Week by week breakdown of launching an Answer Engine Optimization program from zero. Real actions, real tools, real metrics. The exact sequence we run for new Orkkid clients.
Most businesses ask "where do I start with AEO?" and freeze. The category is new, the literature is thin, and every agency pitch sounds the same.
This is the playbook we run for every new Orkkid retainer client in their first 30 days. Week by week. Specific actions, specific tools, specific outputs. By day 30 you have a working baseline, the technical foundation, your first round of citable content, and a tracking system that runs monthly.
It is structured around our free tools so you can run most of it yourself. If you want us to do it for you, book a free AI citation audit and we will start at week one with you.
Before you start: the success metric
The metric is citation share: the percentage of relevant AI prompts that name your business.
Day-1 baseline expectations for an Australian service business with no prior AEO work:
- New domain (under 1 year): 0-5% citation share across major engines
- Established domain with traditional SEO: 5-20%
- Established domain with strong PR or industry mentions: 15-35%
By day 30 we expect the baseline measured. By day 90 we expect a measurable lift. By day 180 we expect a meaningful lift. AEO is a discipline of compounding, not of overnight wins.
For the deeper context on the metric, read how to track ChatGPT citations.
Week 1: Audit and baseline
The first week is diagnostic. You do not change anything. You measure.
Day 1-2: Build the prompt panel
Use the Prompt Panel Generator to produce 30 customer-shaped prompts split into branded, category, and long-tail problem queries. Plug in your industry, country, city, and competitors.
Save the panel as a CSV. This is the same panel you will run every month from now on. Consistency matters more than cleverness.
Day 2: Run the citation checker for the first time
Use the AI Citation Checker with 5 of your most representative prompts (one branded, two category, two long-tail). Run it across ChatGPT and Perplexity.
Record the results. This is your day-1 baseline. Note:
- The percentage cited (your starting citation share)
- The competitors cited instead
- The brands the AI mentions in the same response (these are your real competitors for AI mindshare, often different from the competitors you think you have)
Day 3-4: Run the AI Visibility Audit
Use the AI Visibility Audit on your domain. You will get a 0-100 score across six categories: AI crawler accessibility, schema, llms.txt, sitemap, on-page citability, and security.
Print the report. The categories that scored under 60 are your week-2 priorities.
Day 5-7: Document the gap
Write down three things in plain language:
- What did AI engines say about your business when they did mention you? Was the description accurate?
- Who got cited instead? Why do you think they were chosen?
- Which audit categories scored lowest, and what would it take to fix each?
This document becomes the brief for week 2. Without it, week 2 becomes scattered.
Week 2: Technical foundation
Week 2 is pure technical hygiene. No content writing yet. We are making sure AI engines can read your site cleanly.
Day 8: Fix robots.txt
Open yourdomain.com/robots.txt. Confirm it explicitly allows the major AI crawlers: GPTBot, ChatGPT-User, ClaudeBot, anthropic-ai, Google-Extended, PerplexityBot, CCBot, Applebot-Extended.
If any are missing or blocked, fix it. This is the lowest-effort highest-impact item on the list. If AI bots cannot crawl your site, nothing else you do matters.
Day 9: Publish llms.txt
Use the llms.txt Generator. Paste your URL, copy the output, save it as llms.txt in your site root.
For Next.js: public/llms.txt. For WordPress: upload via FTP. For Webflow: site-wide custom file. Verify it loads at yourdomain.com/llms.txt.
For more on what this file does, read our llms.txt beginner guide or the glossary entry.
Day 10: Schema audit
Check your homepage source for application/ld+json blocks. Use the Google Rich Results Test to validate.
You should have at minimum:
- Organization or LocalBusiness on the homepage
- WebSite with sitelinks search box
- BreadcrumbList on internal pages
- Service on each service page
If anything is missing, use the Schema Generator to fill the gap.
For the full schema strategy, read our AEO schema markup complete guide.
Day 11: Sitemap
Confirm your sitemap is at /sitemap.xml, returns 200, contains all your important URLs, and is referenced in robots.txt.
If your CMS does not auto-generate one, use a sitemap plugin or build one. The glossary entry explains why.
Day 12: Canonical tags
Every page should have a self-referencing canonical pointing to itself with the correct protocol (https) and domain (apex or www, pick one). For more context see our canonical URL glossary entry.
Day 13: Mobile and Core Web Vitals
Run PageSpeed Insights on your homepage and three internal templates. Aim for Core Web Vitals in the green band. AEO depends on Google rankings for live retrieval, and CWV is a confirmed ranking signal.
Day 14: Re-run the audit
Run the AI Visibility Audit again. The technical categories should now be in the 80+ range. If anything is still under 70, that is your week-3 carryover.
Week 3: Content rewrites for citability
Week 3 is where most agencies fail. They redesign the site instead of rewriting it. We do the opposite.
Day 15-16: Pick your top three pages
Identify the three pages on your site most likely to be cited if rewritten:
- Your homepage (the AI default for "tell me about [brand]")
- Your most-trafficked service page
- Your most-trafficked blog post or guide
For each, ask: which prompts in our panel should this page answer? Note them next to the page.
Day 17-19: Rewrite paragraphs
This is the practical heart of the playbook. For each of the three pages, rewrite paragraphs using the four-property framework: self-contained, definitive opening, specific facts, schema-attached.
For the writing technique, read how to write content that AI will cite.
Per page, expect 2-4 hours of focused rewriting. The total time is 6-12 hours across three days. Anything less than that and you have not actually rewritten; you have surface-edited.
