Free course · 12 lessons · self-paced
Become the answer.
Answer engine optimization, from first principles to a 90-day plan. Learn how AI assistants decide what to say, measure where your brand stands with the MRCA scorecard, and fix what keeps you out of the answer.
0 of 12 lessons complete
12 lessons · 61 min · free · Print the whole courseCe cours n'est pour l'instant disponible qu'en anglais. Le reste du site est en français.
/ 01 — LESSONS
Five modules, twelve lessons
12 lessons
Module 1
How answer engines decide
What AEO is, how an AI answer is assembled, and how the engines differ.
LESSON 01 · 6 minWhat AEO is, and what “being the answer” meansAnswer engine optimization is the work of making your brand easy for AI assistants to find, trust, recommend and describe correctly. It builds on SEO rather than replacing it.LESSON 02 · 5 minHow AI answers are builtAn assistant answers from what its model already knows, from pages it retrieves live, or from both. Understanding which one is in play tells you what you can influence.LESSON 03 · 5 minAnswer engines comparedChatGPT, Gemini and Google's AI features, Perplexity, Claude and Copilot each find and show sources differently. Results from one do not transfer to another.
Module 2
Measure where you stand
A fixed prompt panel and the MRCA scorecard: your measuring stick.
LESSON 04 · 5 minBuild your prompt panelA prompt panel is a fixed set of real buyer questions you ask the same way, on the same engines, on a schedule. It is the measuring stick for everything else in this course.LESSON 05 · 5 minThe MRCA scorecardScore every answer in four stages: Mentioned, Recommended, Cited, Accurate. The stage where a brand drops out tells you which fix to make.
Module 3
Make your site answerable
What crawlers read, who may read it, and pages written to be quoted.
LESSON 06 · 5 minWhat machines actually readMany AI crawlers read the raw HTML your server sends and do not run JavaScript. Content that appears only after scripts run may be invisible to them.LESSON 07 · 5 minCrawler access: robots.txt, AI user agents and llms.txtDecide, agent by agent, which AI crawlers may read your site and for what purpose, then check that your CDN and firewall respect that decision.LESSON 08 · 5 minWriting answer-ready pagesPut the answer first, make each passage stand on its own, name things explicitly, date what changes, and keep structured data true to the visible page.
Module 4
Earn trust beyond your site
Reviews, consistent facts and independent mentions as evidence.
LESSON 09 · 5 minReviews and reputation as evidenceFor “best X near me” and “is X any good” questions, assistants lean on what customers and directories say about you. Review recency, responses and consistent business facts are evidence you can improve honestly.LESSON 10 · 5 minMentions, communities and videoIndependent coverage, honest comparison pages, helpful community answers and well-structured video all add evidence about you. Earn them; do not seed them.
Module 5
Run it as a program
Honest reporting and a 90-day rhythm you can keep up.
LESSON 11 · 5 minAttribution and reportingAI referrals show up in analytics only partly. Combine them with bot activity, a “how did you hear about us” question and your MRCA trend, and report monthly.LESSON 12 · 5 minYour 90-day AEO planTurn the course into a program: a first week of setup, a 30, 60 and 90-day rhythm, a process for wrong answers about your brand, and clear rules for when to re-baseline.
/ 02 — MRCA
One scoring model, all the way through
Echorank scores every AI answer in four stages. A brand can pass one and fail the next, and each failure has a different fix. You will meet the model in lesson 1, build it in lesson 5 and report on it in lesson 11.
| Stage | Passes when |
|---|---|
| M · Mentioned | The assistant names you. |
| R · Recommended | It suggests you for the need. |
| C · Cited | It shows a page you own as a source. |
| A · Accurate | What it says about you is true. |
/ 03 — Course at a glance
Course at a glance
| Lesson | You will learn | You will produce | Minutes |
|---|---|---|---|
| 1. What AEO is, and what “being the answer” means | Define AEO in one sentence you can repeat to a colleague | A one-line AEO goal for your brand | 6 |
| 2. How AI answers are built | Tell model knowledge apart from live retrieval in a real answer | A list of the sub-questions behind your top buyer question | 5 |
| 3. Answer engines compared | Know how each major engine tends to show its sources | Your shortlist of engines to track, with a reason for each | 5 |
| 4. Build your prompt panel | Choose 20 to 30 prompts from real buyer decisions rather than keywords | Your first prompt panel, with a type and stage for every prompt | 5 |
| 5. The MRCA scorecard | Score any AI answer on the four MRCA stages | A scored baseline for your prompt panel | 5 |
| 6. What machines actually read | Tell the raw HTML of a page apart from what a browser renders | A list of key pages with content a crawler cannot see | 5 |
| 7. Crawler access: robots.txt, AI user agents and llms.txt | Separate AI crawlers that fetch for search from those that collect training data | A crawler access decision for each AI user agent | 5 |
| 8. Writing answer-ready pages | Rewrite a page so its first paragraph answers the question it targets | A rewrite plan for your three most important answer pages | 5 |
| 9. Reviews and reputation as evidence | Identify the reputation signals assistants can draw on for your business | A reputation evidence map with the inconsistencies to fix | 5 |
| 10. Mentions, communities and video | Know which kinds of independent mention assistants tend to draw on | A list of earned-media targets and corrections to request | 5 |
| 11. Attribution and reporting | Find AI referral traffic in your analytics and know what it misses | A one-page monthly AEO report template | 5 |
| 12. Your 90-day AEO plan | Plan the first week and the first 90 days of an AEO program | Your completed AEO plan, ready to share | 5 |