The first quarter of an AEO program usually shows almost no citations, and that silence is when most companies quit. The fix is to stop watching the citation count and watch the leading indicators underneath it: branded search volume, impressions on new pages, AI-bot crawl frequency, and long-tail share of answer. If three of those four are climbing, the foundation is working and you should hold.
Six weeks into an AEO program, a founder emailed me one line: "Is this actually working?" The citation count was still zero. Nothing in ChatGPT, nothing in Perplexity, nothing in Google's AI answers. On the dashboard, we had spent six weeks and moved nothing visible. I understood the email completely. I also knew that the quarter he wanted to quit was the exact quarter that decides whether the whole program pays off.
Almost everyone quits in this quarter. The ones who do not are the ones who get cited for years.
I have written elsewhere about the mechanics, the three speeds AI citations actually move at and what gets built underneath, in how long AI citations take. The mechanics are the easy part. The hard part is the quarter where you cannot see any of it yet, and whether you will still be in the room when it lands.
Why month one lies to you
The cruel structural fact of AEO is that the signals you watch are the last ones to move, and the work that decides the outcome is invisible by design.
When an AI engine decides whether to cite you, it is drawing on an entity graph it has built about your brand over time, the structured content it can extract cleanly, and the third-party sources it already trusts. None of that is fast to build, and none of it shows up on a citation counter while it is being built. So for the first stretch, you are pouring concrete. Concrete does not look like anything. It looks like a hole in the ground and a large invoice.
Then one part of the work pays off almost instantly, which makes the silence on everything else feel like failure. A single clean answer block on a page that already ranks can get pulled into a Google AI Overview within a day. That speed is real. It is also a trap, because it teaches you to expect the rest of the program to move at the same pace, and the rest of the program does not. The durable work, the entity graph and the citation surface across engines, compounds over a quarter rather than a day. The instant win and the slow compound run on the same program at the same time, and the gap between them is where people lose their nerve.
How to tell a working silence from a broken one
Here is the part nobody writes, and it is the whole game. "Just wait" is what good agencies say and it is also what bad agencies say to dodge accountability. The skill is telling the two apart while the citation count is still zero.
A working silence has leading indicators moving underneath it. A broken program has flat lines all the way down. During the quiet quarter, stop staring at the citation count and watch these instead:
- Branded search volume. Early citations quietly pull people to Google your brand, and that branded search feeds the same entity-graph loop that makes the next citation land faster. For the diagnostic, the point is simpler: if brand-modifier searches that did not exist before the engagement are creeping up, the loop has started, even while the citation dashboard reads zero.
- Impressions on the new architecture. The resources hub, the answer pages, the structured directory you built in month one should be accumulating impressions in Search Console, even at bad positions. Impressions before clicks is the normal order. No impressions at all is a problem.
- AI bot crawl frequency. ChatGPT, Perplexity, and Google's AI crawlers should be hitting the new pages more often over time. You can see this in your server logs. Rising crawl frequency means the engines are reading the foundation. Flat crawl means they cannot find it or do not care yet.
- Long-tail share of answer. You will get cited on narrow, low-competition questions before you get cited on the head terms. Track share of answer on the long tail. Movement there is the first green shoot.
If three of those four are climbing, the silence is working and you should hold. If all four are flat after a full quarter, the skeptic is right, the foundation is not landing, and you should stop paying for patience. That is the honest line, and most agencies will not draw it for you because the flat-line answer costs them the renewal.
Why everyone quits right before it pays, including the agency
The reason the invisible quarter kills so many programs is not that people lack patience. It is that the incentives on both sides of the table push toward quitting at exactly the wrong moment.
On the client side, the math is brutal. Three months in, finance sees a real number going out and a zero coming back. "We have spent $30,000 and we are not in a single AI answer" is a hard sentence to defend in a budget meeting, and it does not matter that the foundation is two weeks from compounding. The person who approved the program spends down their political capital defending a flat line, and the safest move is to cut it and look decisive.
The agency side is worse, and less discussed. An agency staring at a renewal conversation it is not sure it will win is tempted to manufacture visible motion. Same-day AI Overview pickups look like progress, so the agency pivots to chasing instant wins on easy pages instead of finishing the slow foundation. That feels like content marketing, not the retainer work that actually compounds. The client gets a prettier month-two dashboard and a weaker month-six outcome. Both sides optimize for the meeting instead of the result.
Steel-man the other view, because it is strong: if a program shows nothing after a quarter, maybe that is not patience, maybe it is a failing program, and "wait" is the oldest excuse in the agency playbook. That critique is correct exactly when the leading indicators are flat. It is wrong when branded search, impressions, and crawl are climbing and only the headline citation count is lagging. The difference between a disciplined hold and a sunk-cost trap is the diagnostic in the section above. If you cannot see the leading indicators, you are not being patient, you are guessing, and you should leave.
What the other side of the quarter looks like
The payoff, when you hold the line on a foundation that is actually working, is not linear. The curve is flat for ten weeks and then it bends.
A stablecoin payroll company we work with, Toku, sat in the quiet quarter like everyone else. The foundation went in, the dashboard barely moved, and the first citations did not arrive on anyone's preferred timeline. We held the line. Toku is now cited in about 86% of AI answers on its core buyer question, a 30-day share-of-answer reading at the far end of a curve that began in a quarter exactly like the quiet one I am describing. None of it was visible at the start. All of it was being built at the start. The stretch that looked like nothing was the foundation the 86% now stands on.
That shape, flat then bent, is the normal shape of a working AEO program. It is also the precise shape that punishes anyone who measures it linearly and quits at the bottom of the curve.
The position
The single most important decision in an AEO program is not a tactic. It is not the schema, the answer block, or the engine you prioritize. It is the decision not to quit in the invisible quarter, made for the right reason, with the leading indicators in front of you.
So here is the stance I will defend. If you can only commit one quarter to AEO, do not start. You will spend the money, build the foundation, see nothing, and cut it one month before it pays, which is the most expensive way to do AEO there is. Commit to two quarters or do not begin. And in the first one, stop watching the citation count. Watch whether the foundation is breathing underneath it. That is the only number that tells you the truth while the dashboard is still lying.
