Recon is the work you do before you talk to a single business. Choose the market, find out what those owners are actually struggling with, and score it honestly. Do this properly and the offer in Mission 02 writes itself. Skip it and you spend three months pitching people who were never going to buy.
Lesson 1 of 3
Picking a market you can win
Most beginners pick a market because it sounds cool or because someone on YouTube said it prints money. Then they find out the buyers do not exist on Meta, or the jobs are too small to pay you. Pick on evidence instead. This lesson gives you the filters and the shortlist.
01 Market first, offer second
You cannot write an offer until you know who it is for. The market decides your price, your guarantee, your ad angles and how hard your job is going to be. Everything downstream is locked by this one choice, so it gets its own mission.
One market. Not three. You are going to say the same thing to the same kind of business until you have proof, and then you are going to use that proof to get the next one. Three markets means three sets of proof, three sets of research, three offers, and nothing stacks.
The rule
One market. Stack proof inside it. Expand only when you have results you can show without flinching.
02 Lead generation, not ecommerce
This is the first fork and it catches nearly everyone. Two completely different businesses hide behind the words "run ads".
Lead generation
Local service businesses. Roofers, dentists, landscapers, builders. They need phone calls and booked quotes. They already pay for leads somewhere, usually badly.
How you get paid: a retainer, every month, in advance. Ad spend is separate and goes straight to Meta on their card.
Ecommerce
Online stores selling product. Sexier, better looking case studies, far more competitive.
How you get paid: usually they push for revenue share, which means you carry the risk, fund the learning and earn almost nothing for months.
Start lead generation. It is not a downgrade, it is the version that pays you while you are learning. Service businesses have a real number attached to every lead, which means you can promise something specific and prove you hit it.
If an ecommerce brand approaches you anyway
Take it only if all three are true: they are already doing real revenue, the ad account is funded and the inventory exists, and there is a base fee before any revenue share. Never a pure revenue share. If any one of those fails, tell them to come back when it does. That is a real answer, not a rejection.
03 The five filters
Run every candidate market through these. A market that fails filter one or filter two is not a market you can win in yet, no matter how much you like it.
1
Ticket size
What is one customer worth to this business? A roof is worth thousands. A haircut is worth thirty dollars. High ticket means you can guarantee fewer leads and still make them money, and your fee looks small next to the job value. Low ticket means you have to deliver volume every single month to stay worth paying.
2
Meta fit
Meta is interruption. People are scrolling, not searching. That works when the need can be created or reminded: a homeowner who has been putting off a job, a person who sees a before and after and thinks about their own house. It works badly when the need is an emergency someone types into Google at the moment it happens.
Test it on yourself. If the only way a customer buys this is by searching for it in the ten minutes after something breaks, Meta is the wrong channel and no amount of good creative fixes that.
3
Reachable volume
Enough businesses to fill a prospect list, and enough of their customers inside a geography you can target. You are not limited to your own town. A market being "exhausted" almost always means the person ran out of people near them, not that the market ran out.
4
Ability to fund spend
They pay you, and separately they pay Meta. If a business cannot put real money behind the ads every month on top of your fee, they are not a client yet. Decide the floor you will work with before you start calling, so you are not negotiating it live on a call.
5
Proof you can get
Can you picture the case study? A number of booked quotes, a cost per lead, a job value. If you cannot describe what winning looks like in one sentence with a number in it, you will not be able to sell it or defend it later.
04 A shortlist to start from
These are service markets that pass the filters often. The point is not that these are the only options. It is that you can see how ticket size and channel fit change the difficulty of the promise you have to make.
Market
Ticket
Why it works, or does not
Pool builders
Very high
One job funds months of fees. Few leads needed. Long sales cycle, so set expectations early.
Custom home builders
Very high
Same shape. Slow, high value, visual work that performs on Meta.
Roofing
High
Storm and season driven. Homeowners delay it, which is exactly what interruption is good at.
Remodeling and additions
High
Before and after creative is native to the feed. Long consideration.
Med spa
Mid to high
Visual, repeat purchase, strong on Instagram. Watch ad policy on body and health claims.
Portable buildings and sheds
High
Considered purchase, big visual appeal, buyers research for weeks.
