Generative Engine Optimization
Shape how generative models describe your organization, so summaries about your cause are accurate, current and favourable.
What you get
- A baseline audit of what ChatGPT, Gemini, Perplexity and Copilot currently say about you
- Correction of outdated or inaccurate claims that models keep repeating
- Training-data-friendly public facts: mission, programs, impact numbers, service areas
- Quarterly re-testing to catch drift as models are retrained
Control the story models tell about you
Ask an AI assistant to recommend charities working on watershed conservation in British Columbia. It will produce a confident list with short descriptions. Those descriptions came from somewhere, they are often two years out of date, and no one at your organization approved them.
Generative engine optimization is the discipline of making that synthesized description accurate.
Our process
- Model audit. We prompt the major assistants with the questions your funders, donors and partners would ask, and record exactly how each one describes you, who it cites, and what it gets wrong.
- Source tracing. For every inaccuracy we find where it likely originated: a stale directory entry, an old press release, a Wikipedia stub, a funder profile from a previous strategic plan.
- Canonical facts. We publish a clear, machine-readable set of current facts on your own site - mission, programs, service areas, impact figures, governance - and mark it up so it is unmistakable.
- Distribution. We update the third-party sources that models lean on most, so corroboration works in your favour.
- Re-testing. Models change. We re-run the audit quarterly and report on drift.
Why this matters for fundraising
Major donors and grant officers now open a chat window as part of preliminary research. The summary they read is your first impression, whether or not you wrote it.
Frequently asked questions
Is GEO the same thing as AEO?
They overlap but differ in scope. AEO focuses on being cited inside a specific answer to a specific question. GEO is broader: it is about how generative models represent your organization in general, including summaries, comparisons and recommendations where no single question was asked.
Can you really influence what an AI model says about us?
Not by instruction, but yes by evidence. Models synthesize what the public web says about you. When your own pages, your funders, your partners and reputable directories all state the same current facts, the model's summary converges on those facts. When the web is silent or contradictory, the model guesses.
What if a model is saying something wrong about our charity?
That is one of the most common reasons organizations come to us. We trace which sources are feeding the error, correct or supersede them where possible, publish clearer canonical facts on your own site, and then re-test across models until the summaries reflect reality.