You do not need Slack, an ad account, or another app for this. The Bottleneck Diagnosis worker is already in JeremyAI. You give it the offer and the numbers for a clearly defined period. It names the first drop-off to investigate, and one thing to check before you change the ads or the offer.
Watch the walkthrough
The walkthrough is one run inside JeremyAI. The worker opens, the numbers go in, and the diagnosis comes back in the same chat. If the player does not load, open the walkthrough here.
This example shows a first-pass diagnosis. Check its assumptions against your appointment status, reminder process, and sales notes before making changes. Your answer may differ.
What this does
Takes a real path, such as leads, booked calls, shows, and buyers, for one bounded period.
Works out the rate at each stage instead of guessing from the top of the funnel.
Names a first hypothesis, not a proven cause, and the evidence that would confirm or kill it.
Gives you a short check to run before you rewrite ads, rebuild the offer, or add spend.
Stays in JeremyAI. Nothing is connected, scheduled, or sent out.
When to use it
Use it when you have the offer and numbers for a clearly defined period, and you are not sure whether the problem is the ads, the booking, the show rate, the consult, or the follow-up. Last 7 days, last 30 days, or one launch all work. A single day is usually too thin.
Use a normal chat when you do not have counts yet. This worker is for the next operating call, not a general marketing review.
Before you start
Signed in at
jeremy.utari.ai.The Bottleneck Diagnosis worker visible in your worker list. It is a shared JeremyAI worker, so you do not build it.
The offer, the price, and the counts for each stage in that period. Spend and follow-up speed help, but the stage counts are what make the diagnosis usable.
Do it
Open the Bottleneck Diagnosis worker. Do not open its settings. Those are locked on a shared worker, and this run does not need them.
Paste the path. Name the period, the offer, and the counts. The walkthrough uses the prompt below.
Read the reply as a hypothesis. Write down the stage it names, the rate it used, and the first check it wants. Do not treat the first fix as proven.
Run the check in the next section before you change anything else. Then come back with what you found.
How to read the sample
The walkthrough uses last week for a $2,500 med spa package, about $500 a day on Meta: 90 leads, 42 booked consults, 18 showed, 3 bought. Follow-up was usually 12 to 24 hours. Work the rates yourself so you can see what the worker is reacting to:
42 booked from 90 leads is 47%. Booking is not the obvious leak.
18 showed from 42 booked is 42.9%. That is the largest drop in this path.
3 bought from 18 shows is 17%. The consult is a real second leak. It is not nothing.
On these counts, the first hypothesis is the show rate, not the ads. The reply in the walkthrough says to cut follow-up to under 5 minutes. That is a sensible first check. It is not proof that slow follow-up caused the no-shows. People also miss appointments because the reminder never went out, the time was wrong, the lead was never qualified, or the consult was booked too far out.
What to check before you change the ads
Do this before you rewrite creative, change the offer, or add spend. Use the same period as the prompt.
Appointment status. Split the 24 who did not show into canceled, rescheduled, and no-show. A pile of cancels is a different problem from a pile of silent no-shows.
Reminder process. Confirm the confirmation and reminder actually send, and how long before the consult. A missing reminder will look like a show-rate problem.
Speed to lead. Note how fast the first human or text reached the lead, not the average you think you hit. The 12 to 24 hour line in the prompt is the clue the worker used. Time a real sample of 10 leads.
Sales notes on the 18 who showed. If those consults were a poor fit, the show rate is not the only constraint. The 17% close rate may be the attack point once no-shows are explained.
Bring that back into the same chat. A useful follow-up is: "Here is what the 24 missed appointments actually were, and how fast we reached the last 10 leads. Does the show-rate hypothesis still hold, or is the consult the first fix?"
What not to do yet
Do not scale the $500 a day until the show rate is explained. More leads will not fix a 42.9% show rate.
Do not rebuild the offer from this one reply. The booking rate is already 47%.
Do not treat "cut follow-up to under 5 minutes" as the finished plan. Run it as a 7-day test, then compare show rate on the new bookings.
The prompt from the walkthrough
Copy the block below for the same run as the video. For your own business, keep the shape and replace the offer, the period, and the counts.
I own a high-ticket med spa.
Our main offer is a $2,500 body contouring and skin tightening package.
We spend about $500/day on Meta ads. Last week we got 90 leads, 42 booked consults, 18 showed up, and 3 bought.
Our team usually follows up with leads within 12-24 hours.
I am not sure if the problem is the ads, the offer, the booking process, the show rate, the sales consult, or the follow-up.
Please diagnose the biggest bottleneck and tell me what to fix first.
Write your own in the prompt builder
If you do not want to start from the sample, open the prompt builder. Pick Show rate, then Find the cause first. Paste your period and counts, leave the rest blank, and copy the result into the Bottleneck Diagnosis worker.
The builder marks empty boxes as [MISSING] and tells Jeremy not to assume the fix. Do not assume reminders caused the drop. That matches this article. Nothing you type there is sent until you paste it into JeremyAI.
If the worker is missing
Refresh the worker list. If it is still not there, email [email protected] and include the account email. Do not build a replacement for this walkthrough. Creating your own worker is a different article: Create a Focused JeremyAI Worker.
The how-to for this worker is Bottleneck Diagnosis Worker.
Support
Stuck on a step? Email [email protected]. Include the account email, the worker name, and a screenshot of the chat.