Every New Member Gets
A Personal Welcome Video.
When someone joins one of my programs, an AI clone of me records a personal welcome video on the fly. It calls them by name, mentions the date they joined, the package they took, and the goal they came for. I never hit record, and they always feel seen.
The welcome that finally feels personal.
Most onboardings are a templated email, a login link, and silence. The member joins, pays, and is left to figure out the rest. The single moment when goodwill is at its peak (the seconds right after they say yes) gets handed to an automated email that reads like every other automated email. There is a different way. Let a small automation generate a personal welcome video on the fly, with their name and their goal in it, recorded by an AI clone of you, then send it by email a few seconds later. Here is who it is for, what goes wrong without it, how it works, and what you get back.
Who it’s for
Course creators, membership owners, consultants, coaches, and agencies who sell anything with an onboarding moment. Especially useful if you have a recurring offer, because the welcome video is the first lever on retention. If a real human shaking the new member’s hand would change how they feel about week one, this is for you.
What goes wrong
Without the welcome video, the new member gets a receipt, a login, and the same canned email everyone else gets. The peak goodwill moment passes in silence. In week two, the doubt sets in, the refund window is still open, and there is nothing personal to point back to. A new member who does not feel seen in week one is a refund or a non-renewal waiting to happen.
How the machine works
The signup webhook fires. The automation pulls the new member’s first name, the date they joined, the package they bought, and the goal they wrote down. It writes a short script with those fields baked in. It sends the script to an AI avatar API trained on a few minutes of my real footage. The video renders in a few seconds. It gets attached to a warm welcome email and sent. You sleep. They feel seen.
What you get back
Roughly 8 to 15 personal welcomes a day, at about forty cents a render. Retention up by a few points in the first 90 days. Refund rate in week one quietly down. About one extra renewal a week saved, worth around 97 euros each, which is around 5,000 euros a year kept. Compounded over five years, that is 25,000 euros plus, for an automation that runs without me.
The welcome moment was going to waste.
For a long time my onboarding was a receipt and a login link, like everyone else’s. People would pay, the platform would send them the standard welcome, and then I would not speak to them again until they wrote in with a question. The single highest goodwill moment in the whole customer journey, the few minutes right after they said yes, was being spent by an automated email that nobody remembers. The cost was not loud. There was no error in the logs. It just leaked, quietly, week after week.
I could see it in the numbers if I looked. People who joined and felt nothing in the first week were the same people who refunded in week two, or who did not renew at month three. The receipt does not create a relationship. The login link does not create a relationship. A real person taking thirty seconds to say “hi, glad you are here, here is what week one looks like” would have changed the feeling. But there was no way I could do that for every new member. I have 8 to 15 new ones a day on a good week.
So I built a small automation. The moment someone joins a program, the signup webhook fires. The automation pulls their first name, the date they joined, the package they bought, and the goal they wrote on the order form. It writes a short script with those fields baked in. It sends the script to an AI avatar API that has been trained on a few minutes of my real footage. The video renders in a few seconds. It gets attached to a warm welcome email and sent. The whole thing runs without me.
The cost is about forty cents per render. The time I spend on a welcome is zero. Every new member now opens an email a minute after joining and watches a short video of me calling them by name, mentioning the date they joined and the goal they came for, and pointing them to the first thing to do. People reply. They say they were not expecting that. A few of them ask if I really sat down and recorded it. Most of them just feel seen, and they stay.
Proof point: the mechanics behind this automation are taught inside Automations Made Easy, the same playbook 1,000+ students use to build small machines like this one without writing code. The welcome video runs on my real audience, and the retention and renewal numbers on this page come from the same offers I sell, not a demo account.
Three moves that turn every signup into a moment the member remembers
What made this work was treating the welcome video as a small handshake, not a piece of content. The automation is not a marketing tool in the open sense, it is a stand-in for the human gesture I would have made at every signup if I had been able to. It hangs on three moves: pull the right fields off the signup, write a script that uses those fields like a friend would, and render it through an AI avatar that sounds like me on a normal day. Done right, it costs about forty cents per render, lifts retention by a few points in the first 90 days, and quietly keeps around one extra renewal a week (about 5,000 euros a year, around 25,000 euros plus over five years).
