Agency AI Discovery Call Kit
Walk into a discovery call with a plan and fifteen questions, and walk out with structured notes, a
same-day recap and ...
Show more
Instructor
David Tainsh
Course details
Lectures
:
4
Level
:
Beginner
- Description
- Curriculum
- FAQ
- Reviews
Free course · For SMMA, marketing and AI agency owners
Every call ends in a decision — not a maybe you chase until the next year.
Six prompts that put a plan and fifteen questions in front of you before the call, then turn what you heard into structured notes, a same-day recap email, and an honest verdict on whether to write the proposal at all.
- A decision, not a maybeEvery call ends in propose now, go back with questions, or walk away.
- Notes you can price fromTheir words and your read kept separate, with the gaps named rather than filled in.
- An email sent todayUnder 150 words, one question, proving you listened while it still counts.
Who this is for
New and growing agencies who want calls that turn into work.
Whether it is your first sales call or your fiftieth, the risk is the same: the call goes fine, produces nothing you can price from, and ends with you saying you will send something over.
- 01SMMA owners
Calls that arrive from ads or outreach, with a prospect who has not thought about the problem before you asked. Structure is what turns that into something.
- 02Marketing and digital agencies
Selling retained work, where the second decision-maker is always somewhere and almost never on the first call.
- 03AI and automation agencies
Selling something with no budget line yet, so the whole call turns on establishing what it costs them to leave it alone.
- 04Starting out, or going out on your own
No sales background and no script. This gives you a shape you can hold while listening, which is the only part that matters.
- 05Anyone writing proposals into silence
If your last three went unanswered, the fault was on the call rather than in the document.
You do not need a CRM, a call-recording subscription, a sales background or a script. You need one call in the diary.
What you walk away with
Eight things, from one thirty-minute call.
All of it on a real prospect of your own, or one you would like to call, not a worked example. The measure of this course is whether your next call ends in something you can act on rather than something you have to follow up.
- A call plan you can hold in your head while listening
- The three things you must find out, ranked for when it overruns
- Fifteen questions grouped by job, none of them sounding like a form
- Six ranked objections, and the one you cannot yet answer named
- Structured discovery notes with their words and your read kept apart
- A written list of what the call did not establish
- A recap email under 150 words, sent the same day
- A verdict: propose now, go back with questions, or walk away
The notes are also the exact input the free Proposal Generator takes, so if the verdict says propose, the next step needs no rewriting.
Why the call produced nothing
It went well. That is not a signal.
A good conversation feels like progress, so nothing gets fixed. Four things went wrong, none of them were charisma, and all four are decided before you dial rather than during.
- ×You took the stated problem at face value
They led with the thing that annoyed them that week. It is almost never the thing worth paying to fix, and the real one surfaces about ten minutes in.
- ×You never found what it costs them
No number means no reason to act this quarter rather than next year. A shrug is not a cost, and a proposal into a shrug goes unanswered.
- ×Somebody else was mentioned once
In a subordinate clause, in passing, and never followed up. That person reads your proposal cold and is usually the one who kills it.
- ×You wrote it anyway
Because the call felt good and you liked them. A week writing, a month chasing, and a pipeline full of polite maybes that makes you feel busy and broke.
Fix them by deciding what the call has to come back with before it starts. That is the whole method, and it takes twenty minutes the day before.
The trap
A conversation is not discovery.
The instinct on a first call is to be interesting. To explain who you are, what you have done and how you work — which is exactly what stops the other person thinking out loud, and thinking out loud is where everything useful arrives.
The conversation
- Opens with ten minutes on your credentials
- Solves the problem they led with
- Never establishes a cost
- One person, no idea who else is involved
- Ends with “I’ll send something over”
The diagnosis
- Opens with one line, then a question
- Finds the problem underneath the stated one
- Establishes cost in their own figures
- Names the second decision-maker
- Ends with an agreed next step and a date
The through-line
AI structures the call. It cannot listen for you.
The prompts can plan, sort and decide. What they cannot do is be in the room — and everything that goes wrong with generated discovery notes goes wrong on the wrong side of that line.
- 1You ask and you write down
Their phrases rather than your summaries, numbers exactly as said, and a question mark wherever you were not sure. A summary is already an interpretation.
- 2The prompts sort and separate
Nine headings, with what they said kept apart from what you concluded — so in a week you can still tell which was which.
- 3Gaps stay gaps
Anything you did not actually hear is written down as unknown rather than filled in with something reasonable. That gap becomes your follow-up question.
The most expensive habit in discovery is quietly upgrading an implication into a fact. A figure mentioned in passing becomes the budget; “we” becomes sole authority.
The framework
Six things a call has to produce.
If it produced none of these, it was a chat. Learn the six and you stop guessing at what went wrong when a proposal goes quiet, because the answer is always one of them.
- 1The real problem
What is actually wrong, rather than the thing they led with. It usually arrives ten minutes in, once they stop answering and start thinking.
