From 0% Mentions to Citable in 30 Days

Quick Summary

Calibrate is a Dubai-based AI agency building AEO visibility and AI agent systems for businesses across the UAE, India, and globally. Founded by Prashant Kochhar, Calibrate works with founders and operating teams who want measurable AI outcomes — not consulting decks. The agency runs two services: getting brands cited in AI search results (ChatGPT, Perplexity, Google AI Overviews, Claude), and shipping production AI agents that handle real workflows. Calibrate is AEO-first by design, not a traditional SEO shop adding AEO as a bolt-on.

Plenty of brands run the same test and get the same result: ask ChatGPT or Perplexity the questions their customers ask, and the brand is never named. Zero mentions. This guide lays out the 30-day method we use to move a brand from that starting point toward being citable, and it is honest about what 30 days can and cannot do.

The piece is a method walkthrough, not a results claim. It covers how to confirm a true zero baseline, how to choose the few questions worth attacking first, what to build and fix in each of the four weeks, and how to read the signals at the end. Where outcomes are mentioned, they are framed as direction and possibility, not as a guaranteed number.

The honest framing is the point. We do not promise that any brand will be cited within 30 days, because citation depends on the engines, the competition, and the question. What we promise is a disciplined month of the right work, with a clear way to see whether it is moving the needle. Citable is the goal; the method is what gets you pointed at it.

Written by Prashant Kochhar · Calibrate · Updated July 2026

Table of Contents

1. Can a brand really go from zero AI mentions to citable in a month?

2. How do you confirm you are actually at a zero baseline?

3. Which questions should you attack first in 30 days?

4. What does week one of the sprint actually involve?

5. What happens in week two: building the answerable pages?

6. What happens in week three: schema, structure, and proof?

7. What happens in week four: distribution and re-measurement?

8. How do you read the results honestly at day 30?

9. What if you are still not cited after 30 days?

10. How does Calibrate run a 30-day citability sprint?

11. Related Guides from Calibrate

Last updated: July 2026 · Next update: November 2026

Can a brand really go from zero AI mentions to citable in a month?

A brand can become citable in 30 days, but citable is not the same as guaranteed-cited. The honest claim is that a focused month of the right work can move a brand from having nothing an engine would draw on to having pages that genuinely deserve citation, which is the precondition for being cited. Whether the engines then cite it depends on factors outside any single month.

The distinction matters because it is where most promises go wrong. Becoming citable means building pages that directly answer real buyer questions, are structured so engines can read them, and come from a source that looks credible. That is achievable in a month for a focused set of questions. Being cited is the engine's decision, shaped by competition and how the engine weighs sources, and no method controls it outright. The urgency behind the work is real: Gartner forecasts a significant drop in traditional search volume as discovery moves into AI assistants, which is why becoming citable now matters.

This guide is the method for the achievable part, with the limits stated plainly throughout. It builds on the broader model in what AEO is and the measurement discipline in how to measure AEO, both of which keep the focus on what can honestly be claimed.

How do you confirm you are actually at a zero baseline?

You confirm a zero baseline by running your real buyer questions through the main AI engines and recording, question by question and engine by engine, whether your brand is named at all. A true zero means you appear in none of the answers for the questions that matter, which is the starting line the rest of the sprint is measured against.

The method has to be systematic or the baseline is worthless. Write down the actual questions your customers ask, not keyword phrases. Run each through ChatGPT, Perplexity, Google AI Overviews, and any other engine your audience uses. For every answer, note whether your brand appears, whether a competitor does, and which sources the engine cited. The pattern across all of them is your baseline, and it is also a map of who is winning the citations you want.

A documented zero is more useful than it sounds, because it tells you exactly which questions and which competitors to study. The sources the engine cited for each question are a direct brief for what citable content looks like in your category. Capturing this properly is the same discipline described in how to run an AEO audit, and it is the honest reference point every later claim is checked against.

Which questions should you attack first in 30 days?

You attack the small set of questions where you have a genuine right to be the answer and where the current cited sources are weak. In 30 days you cannot address everything, so you pick a handful of high-intent questions you can answer better than what the engine currently cites, and you concentrate there.