Pull the dynamic fields that make the welcome feel personal
The automation starts by reading the signup payload and pulling the few fields that turn a generic welcome into a personal one: first name, date joined, the package or product they took, and the one-line goal they wrote on the order form. The trick is to pull the goal field, not just the name. A welcome that mentions the goal lands ten times harder than a welcome that only uses the first name. Without the goal, the video sounds like every other mail-merge. With the goal, the member realizes you read what they wrote.
Write the script like a friend would, with the fields baked in
The script is short, 40 to 60 seconds at most. It says hi by name, names the date and the package, names the goal they came for, names one small thing to do in the first 24 hours, and ends with a warm line. There is no sales pitch and no marketing voice. The whole point is to sound like a friend who happens to be the founder. The script template has placeholders for every dynamic field and conditional fallbacks for empty fields, so a missing goal does not break the sentence, it just gets skipped.
Render through an AI clone that sounds like you on a normal day
The last move is the render. The script goes to an AI avatar API trained on a few minutes of my real footage. The voice carries my cadence, the face carries my expressions, and the energy stays at the level of a relaxed voice note rather than a polished sales video. That is the right level. A polished render reads as marketing. A relaxed render reads as the founder taking a minute. The output is a short MP4 that the automation attaches to a warm welcome email and sends, all inside the first minute after signup.
Once those three moves are in place, the welcome stops being a templated email. It becomes the moment in the whole journey that the member is most likely to mention to a friend.
Before the engine
- The welcome was a receipt, a login link, and a templated email everyone else gets
- The peak goodwill moment was spent on automation that nobody remembers
- Refund rate in week one was higher than it needed to be, with no warm point to refer back to
- Non-renewals at month three crept up because nobody felt personally welcomed in week one
- I could not scale a real welcome, so I shipped no welcome at all and lost the moment
After the engine
- Every new member opens an email with a video of me saying their name and their goal
- The peak goodwill moment is captured by a 40-second handshake that runs without me
- Refund rate in week one drops because the member already has a personal moment to point back to
- Retention in the first 90 days lifts by a few points, around one extra renewal a week kept
- The whole thing costs about forty cents per render, around 5,000 euros a year kept
Prompt 1: extract the dynamic fields off the signup payload
The signup webhook gives you a lot of fields, most of them useless for the welcome video. Use this prompt to pull only the fields the welcome script needs, with safe fallbacks for empty fields, so the rest of the automation works against a clean dataset.
Dynamic field extractor for the welcome script
Act as a workflow operator. I have a signup webhook payload from a new member of one of my programs, and I need a clean dataset to feed a welcome-video script. Give me only the fields the script needs. Full signup payload as JSON: [paste] The list of programs I sell and their internal names: [paste] The goal-field name on my order form (the line where they wrote what they want to achieve): [paste] My fallback text for each empty field (one short line per field): [paste] Return a single JSON object with exactly these keys: first_name, join_date (formatted day month year), package (the internal program name), goal (the one-line goal they wrote, trimmed to 120 characters). If a field is empty, use the fallback I gave you, not a blank string. Do not invent values. Do not return any other keys. Return nothing else, just the JSON.
The output is a small JSON object the script writer in prompt 2 can read directly. Lock the extractor once, every future signup is parsed the same way.
Prompt 2: write the 40-second welcome script in the founder’s voice
The dataset is clean. Now use this prompt to write the welcome script that the AI avatar will read. The whole point is that it sounds like a friend, not a marketing video, and that the dynamic fields land where they matter.
Personal welcome script writer
Act as a copywriter writing a 40 to 60 second video script for a personal welcome video that a founder is sending to one specific new member. The voice should sound like a relaxed founder leaving a friendly voice note, not like a sales video. The new member's dynamic fields as JSON: [paste output of prompt 1] A short founder-voice profile (how I sound when I am being a friend, three sentences): [paste] The one small action I want them to take in the first 24 hours: [paste] Two phrases I never say (banned wording): [paste] Write the script as a single block of spoken text, no scene directions, no stage notes. Start with hi by first name. Mention the join date and the package in one short sentence. Mention the goal they came for in the next sentence, in their own words if possible. Name one small action they should take in the first 24 hours. End with one warm line, no sales pitch, no closer. Keep sentences under 14 words. The whole script must read aloud in 40 to 60 seconds.
The output is the spoken text the AI avatar API will read in the next pass. Lock the script template once, every member gets a script that sounds like a friend.
