the vocabulary relevant to their current project. If you're in the middle of a portal build, you're probably not confused about the AI terms right now.
The Overloaded Words Section
The part we spent the most time on was a set of 12 words that cause the most confusion in client conversations: words like "model," "agent," "context," "integration," and "workflow." These mean different things depending on whether the speaker is an AI engineer, a CRM vendor, or a no-code automation builder.
"Integration" in a SaaS tool means a pre-built connector in their marketplace. In a developer context it means custom API work. These two things are not the same commitment. A client who hears "we integrate with that" from both a vendor and a developer might assume the effort is comparable when it isn't.
We documented the range of meanings for each of these words because the confusion usually shows up in scoping calls, not implementation. Someone quotes a price based on one meaning, the client heard a different one, and six weeks in everyone is surprised.
Why It's Ungated
A form on a top-of-funnel reference page is a tax on someone who is trying to do a legitimate thing: learn what a word means. The people most likely to fill out that form are not the best leads. The people most likely to bookmark the page, share it with their team, and come back when they have a real project are exactly the clients worth reaching.
Search engines index it. Someone Googling "what is a webhook" can land there. Someone in a Slack thread about AI tools can drop the link. That reach disappears the moment there's a gate, and the thing behind the gate isn't valuable enough to trade an email address for.
We keep the forms for things that represent a real commitment: audit requests, project inquiries, the AI Readiness Assessment. The glossary is a reference tool. It belongs on the open web.
What It's Next To
The glossary sits on the same site as the AI Operations Academy, which starts with a $750 AI Readiness Assessment. Someone who finds the glossary while trying to parse what their vendor said, realizes they need more than definitions, and sees a clear next step. We didn't build it as a funnel. The gap existed. The two things fit together in a way that feels honest rather than engineered.