Landscaping and hardscaping
Mid to high
Project work photographs well. Avoid the mow and blow end of it.
Pressure washing and exterior cleaning
Low to mid
Easy to fill the calendar, thin margin per job. You need volume every month to stay valuable.
Junk removal
Low to mid
Same trade off. Fast wins, harder to hold a real fee long term.
Auto repair
Mid
Weak Meta fit. Most demand is someone searching the moment the car breaks. Rank it near the bottom unless it is fleet or a specialty shop.
Read the pattern, not the list
The higher the ticket, the fewer results you have to promise, and the easier your offer is to say out loud. That is the whole reason ticket size is filter number one.
05 Your task
Lesson 10 / 5 complete
Next
You have a shortlist. Now stop guessing what those businesses struggle with and go find out, in their own words. That is Lesson 2.
Lesson 2 of 3
Finding what actually hurts, without guessing
Every bad ad in this industry was written by someone imagining what a customer cares about. You are not going to do that. You are going to collect what real buyers and real owners already said, in public, in their own words, and build from that. It takes about two hours and it is the single highest leverage research you will ever do.
01 Two different pains, do not mix them
Buyer pain
What the homeowner or customer is frustrated about. This becomes the ad copy, the hooks and the creative angles you hand a client in Mission 06.
Owner pain
What the business owner is frustrated about. Bad leads, tyre kickers, quote chasers, no shows. This becomes your pitch on the call in Mission 08.
Collect both. Keep them in separate columns. Confusing the two is how people end up pitching a roofer about roof leaks.
02 Where the honest answers live
1
Reddit threads
The best source, by a distance, because nobody is performing. Search the market plus the complaint: the niche and "quote", "ripped off", "regret", "how much did you pay". Read the whole thread, not the top comment.
2
One and two star reviews of local competitors
Google and Facebook. The angry ones tell you exactly what the market fails at, which is what your client can promise against. The five star ones tell you what people actually value, which is what your creative should show.
3
Facebook groups and local community pages
"Anyone know a good ..." posts are pure demand. The replies are the objections you will have to beat.
4
The Meta Ad Library
Search competitors in the market. Anything running for months is working. Note the angle, not the design. You are looking for what they are saying, and just as usefully, what nobody is saying.
5
The owners themselves
Once you start calling in Mission 07, every call is research. The complaint you hear three times is a real pattern. Write it down the same day.
03 The method
Ten threads minimum before you conclude anything. Fewer than that and you are reading one person's bad day.
Collect ten to fifteen threads or review sets. Paste them somewhere you can keep them, because sources disappear.
Pull verbatim lines. Their words, exactly. Do not tidy the grammar. The phrasing is the asset.
Group the lines into themes. You will usually find four to six.
Rank the themes by how often they came up, strongest first.
Note who said it: buyer or owner.
What a finished pain bank looks like
Research for a portable building company produced five buyer themes, ranked by how often they showed up across the threads: getting conned by a company that turns up, sells and disappears. The cheap one already fell apart. The quote runaround, where you fill in a form and your details get sold and you still never get a real price. Winter, and not wanting to dig anything out again. Hidden costs and specs the buyer did not understand until the invoice.
Five themes, no invention, each one traceable to a real thread. Every ad angle for that account came out of that list.
04 The capture format
Paste this into your research sheet as the header row. One line per quote.
SOURCE | DATE | WHO (buyer / owner) | VERBATIM QUOTE | THEME | STRENGTH (1-5)
And this is the prompt to run once the sheet is full. Paste your quotes underneath it.
Below are real quotes I collected from forums and reviews about [MARKET].
Group them into themes. For each theme give me:
1. The theme in plain language, one line
2. How many quotes support it
3. The two strongest verbatim quotes, unedited
4. Whether it is buyer pain or owner pain
Do not invent a theme that is not supported by at least three quotes.
Do not rewrite the quotes. Rank the themes by how often they appear.
QUOTES:
[paste]
05 What to leave out, and why
Research turns up angles you are not allowed to use, or not yet. Mark them as excluded with the reason, so you do not rediscover them in three weeks and put them live by mistake.
Excluded until proven
Anything about finance or monthly payments. Meta treats credit offers as a Special Ad Category, which restricts your targeting and puts the ad account at risk if you get it wrong. Never run it because a competitor does.
