Your ad is not the problem. The research you skipped is.
If your ads are not working, the problem is usually not the ad. It is the 5 things nobody wrote down before it was made.If your ads are not working, the problem is usually the 5 things nobody wrote down first.
- All 5 things are research. Not 1 of them asks you to be clever.
- The pain and the want are 2 different lists. Most ads only ever reach for the pain.
- Write both lists in your buyer's own words, never in your category words.
- 1 focused day gives you about 20 lines and 2 decisions, and that page feeds every ad for 3 months.
- People skip it because nothing ships at the end of it. That is the only reason.
In this article
I know how this sounds. Do your market research before you write ads. Nothing new. I keep saying it because it is the real gap. Everyone nods at it, and then goes straight to writing hooks anyway.
If your ads are not converting, this is almost always the part that got skipped. Not the creative. Not the budget. Not the targeting. The 5 things nobody wrote down before the ad was made.
There are 5 of them, and all 5 are research, not creative choices. You can finish all 5 in 1 focused day. Every ad you write for the next 3 months gets faster and better before it even goes live.
01All 5 are research, not a brainstorm
Here is the whole list. Read it once and notice that not 1 of them asks you to invent anything. Every answer already exists somewhere, in something a real buyer said.
The 5 things, before you write a line
- 01Pain points
The top 3 to 10 pains your buyer actually feels, written the way they would say it.
- 02What they want
The top 3 to 10 things they are actively going after. Not the absence of pain.
- 03What they get
What is true about their life on the Tuesday after they buy from you.
- 04What they already know
Which of the 5 awareness levels they are standing on right now.
- 05How tired the market is
How much of your promise they have already heard from everyone else.
A brainstorm gives you opinions. Research gives you lines you can quote. That is the whole difference, and it shows up in the first line of the ad.
02The pain and the want are 2 different lists
Start with the pains. The top 3 to 10, written the way the person would actually say it out loud. That last part is the whole thing.
A physio does not write "reduced lumbar mobility". The patient says "I cannot pick up my kid without thinking about it first". A consultant does not write "operational inefficiency". The owner says "everything still comes through me and I cannot take a week off". Your category words are precise, and they do not sell, because nobody sees their own life inside them.
Your category words are correct and they still do not sell. Nobody recognises their own life in them.
Then do the wants, and keep them on a separate list. The pain is "my back hurts at 3pm". The want is "I want to play cricket on Sunday without paying for it on Monday". You can speak to 1 and completely miss the other. Most ads only ever name the pain, because pain is easier to name.
The words matter more now than they did in 2024. People type their problem into a chatbot in their own plain language and ask who can fix it. The firms that wrote those exact words down are the ones the answer tends to name. I wrote about that in people now ask a chatbot who to hire.
03The other 3 decide how hard you have to work
Number 3 is benefits and outcomes. What they actually get if they buy. What changes on the Tuesday after. Keep this away from your features and away from your process. Nobody buys the process. They buy what is true about their life afterwards, and the more exactly you can describe that Tuesday, the less convincing you have to do.
Number 4 is the awareness level. There are 5 of them: unaware, problem aware, solution aware, brand aware, most aware. This decides what you are allowed to say. An offer sent to someone who does not even know they have a problem is not a bad offer. It is a good offer arriving 3 steps too early.
Number 5 is how tired your market already is of the promise you are about to make. If everyone in your category has said the same line for 5 years, the plain claim is dead before you spend a rupee and you have to show the how, the reason your way works. If you are genuinely first, the plain claim works and anything clever is wasted effort. Same offer, completely different message, decided only by how much your audience has already heard.
What each input decides
Pain points
The opening line
What they want
What you promise
Benefits
What you prove
Awareness level
How much you are allowed to assume
How tired the market is
How hard you have to work to stand out
04Where you actually get this
Not from a keyword tool. Not from a workshop with sticky notes. From conversations that already happened, most of them already sitting on your phone.
Go and read these
- The first message every prospect ever sent you, before your framing got involved. This is the richest source you have and it is in your inbox.