The 3-minute overview of how this works
Before the build steps, watch this short overview. It is the exact video from the Automations Made Easy page, and it walks through the mechanics behind machines like this one. 1,000+ students have used these mechanics to save two hours a day, with zero coding.
Want to learn the mechanics behind personal welcome engines like this one?
I teach the same mechanics that make this welcome engine run inside Automations Made Easy. 1,000+ students have used these mechanics to save two hours a day and turn the cold parts of their onboarding into moments people actually remember. No coding required.
Week one: about 60 personal welcomes, none recorded by me.
Most people quit a system like this in the first week because the first welcome video feels too quiet. There is no traffic spike, no inbox flood. Just a few replies from new members saying they were not expecting that. The first week is meant to feel small. The point is to confirm the webhook fires on every signup, the dataset is clean, the script reads like a friend, the AI clone renders in seconds, the email lands, and the member opens it on their phone.
Week one looks like this. About 8 to 15 new members join the various programs each day. That is roughly 60 personal welcomes across the week. Each one renders for about forty cents, so the whole week of welcomes costs around 25 euros. None of them are recorded by me. Each video calls the member by name, names the date they joined, mentions the package they took and the goal they came for, and points to the first small action in week one. Reply rates climb. Refund requests in the week one window drop. Quietly, in the background, while I work on the next program.
The point of the first week is not the open rate. The point is to prove the loop closes: signup fires, fields extract clean, script reads like a friend, AI clone renders in seconds, email lands warm, the member feels seen at the moment they joined.
From there the maths is simple. Around 8 to 15 new members a day across the programs is roughly 3,500 personal welcomes a year, at around forty cents each, around 1,400 euros a year in render cost. Retention in the first 90 days lifts by a few points. That works out to around one extra renewal a week kept, worth around 97 euros each, around 5,000 euros a year. Compounded over five years, that is around 25,000 euros plus kept on the books, for an automation that runs without me.
All of this runs while you sleep, take a week off, or write the next program. The webhook watches the signups, the extractor cleans the fields, the writer drafts the script, the clone renders the face, the mailer sends the email. The welcome stops being the moment you missed, and starts being the moment members mention to a friend.
Prompt 3: pick the fallback rules for empty dynamic fields
Dynamic fields are only as good as their fallbacks. Use this prompt to design the fallback rule for every field so the script still reads like a friend even when the order form was half-filled. The whole point is that no broken sentence ever ships.
Fallback rule designer for the welcome script
Act as a careful copy editor for an automated welcome script. The script reads a dataset that sometimes has empty fields, and I need a fallback rule for every field so the spoken script still sounds like a friend, never a glitch. The dataset schema as JSON (first_name, join_date, package, goal): [paste] The script template with placeholders (the same one prompt 2 uses): [paste] My founder voice profile (three sentences): [paste] Return a JSON object with one key per field. For each field give me three things: the fallback string to use when the field is empty (kept short and warm), the rewriting rule for the sentence that uses that field (whether to drop the sentence, swap it for a generic line, or skip the field but keep the sentence), and an example of the full spoken sentence with the fallback in place. Do not return values that sound like a system message or an apology. Keep every fallback in the founder voice.
The output is the fallback profile the writer in prompt 2 reads at the top of every run. Lock it once, every future signup is handled cleanly.
Prompt 4: write the warm email body that the welcome video rides in
The video is the moment, but the email is the envelope. Use this prompt to write the email body the AI welcome video rides in, so the member opens, watches, and then knows the one small thing to do next.
Welcome email body writer
Act as a warm onboarding email writer. I have a personal welcome video that was just rendered for a new member, and I need the email body the video rides in. The email is short, warm, names the member, points to the video, and ends with the one small first step. The new member's first name and the program they joined: [paste] The link or attachment for the personal welcome video: [paste] The one small action for the first 24 hours (the same one in the script): [paste] My standard email signature line (one sentence): [paste] Return the email body as plain text, prepared to paste into the mailer. Subject line in the first row. Body underneath. Start with hi by first name. Say in one short sentence that I recorded a short welcome video for them. Put the watch link or attachment cue right after. Then in one short paragraph, name the one small first step for the first 24 hours. End with the standard signature line. Do not use marketing closers. Keep paragraphs short, two lines maximum.