Day 20-21: About page deep work
The About page is your highest-leverage rewrite. AI engines reach for it when asked "who is [your business]" or "tell me about [your business]". Most About pages are brand poetry; yours needs to be a structured fact dump.
Required elements:
- Founder name with credentials
- Founding date
- Location with city and state
- Specific services
- Industries served
- Notable clients or projects (real ones)
- Awards or recognition
- One-sentence positioning that ChatGPT can quote
This pattern is closer to a Wikipedia entry than a marketing page. That is the point.
Week 4: Brand mentions and tracking
Week 4 is off-page. AI engines weight signals from outside your own site at least as heavily as signals from inside it.
Day 22-23: Audit your brand entity
Search for your business name in:
- Wikipedia (do you have an entry? probably not)
- Wikidata (the structured-data layer that feeds Knowledge Graph)
- Crunchbase
- Industry directories (hipages, Oneflare, ProductReview if Australia)
- Google Business Profile
Note where you exist and where you do not. The entries that exist should be accurate, complete, and use consistent NAP (name, address, phone). For more on this see our NAP consistency glossary entry.
Day 24-25: Build first brand mentions
Pick three places to land a brand mention in week 4:
- One industry publication. Write a guest post, do a quoted comment via HARO/SourceBottle, or contribute to a roundup.
- One podcast appearance. Even a small podcast counts. AI engines train on podcast transcripts.
- One Reddit or forum thread. Genuinely participate in your industry's largest community. Get your brand named once organically.
These first three mentions are not a campaign; they are proof you can do it. Maintain a steady cadence after week 4.
For the deeper context on why brand mentions matter, read brand mentions: the underrated AEO signal.
Day 26-27: Re-run the citation checker
Use the AI Citation Checker again with the same 5 prompts you used on day 2. Compare to baseline.
If you have done all of week 1-4 properly, expect:
- 0-3 percentage points lift on category queries (slow, this is fine)
- 5-15 percentage points lift on branded queries (faster, AI engines pick up new info quickly)
- More accurate descriptions of your business when you are cited
Real citation share movement on category queries usually shows up in week 8 to 12, not week 4. The point of the day-26 re-run is to detect whether the technical foundation is working, not to declare victory.
Day 28-30: Set up monthly tracking
Lock in the recurring discipline.
- Calendar a recurring monthly slot for the prompt panel run.
- Pick a tool: free (the AI Citation Checker on your top 5 prompts), or paid (see our citation tracking tools comparison).
- Decide one person who is accountable for running it, recording results, and flagging trend changes.
- Set a quarterly review where you read three months of trend data and decide whether to adjust.
This is the discipline that separates programs that work from programs that quit at week 8.
Week 5 and beyond
The 30-day playbook gets you to a working program. Months 2-6 are about consistency.
Monthly:
- Run the prompt panel
- Publish 2-4 new pieces of citable content
- Build 2-4 new brand mentions
- Update llms.txt with new URLs
Quarterly:
- Re-audit the technical foundation (re-run the visibility audit)
- Refresh the prompt panel (questions drift over time)
- Review trend and adjust
Annually:
- Refresh About page
- Audit Wikipedia/Wikidata/Crunchbase entries
- Review schema for new types Google adds
What to expect at each milestone
| Milestone | Citation share lift | What's happening |
|---|---|---|
| Day 30 | 0-5 points on category queries | Technical foundation set, first content rewrites live |
| Day 60 | 3-10 points | AI engines re-crawled the rewritten pages |
| Day 90 | 5-20 points | First brand mentions filtered into training/retrieval |
| Day 180 | 10-30 points | Compounding effect kicks in |
| Day 365 | 20-50+ points | Mature program; you are the cited brand for category queries |
These are illustrative ranges based on what we see across Australian service business clients. Individual results depend heavily on starting position, market saturation, and consistency of execution.
For the financial side, see our AEO vs traditional SEO ROI breakdown.
Common reasons playbooks like this fail
- Skipping week 1 baselining. If you do not measure before you start, you cannot tell if it is working.
- Doing week 2 (technical) but skipping week 3 (content). Schema is necessary but not sufficient.
- Treating week 4 (brand mentions) as optional. It is the off-page leverage that compounds the on-page work.
- Quitting at week 8 because there's no visible movement. The first 8 weeks are foundation. Weeks 8-24 are when the metrics actually move.
Frequently asked questions
Can I run this playbook in parallel with traditional SEO work?
Yes, and you should. AEO and traditional SEO share most foundational signals. The same content rewrites that improve AI citation also tend to improve Google rankings. The same schema work helps both.
What if I am not in Australia?
The playbook works anywhere. The Australian-specific elements (hipages, ProductReview, AU directories) swap to your local equivalents (Yelp, Trustpilot, BBB, etc.).
What is the minimum budget to do this?
Pure DIY: zero, using our free tools. It takes 30-50 hours of your time across 30 days.
With an agency: $2,500 to $6,000 a month for a serious AU AEO program.
How does AEO interact with paid search?
Paid search buys you visibility now. AEO builds it over time. The brands that win are doing both. As paid CPA inflates (AI engines intercept early-funnel demand), AEO becomes the more efficient long-term spend.
If you want this run for you instead of by you, book a free AI citation audit. We will start with week 1 baselining and send a personalised report inside 72 hours.
For deeper context, read the GEO SEO complete guide, how to write content AI will cite, and our AEO schema markup guide. For service work, see our AEO services.