Rules, permits and legal thresholds. These change by county and state. If the client has not confirmed it in writing, it does not go in an ad.
Review counts, years in business, customer numbers. If the client did not supply the number, the number does not exist. Never invent one.
The standard
Every claim in an ad traces back to something the client can show you. That is not being careful for its own sake. It is what keeps the ad account alive and the client out of trouble.
06 Your task
Lesson 20 / 6 complete
Lesson 3 of 3
Scoring it: the market scorecard
You have a shortlist and you have real research. Now pick one, on paper, with a score, so the decision holds on a Tuesday when a stranger in a different market says yes. This is the last thing you do before you write an offer.
01 Seven criteria, one to five
Score every shortlisted market. Be honest. A score you talked yourself into is worth nothing later.
Criteria
1 looks like
5 looks like
Ticket size
Small jobs, needs constant volume
One job pays for months of your fee
Meta fit
Emergency, searched for on Google
Delayed, visual, decided over weeks
Business volume
You would run out of prospects
Thousands, nationally
Spend capacity
Cannot fund ads on top of a fee
Already spending somewhere
Provable result
Cannot describe the win with a number
One clean sentence with a number in it
Research depth
Found almost nothing in Lesson 2
Five ranked themes, dozens of quotes
Your access
You know nobody, nothing, no angle in
You have worked near it, or you know people in it
02 The kill rules
Two of the seven are gates, not scores. They override the total.
Automatic out
Meta fit scores 2 or below. You cannot fix channel mismatch with creative. You will spend a client's money proving that and lose the account.
Spend capacity scores 2 or below. If the market cannot fund ads, your results are capped before you start, and the client will blame you for it.
Everything else is a trade. Low ticket with huge volume is a real business, it is just a harder one. Weak research depth is fixable, you just have not done the work yet.
03 Break the tie
Two markets within a point or two of each other is normal. Do not agonise. Use these, in order.
Which one can you get proof in fastest? The first case study is worth more than the perfect market.
Which one do you have any access to? A cousin who roofs beats a spreadsheet every time.
Which one can you stand talking about every day for six months? You are going to say the same sentences hundreds of times.
Commit for thirty days
Pick one and do not reopen the decision for thirty days, whatever happens. Market hopping feels like progress and produces nothing. The only thing that changes the decision inside that window is the kill rules turning out to be true.
04 Write it down
Fill this in and keep it at the top of your Target List. Everything in Mission 02 is built on these five lines.
MARKET:
Score: / 35 Decided on:
The buyer:
The owner's number one complaint:
The result I will promise:
Why Meta works for this market:
The reason I would walk away from this market:
05 Your task
Lesson 30 / 5 complete
Mission 01 complete
You have a market, a ranked pain bank in real words, and a scorecard you can defend. Mission 02 turns that into an offer with a scope you can deliver and a price you can say out loud without flinching.
Working assets
Target List and research sheet
Two tabs in one spreadsheet. Build it once and it carries through the whole programme. The Target List becomes your prospect list in Mission 07. The research sheet becomes your ad angles in Mission 06.
01 Tab one: Target List
One row per candidate market while you are deciding. Once you have picked, one row per business you intend to contact.
MARKET | TICKET (1-5) | META FIT (1-5) | VOLUME (1-5) | SPEND CAPACITY (1-5) | PROVABLE RESULT (1-5) | RESEARCH DEPTH (1-5) | ACCESS (1-5) | TOTAL | KILLED? | NOTES
02 Tab two: research sheet
One row per quote. Never summarise into this sheet. Raw words only.
SOURCE | DATE | WHO (buyer / owner) | VERBATIM QUOTE | THEME | STRENGTH (1-5)
03 Tab three: excluded
Short, and it will save you. Anything you found that cannot be used, with the reason, so nobody puts it in an ad six weeks from now.
ANGLE | WHY EXCLUDED | WHAT WOULD MAKE IT USABLE
Field note
Keep the raw threads themselves, pasted in, not just the links. Forum posts get deleted and review pages change. The archive is the thing you own.
The Brand Command System · Mission 01 Recon · Recon. Deploy. Payload.