This is the email body the mailer sends, every signup, every time. Lock it once, every future welcome ships with the same warmth as the first one.
The exact build, step by step
Wire the signup webhook that fires on every new member
The trigger is the source of truth for the whole automation. Point a small workflow (n8n, Make, or Zapier all work) at the signup event of your platform: ThriveCart, Stripe, Memberstack, Circle, Kajabi, whichever sells your programs. Every new member fires the webhook once. Filter so the workflow fires only on completed payments, not on abandoned checkouts. The trigger does not render anything yet, it only carries the new member’s signup payload forward to the next step. This single piece is what makes the whole engine event-driven instead of scheduled.
Extract the dynamic fields off the payload with safe fallbacks
When the webhook fires, the workflow reads the payload and pulls the four fields the welcome script needs: first name, join date, package, and the one-line goal they wrote on the order form. Every field has a fallback rule, so an empty goal does not break the script, it just gets skipped. The output of this step is a small clean JSON object the script writer in the next step can read directly. The dataset is small on purpose, four fields not forty. Most of what the payload sends is noise for a welcome video.
Write the 40-second script in the founder voice
This is the most important pass in the whole engine. The dataset goes through the script writer in prompt 2, which produces a short spoken script in the founder voice. It says hi by name, names the join date and the package in one short sentence, names the goal in the next, gives one small first action, and ends warm. No marketing voice. No closer. The whole point is that it reads like a friend leaving a voice note, not a sales video. That voice is what makes the AI render land as personal instead of as content.
Render through the AI clone API trained on my real footage
This is where the engine builds the actual video. The script gets sent to an AI avatar API (HeyGen, Synthesia, D-ID, all work) trained on a few minutes of my real footage. The clone keeps my voice, my cadence, and my expressions. The render takes around 20 to 40 seconds for a 40 to 60 second clip. The cost is around forty cents per render at the volume I run. The output is a short MP4 file the next step will attach to the welcome email. No editing pass. No human review. The whole point is that it runs without me.
Build the warm welcome email with the video attached
Before anything ships, the workflow builds the welcome email the video rides in. The subject names them by first name. The body is short, names the package, points to the video, and ends with the one small first action. There is no marketing copy, no offer, no closer. The whole point is to keep the email out of the way so the video does the work. The email body is built by the writer in prompt 4. The video file is either attached or uploaded to a short link the email points to. The mailer holds the send for one final check.
SEND the welcome email and log the open / reply for retention
The last step is a single send. The mailer sends the warm welcome email to the new member with the personal video sitting inside it. The whole loop, from signup to inbox, takes under a minute on a normal day. The workflow then writes a small row to a tracking sheet: the member’s first name, the join date, the package, the goal, the video URL, the send time, and a field for whether the email was opened and whether the member replied. That tracking row is what lets you see retention by cohort later. No log in. No Studio tab. The welcome is done before they have closed the receipt page.
Webhook fired
The platform sees the completed signup, the workflow fires once with the signup payload.
Fields cleaned
First name, join date, package, and goal pulled off the payload with safe fallbacks, the dataset is four fields wide and ready for the writer.
Script written and rendered
Founder-voice writer drafts a 40 to 60 second script, the AI clone API renders the short MP4 in around 30 seconds for about forty cents.
Email sent and logged
The warm email goes out with the video sitting inside it, the retention row is written to the tracking sheet, the welcome is live before the receipt page closes.
Build this personal welcome engine inside the same playbook 1,000+ students use
Automations Made Easy teaches the mechanics behind engines like this one. Step by step, no code, plain English. Turn the coldest part of onboarding into the moment your members are most likely to mention to a friend.
The six months after I switched it on
Here is the shape of the first six months after I turned the personal welcome engine on. The line is intentionally modest in the early months and steady from there, because that is how retention actually behaves: every cohort of new members adds a small layer, and the layers stack.
Retained revenue from extra renewals kept, per month (in euros)
Caption: monthly retained revenue from extra renewals kept on cohorts the welcome engine touched. Three things matter on this chart. The line stabilises near 5,000 euros a year (around 415 euros a month) and stays there. The revenue comes from members who would have quietly drifted off without the welcome. And every single one of those months happens without me recording a single video.
What other students built
I teach the simple skills behind machines like this in Automations Made Easy. Students who built their own version sent back what changed in their first month.