- 2What it costs them
Money, hours or work lost, in their figures rather than your estimate. No cost means no reason for anybody to move.
- 3Who else has to agree
The partner or co-owner who was not on the call. Almost always someone, almost always mentioned once.
- 4Why now
What changed this month. A date, a season, a hire, a contract. Without one it waits a year and sits in your pipeline.
- 5What they can spend
Established without ever asking the budget question — by finding what they were already about to spend on the alternative.
- 6What you still do not know
Written down as a gap. Guessed at, it becomes the reason the proposal misses by an amount you cannot explain.
Number five is the practical trick of the whole course. Nobody answers the budget question honestly, and everybody tells you what they were about to spend instead.
What’s included
Every lesson comes with the prompts that build it.
The classes are short. The work happens in the prompt pack — a numbered SOP, the six prompts in order, and a note on what a good output actually looks like so you can tell when to run one again.
- SOPA step table in the pack
Seven numbered steps. Three before the call, four within a few hours of it ending. Work down it and the call produces something.
- AICopy-paste prompts, in plain English
No codes and no variables to install. Square brackets mark the few things only you can supply.
- QAWhat good looks like
A concrete standard for every prompt — including how to tell when your notes have come back more confident than the call justified.
- DEMOA real thirty-eight-minute call
Apex AI and Pebble & Co, transcript and scribbled notes included, with three things the call never established.
How the course runs
Three short lessons, then the whole chain run live.
Twenty minutes of preparation the day before, thirty minutes on the call, and about an hour afterwards. That is the entire commitment, and it is the difference between a call you can price from and one you cannot.
Lesson one
Why discovery calls fail
The six outputs, why the stated problem is a symptom, and how to establish what they can spend without asking. Nine minutes.
Lesson two
Before the call
Prompts 01 to 03. A call shape, fifteen questions grouped by job, and the six things this person will push back on. Nine minutes.
The call
Ask, do not present
One line on who you are, then a question. Twenty of the thirty minutes should be them talking.
Lesson three
After the call
Prompts 04 to 06. Structured notes with the gaps named, the recap email, and the verdict. Ten minutes.
Live demo
All six prompts, end to end
The full chain on a real call, including the notes that come back too confident and the instruction that corrects them. Four minutes.
The live demo
Including the notes that came back confidently wrong.
Most demonstrations show the run that worked. This one keeps the output that did not, because a tidy set of notes containing three things nobody said is the most expensive failure in the whole course — and it is far easier to recognise once you have watched it happen to somebody else.
The call
Thirty-eight minutes
- Apex AI and Pebble & Co, a Clapham gym
- Booked off a cold opener she replied to
- Transcript and scribbled notes, both included
What it hid
Three things
- The stated problem was not the real one
- The cost was a shrug, never pressed
- A second decision-maker, named once
The verdict
Not propose
- Go back with questions, and why
- Which three questions, and where they go
- What that saved: a week and a month
Is it for you
Agency owners who will take the answer they did not want.
A good fit
- You have calls booked, or soon will, and want to know afterwards whether the work is real
- You will spend twenty minutes preparing rather than winging it
- You will ask rather than present for the first ten minutes
- You want notes you can still read honestly in a week
- You will accept a verdict of walk away when it is correct
Not a fit
- You want a script to read from during the call
- You want an AI notetaker rather than a way of running the call
- You would rather have a full pipeline than an honest one
- You will write the proposal regardless of what the verdict says
- You are looking for closing techniques rather than diagnosis
What it costs
Nothing, and nothing attached to it.
Free
No card. Lifetime access, including anything added later.
4 lessons · 33 minutes · 6 prompts · prompt pack · demo call
No upsell inside the lessons. There is a paid course this leads to, and it is mentioned once, at the end, after you already have a call you can price from.
Quick answers
The questions people ask before starting.
Is this a sales course?
No. There are no closing techniques in it and no objection-handling scripts. It is about finding out what is true before you decide whether to sell anything, which is a different skill and an easier one to learn.
Do I need to record my calls?
It helps, and asking plainly at the start means almost nobody refuses. If they do, scribbled notes work fine — prompt 04 is built to take a messy page and sort it, and the demo includes exactly that.
Which AI does this use?
Claude. The prompts are plain English with square brackets marking the few things only you can supply. Nothing to install, no account to connect, no template to configure first.
What if I have no calls booked?
Run it against a prospect you would like to call. The preparation is still real and it makes booking one easier — and the free Cold Outreach course is the one that fills the diary if that is the actual problem.
Is it not rude to follow a plan?
A plan is not a script. It is three ranked priorities and a shape, so that when the call overruns you sacrifice the right thing. The people who sound rehearsed are the ones reading questions off a screen.
What if the verdict says walk away?
Then you have saved a week of writing and a month of following up, which is the single most valuable output in the course. A pipeline of polite maybes is what makes agency owners feel busy and broke at the same time.
How does this connect to the proposal?