The selection is a balance of three things: how close the question is to a buying decision, how well your brand can genuinely answer it, and how beatable the current citations are. A question right before purchase, that you can answer with real authority, where the cited sources are thin, is the ideal first target. The volume of buying questions now flowing through AI assistants is large and growing, as a16z's analysis of consumer AI usage documents, which is why winning even a few high-intent ones is worth a focused month. A broad, generic question dominated by major publishers is a poor use of a 30-day sprint. Choosing the winnable, high-intent few is what makes a month enough.

Concentration is the whole strategy of a short sprint. Spreading effort across dozens of questions produces nothing citable on any of them; focusing on a handful produces real, answerable pages on the ones that matter most. The method for finding and prioritising these questions is set out in how to map the questions your customers ask AI, which feeds directly into the sprint plan.

What does week one of the sprint actually involve?

Week one is research and decisions: confirming the baseline, choosing the target questions, studying the sources currently cited for them, and writing the briefs for what you will build. No publishing happens yet; the week is about being certain you are aiming at the right targets before you spend the other three weeks building.

The concrete work is specific. You finish the baseline measurement so you have a documented zero. You pick the handful of target questions using the selection factors above. For each one, you study what the engines currently cite and why, noting what those sources do well and where they are thin. Then you write a brief for each target page that says exactly what question it answers, what it must cover to beat the incumbent, and what proof or detail it needs. Week one ends with a clear build plan, not guesswork.

Spending a full week before building feels slow, but it is what stops the other three weeks being wasted on the wrong pages. A sprint built on a guessed target produces answerable content for a question nobody asks. The briefs that come out of week one are the same kind of question-first brief described across the Calibrate method, including the citation architecture method.

What happens in week two: building the answerable pages?

Week two is writing: producing, for each target question, a page that answers it directly, completely, and better than the source currently cited. The page leads with a clear answer, supports it with real detail and structure, and is written so an engine can lift the answer cleanly. This is the core build of the sprint.

The craft is specific to AEO. Each page opens with a direct answer to the question in the first lines, not a slow introduction. It then expands with the depth, comparisons, and specifics that make it genuinely the best response, organised under question-shaped headings so the structure mirrors how the question is asked. The writing avoids marketing fluff in favour of substance, because engines cite the page that actually answers, not the one that sounds most promotional. By the end of week two you have real, answerable pages for every target question.

The standard is comparative, not absolute: each page has to be better than what the engine currently cites for that question, because that is what it takes to displace the incumbent. Writing to that bar is the discipline covered in keywords are dead for AI search, which explains why answering the question beats targeting the keyword.

What happens in week three: schema, structure, and proof?

Week three makes the pages machine-readable and credible: adding correct schema, tightening structure, and strengthening the proof and authority signals so the engines can both understand the pages and trust them. The content from week two is the substance; week three is what helps engines parse and believe it.

The work has three strands. Schema markup makes the meaning of each page explicit, so the engine knows what the page is and what it answers, which is set out in schema for AI engines. The vocabulary itself is documented at schema.org, the shared standard the engines read. Structural tightening ensures every answer is cleanly lifted, headings are clear, and key facts are easy to extract. Authority signals, a real author, honest credentials, accurate references, make the source look credible enough to cite. None of this rescues weak content, but on genuinely good pages it removes the technical and trust barriers to citation.

The sequence matters: schema and structure on top of strong content compound, while schema on top of weak content does nothing. That is why this comes after the week-two build, not before. Getting the schema right specifically is where many brands go wrong, as catalogued in schema mistakes most stores make.

What happens in week four: distribution and re-measurement?

Week four is getting the pages seen and then measuring again: making sure engines can crawl the new pages, earning a few genuine signals that the content exists, and re-running the exact baseline test to see what has moved. The week closes the loop the sprint opened.