“My welcome used to be a Stripe receipt and a login link. Now every new client opens a video of me by name, naming their project. Two of them replied that it made them feel like they had picked the right person.”
“Refunds in week one dropped to almost zero on my course after I switched this on. The welcome video gives them a moment to point back to when the doubt creeps in. Same product, same price, fewer refunds.”
“I run a small membership and I cannot personally welcome every new member. This engine does it for me, in my voice, with their goal in it. Renewals at month three lifted by a few points across the first cohort.”
“My new B2B clients now get a 60-second welcome that names their company and their project. Two of them forwarded it inside their company. That is the kind of word-of-mouth I cannot buy with ads.”
What’s inside Automations Made Easy
AME isn’t a library of pre-built automations. Every business is slightly different. What’s reusable across all of them is the underlying mechanics: how to set up little machines that listen, write, and follow up while you sleep, and how to wire the pieces together without writing code.
The program walks you through six modules: The Right Tools (the cost-effective, no-code stack I actually use), Task Selection Mastery (which automations are worth building first), Design Secrets (mapping an automation before you build it), Zero to Hero (complete beginner to confident automator), Real-World Application (we build a full automation together, end to end), and Monetization Mastery (turn the skill into a side-business).
It also includes done-for-you templates you import in two clicks, over-the-shoulder training videos, and the same playbook 1,000+ students have used to save two hours a day. No coding required. If you can copy and paste, you can build this.
The personal welcome engine: common questions
Pulled from what readers and Automations Made Easy students ask most.
Won’t people feel tricked when they realize the video is AI?
No, and the reason is the work the video is actually doing. The video is short, it calls them by name, it names the date they joined, the package they took, and the goal they came for. None of those details are generic. They feel seen, which is the whole point of a welcome. Most people never ask whether it was recorded live or generated, because the feeling lands either way. If someone does ask, I tell them, and they almost always reply that they love it even more knowing it scales. The risk is not honesty, the risk is silence at the moment of joining.
How realistic does the AI video actually look?
Realistic enough that unless I point it out, members assume I sat down and recorded it for them. The AI clone is trained on a few minutes of my real footage, so the voice, the cadence, and the face all carry across. The trick is to keep the script short and the energy honest. A 40 to 60 second welcome with their name and their goal lands cleaner than a two-minute attempt at acting natural. I write the script as the kind of voice note a friend would leave, the clone delivers it, and the result reads as personal because the content is personal, not just the face.
What if the dynamic fields are empty for some users?
The automation has a fallback rule for every field. If the first name is missing, the script switches to a warm greeting that does not name anyone. If the goal field is empty, the line about the goal gets dropped and the script moves on. If the join date is missing, the script says today. The point is that no broken sentence ever ships. The script is built with conditional blocks, not blind merges, so the worst-case version still sounds like a clean welcome rather than a glitch. The dynamic version is the bonus, the clean version is the floor.
Does this work for B2B services or only B2C products?
It works for both, and B2B is often where it lands hardest. A new B2B client signing a retainer is more likely to be on the fence in week one than a B2C buyer who already paid for a digital product. A personal welcome video that names their company, the project they came in for, and the next step in the onboarding lowers the buyer’s remorse window and gives them something to forward to a colleague. The B2B version is usually a touch longer, about 60 to 90 seconds, and the dynamic fields include the company name and the project scope rather than just the first name.
How much does each rendered video cost me?
About forty cents per render at the volume I run, which is roughly 8 to 15 new members a day depending on the season. The cost has two parts: the AI avatar render through the video API, and the small AI call that fills the script with the dynamic fields. At those numbers, a hundred welcome videos cost around forty euros, and the retention lift from those welcomes is worth far more than that. The automation pays itself back inside the first month of any program with a recurring offer, and from month two onward the cost is rounding error against the goodwill it earns.
More Money Makers like this one
Built from the same handful of mechanics. Each one captures, converts, or earns money that would have walked out the door.
Run this welcome engine yourself, or learn the mechanics inside Automations Made Easy.
If you want to learn the mechanics behind personal welcome engines like this and build your own at home, Automations Made Easy is the playbook. Step by step, no code, plain English. If you want to talk through which webhook, dynamic fields, and avatar voice would work for your specific programs first, I take a small number of consulting clients each month.
1,000+ students · Save 2 hours a day · No coding required