The notes from prompt 04 are the exact input the free Proposal Generator takes. Same structure, no translation. If the verdict says propose, you paste and carry on.
What exactly do I get?
Four lessons totalling thirty-three minutes across 45 slides, six plain-English prompts, the prompt pack and SOP in PDF and Word, a real thirty-eight-minute demo call with transcript and notes, and lifetime access including anything added later.
Stop leaving calls with a feeling. Leave with a decision.
Twenty minutes of preparation, thirty minutes of asking, an hour afterwards — and an answer you can act on rather than a proposal you hope somebody reads.
A Digital Plan · adigitalplan.com
Agency AI Discovery Call Kit
-
1
Lesson 01 — Why discovery calls failYou liked each other, the conversation flowed, and you said you would send something over. Three weeks later you were writing a proposal from memory for somebody who turned out not to be able to sign it. That is not a charisma failure. It is what happens when nobody decided, before dialling, what the call had to produce.
This lesson makes the case for treating a discovery call as a document you are collecting rather than a conversation you are having. You will learn the six outputs, why the problem they lead with is almost never the one worth fixing, and how to establish what they can spend without ever asking about budget.In this lesson
• See why a call that went well and a call that produced something are different things
• Learn the six outputs, and which one you missed when a proposal goes silent
• Find the real problem underneath the one they opened with
• Establish what they can spend by asking what they were about to spend on instead
You finish with: The framework and the argument behind it. Nothing to run yet.
-
2
Lesson 02 — Before the call
Twenty minutes of preparation the day before changes what the call produces. Not a script — those make you sound like a bank — but three ranked priorities so you know what to sacrifice when it overruns, the questions in your head rather than on your screen, and a named list of what they will challenge.
You will run the first three prompts and meet the demo: Apex AI calling Pebble & Co, the same prospect the Cold Outreach course wrote the strongest opener to. She replied, and this is the call that reply produced.In this lesson
• Run prompt 01 and read the assumption it flags before anything else
• Learn the three questions you ask no matter how short the call runs
• Avoid the three questions that get an answer which is not true
• Run prompt 03 and fix your weakest objection answer, or accept it is coming
You finish with: A call plan, fifteen questions, and the objection you cannot yet answer.
-
3
Lesson 03 — After the call
Run this half within a few hours of the call ending, while you can still tell the difference between what was said and what was implied. Three days later an interpretation you made on Tuesday reads exactly like something the client said, and you will price against it with total confidence.
The lesson ends on the step most people skip: asking whether to propose at all. The prompt is forbidden from hedging and returns one of three verdicts. On the demo call it returns go back with questions, because a second decision-maker was named once and the cost of the problem was a shrug rather than a number.In this lesson
• Turn a messy transcript into nine structured headings, gaps named rather than filled
• Keep their words and your interpretation separable a week later
• Send a recap under 150 words today, with exactly one question in it
• Take the verdict seriously, including when it says walk away
You finish with: Discovery notes, a sent recap email, and a decision you can defend.
-
4
Lesson 04 — Live demo — all six prompts, end to end
No new teaching. The whole chain run start to finish on screen, using Apex AI and the Pebble & Co call from the demo pack — so you can watch what each prompt actually returns before you spend a call finding out.
It includes one output that comes back too tidy, and the follow-up instruction that fixes it. That is deliberate. A confident set of notes containing three things nobody said is the most expensive failure in this course, and it is much easier to recognise once you have seen it happen to somebody else.In this lesson
• Watch prompt 01 name an assumption that turns out to be wrong
• See nine lines of shorthand become structured discovery notes
• See the notes come back too confident, and the instruction that corrects them
• Follow the verdict to go back with questions rather than propose
You finish with: A clear picture of what good looks like at every step, before you run your own.
Is it really free?
Yes. No card is needed and you get lifetime access, including anything added later. A paid course is mentioned once, at the end, after you already have a call you can price from.
Is this a sales course?
No. There are no closing techniques or objection-handling scripts. It is about finding out what is true on the call before you decide whether to sell anything.
What do I come away with after a call?
Structured notes with their words and your read kept apart, a list of what the call did not establish, a recap email under 150 words sent the same day, and a verdict: propose now, go back with questions, or walk away.
How much time does it take per call?
About twenty minutes of preparation the day before, thirty minutes on the call, and about an hour afterwards. The lessons themselves total 33 minutes.
Do I need to record my calls?
It helps, but it is not required. Prompt 04 is built to take a messy page of scribbled notes and sort it, and the demo includes exactly that.
Do I need a CRM or a sales background?
No. You do not need a CRM, a call-recording subscription, a sales background or a script. You need one call in the diary, and the prompts run in Claude.
What if I have no calls booked yet?
Run it against a prospect you would like to call, as the preparation is still real. If filling the diary is the actual problem, the free Cold Outreach course covers that.
How does it connect to writing a proposal?
The notes from prompt 04 are the exact input the free Proposal Generator takes, in the same structure. If the verdict says propose, you paste them in and carry on.
Please, login to leave a review