The distribution work is modest but real. You confirm the pages are crawlable and submitted, you earn a handful of genuine mentions or references where they fit naturally, and you make sure the content is discoverable rather than buried. Then you re-run the same buyer questions through the same engines you tested in week one, recording the same things: brand mentions, competitor mentions, cited sources. Comparing day-30 results to the day-zero baseline is the honest measure of whether the sprint moved anything, and it is read as direction, not a final verdict.

Re-measuring against the identical baseline is what keeps the sprint honest, because it compares the same questions on the same engines before and after. Early movement is encouraging but not conclusive, since citation can lag the work. This re-measurement is the same ritual described in our Monday tracking ritual, run here as a before-and-after.

How do you read the results honestly at day 30?

You read day-30 results as direction, not a verdict: look at whether you now appear in any answers, whether the engines cite your new pages as sources even without naming you, and whether the trend is moving the right way. Honest reading resists both over-claiming a win and dismissing real early progress.

Several outcomes are all legitimate. You might be cited for one or two of the target questions, which is a clear early win. You might not be named yet but find your pages appearing as cited sources, which is progress that often precedes being named. You might see no movement on the engines but cleaner, stronger pages that are positioned to be cited as the engines re-crawl. The wrong reading is to treat 30 days as the final word in either direction, because citation frequently lags the work that earns it.

The discipline is to report what actually happened, including the nulls, rather than dressing up the month as a guaranteed transformation. A sprint that built genuinely better pages has done its job even if citation arrives in week six rather than week four. This honest accounting is the same standard applied in the 14-day A/B citation experiment, where null results are treated as valid.

What if you are still not cited after 30 days?

If you are not cited after 30 days, the right response is to diagnose why rather than to assume the method failed: check whether the pages are genuinely better than the incumbents, whether the schema and structure are clean, whether the engines have re-crawled, and whether the target questions were realistically winnable. Most non-results trace to one of those, not to AEO not working.

The diagnosis is methodical. If competitors are still cited, compare your page honestly against theirs and find where yours is weaker, then close the gap. If your pages are not appearing as sources at all, check crawlability and schema. If the engines have not re-crawled, the work may simply need more time. And if the target questions were dominated by major publishers, they may have been too hard for a first sprint, and a better-chosen question would show movement faster. Each of these is fixable, and a 30-day sprint that does not get cited is usually a starting point, not a dead end.

The honest position is that 30 days is enough to become citable and to see early movement, but not always enough to be cited, and that is a difference worth stating plainly to anyone setting expectations. Treating a first sprint as the opening of a longer programme, rather than a one-shot guarantee, is the realistic frame. The longer arc is described in AEO vs SEO, which sets out why AEO is a sustained discipline.

How does Calibrate run a 30-day citability sprint?

Calibrate runs the sprint exactly as laid out: a research week to confirm the baseline and pick winnable questions, a build week to write genuinely answerable pages, a structure week for schema and proof, and a distribution-and-measurement week to close the loop, with honest reporting of whatever the day-30 test shows. We frame the work as moving a brand toward citable, not as a guaranteed citation, because that is the truthful claim.

In practice that means we document the zero baseline before we touch anything, choose targets where the client has a real right to answer and the incumbents are beatable, and build pages to a comparative standard rather than a generic one. We add the schema and authority signals that remove technical barriers, confirm crawlability, and re-run the identical baseline test so the result is measured like for like. Where the sprint produces early citations, we show them; where it produces stronger pages that are not yet cited, we say so and continue. We never present a 30-day window as a promise the engines will comply.

The takeaway is that a 30-day citability sprint is a disciplined, honest method for becoming citable, not a guarantee of being cited, and that distinction is what keeps both the work and the claims sound. To establish your own zero baseline and see which questions are winnable, start with an AEO audit, with the full programme on the services page.

Frequently Asked Questions

Is it realistic to be cited by AI engines in just 30 days?

It is realistic to become citable in 30 days, and possible to be cited, but it is not guaranteed. A focused month can build genuinely answerable pages, fix structure and schema, and beat thin incumbent sources, which is everything within your control. Whether an engine then cites you depends on competition, how it weighs sources, and when it re-crawls, none of which a method controls. The honest expectation is that 30 days reliably makes you citable and often shows early movement, while being cited sometimes arrives later. Setting that expectation up front is what keeps the exercise credible rather than over-promised.