- Sales call recordings, with consent. Mine the phrasing, not the summary.
- Client group chats. People describe problems casually there in a way they never do on a form.
- Reviews, yours and your competitors'. People explain their situation before they give a verdict.
- Comments that start with "this is so me". That is a pain point announcing itself.
Keep it as a living document. 1 file of exact phrases, added to whenever you hear a good one. Phrases, not summaries. Most of my best performing hooks are lightly edited quotes from people who had no idea they were writing an ad.
In 2026 I do not keep that file on a laptop. It sits in the company brain inside the AiOS command center, so the team writing ads and the team talking to clients are reading the same phrases. When I hear a good line in a meeting I send a voice note to the Telegram assistant and it gets filed. Then STORM, the posts and videos and articles we put out every week, gets written from those same phrases instead of from imagination.
05What 1 pass actually gives you
People treat this as a mindset exercise. It is not. It has an output, and the output has a size you can count.
1 research pass, in lines on a page
Pain points
3 to 10, in their words
What they want
3 to 10, in their words
Benefits and outcomes
3 to 10, concrete
Awareness level
1 decision per group of buyers
How tired the market is
1 decision per market
Take the middle of those ranges and you finish with roughly 20 lines and 2 decisions. That is the whole thing. It fits on 1 page and it feeds every ad you write for the next 3 months.
Now compare that against the other way. Without those 20 lines you write from imagination, and the way you find out you were wrong is by spending money. 1 week of reading against 1 month of budget is not a close call, and yet the reading is the part that gets skipped, because it does not feel like progress.
A week of reading beats a month of guessing with your card on the table.
The count gives you something else too. If you sit down to list 10 pains and you dry up at 3, that is the finding. It means you have been talking about your customers instead of to them, and no amount of creative will fix a page with 3 lines on it.
06The buyer and the user are not always the same person
1 thing sits underneath all 5 and quietly wrecks the research when it is missed. Think about nappies. The user is the baby. The buyer is the parent. The baby does not shop, does not know the brand exists, and cannot be sold to at all. Everything you write is for the parent, even though the baby is the one using it.
It happens more often than people expect. I am not a buyer for a clip on microphone, because I have one and my voice carries anyway. But I could be a user of one, because my team runs workshops and not everyone speaks loud. Different role, different pain, different ad. Get that wrong and all 5 of your research inputs are perfectly correct about the wrong person.
So before you start, write down who pays and who uses. If they are 2 different people, you need the pains and the wants of the payer, and the ad is written to the payer.
The honest timeline is 1 focused day. Half a day reading old messages and call notes, half a day sorting what you found. Nothing ships. There is no ad at the end of it. That is the only reason it gets skipped, and it is a bad reason.
Open your inbox, pull up the first message your last 10 buyers ever sent you, and paste the first 2 lines of each into 1 file. You will have half your pain list before lunch.
07Questions, answered
How long does this really take?
1 focused day. Half a day reading old messages and call notes, half a day sorting what you found. It feels like nothing shipped, and then every ad after it is faster to write.
What if I cannot think of 10 pains?
That is the finding, not a failure. It means you have been talking about your buyers instead of to them. Go and read the first message 10 of them ever sent you and the list fills up.
Are pain points and wants not the same thing?
No. The pain is the back that hurts at 3pm. The want is the Sunday game of cricket. You can speak to 1 of them in your ad and completely miss the other.
Can Ai do this research for me?
It can sort and group what you give it, very fast. It cannot know what your buyers said. Feed it the real messages, reviews and call notes and it is excellent. Feed it nothing and it guesses.
Do I have to redo this for every ad?
No. 1 pass gives you about 20 lines and 2 decisions, and that page feeds every ad for the next 3 months. Just add a phrase whenever you hear a good one.
If you want me to look at your buyers and your last few ads with you, and tell you which of the 5 is missing, book a consulting call and we will go through it together.
Digital PratikA 37 year old, smiling, practical Gujarati, a Kathiyawadi businessman. Founder of JorrDaar LLC, Dubai.