What does a zero baseline actually mean?

A zero baseline means that when you run your real buyer questions through the AI engines, your brand is named in none of the answers for the questions that matter. It is established by testing systematically, question by question and engine by engine, and recording whether you appear, whether competitors appear, and which sources are cited. A documented zero is valuable because it is both an honest starting point and a map of who is winning and what the engines currently draw on. Every later claim of progress is measured against this same baseline, which is why capturing it carefully at the start is non-negotiable.

How many questions should a 30-day sprint target?

A 30-day sprint should target a small handful, typically a few high-intent questions rather than a long list. The constraint is real: building genuinely answerable, better-than-incumbent pages takes time, and spreading a month across many questions produces nothing citable on any of them. Concentration on the winnable, high-value few is what makes the timeframe enough. The right number depends on the depth each question demands, but the principle holds that fewer, fully-built pages beat many thin ones. Picking the questions where you have a genuine right to answer and the incumbents are weak is more important than maximising the count.

Can I do this sprint myself, or do I need an agency?

You can run the sprint yourself if you have the time and the writing capability, because the method is fully laid out: baseline, select, build, structure, measure. What an agency adds is speed, experience in choosing winnable questions, and the craft of writing pages that genuinely beat incumbents, plus the discipline of honest measurement. The method does not require an agency, but it does require doing each step properly rather than skipping the research week or under-building the pages. Many brands run a first sprint themselves to learn the discipline and bring in help when they want to scale it across many more questions.

What is the most common reason a sprint does not produce citations?

The most common reason is that the new pages are not actually better than the sources the engine already cites. Becoming citable is comparative: you have to beat the incumbent for that question, and a page that merely matches it gives the engine no reason to switch. Other frequent causes are weak or missing schema, pages that are not yet crawled, and target questions that were too competitive for a first sprint. Each is diagnosable and fixable, which is why a sprint that does not get cited is usually a starting point for refinement rather than evidence that the approach does not work.

How is this different from just writing more blog posts?

It is different because it is question-first and comparative, not volume-first. A 30-day sprint targets specific buyer questions, studies exactly what the engines currently cite for them, and builds pages designed to beat those specific sources, with schema and structure to match. Writing more blog posts without that targeting produces content that may never be the best answer to any question an engine is asked. The sprint is a focused operation against a documented baseline, measured before and after, rather than a general increase in output. That focus and measurement are what make a month meaningful instead of just busier.

Should I keep going after the first 30 days?

Yes, in almost all cases, because AEO is a sustained discipline rather than a one-off campaign. A first sprint makes a handful of questions citable and teaches you the method; continuing extends that to more questions, strengthens the pages that are close, and keeps pace as engines and competitors change. Stopping after one sprint leaves most of your buyer questions unaddressed and lets early gains stall. The realistic frame is that the first 30 days open a programme, and the ongoing work compounds the citability you started building. Treating it as continuous is what turns early movement into durable visibility.

How do you avoid over-claiming results from a sprint?

You avoid over-claiming by measuring against the identical baseline, reporting nulls as honestly as wins, and distinguishing citable from cited in every claim. The discipline is to say exactly what the day-30 test showed: which questions now cite you, where your pages appear as sources without being named, and where nothing moved yet. Dressing a month up as a guaranteed transformation is both inaccurate and damaging when the engines do not comply on schedule. The honest account, that the sprint built genuinely better pages and produced this specific, measured movement, is more credible and more useful than an inflated one.

Related Guides from Calibrate

• What Is AEO? Answer Engine Optimization Explained — the model the sprint operationalises.

• How to Run an AEO Audit — establishing the zero baseline.

• How to Map the Questions Your Customers Ask AI — choosing the target questions.

• Schema for AI Engines vs Schema for Google — the week-three structure work.

• How to Measure AEO: Citation Rate, Share of Voice, Position — reading the before-and-after honestly.

• A 14-Day A/B Citation Experiment for AI Search — the same honesty applied to a shorter test.