Atiba de Souza.Start a conversation
WritingMarketing

You're Explaining What You Do Before Anyone's Decided to Trust You

By Atiba de Souza

Atiba de Souza wearing a dark-colored shirt and a white baseball cap with a blue Superman logo on the front. The cover reads: MARKETING You're Explaining What You Do Before Anyone's Decided to Trust You

The team that built the wrong thing, fast

July 2026. I own marketing for TrueMD, a three-founder venture. My team is building the LinkedIn presence, the assets, the whole front door.

Work starts getting dropped. Off-target pieces start showing up. I trace it back and find the problem: most of my team never got told what TrueMD actually is. A subset knew. I'd assumed everyone did.

Nobody on that team was incompetent. They were doing exactly what you do when nobody's given you the why: they filled the gap with a guess, and then they built the what on top of the guess. Tasks got dropped. Assets got made that made no sense the second I looked at them. Not because the team was bad. Because I skipped a step and let them start at the end.

That's an internal story. It's also the exact mistake most content calendars make with strangers on the internet, and it's the exact mistake you're about to make with your own team if you don't fix it before Monday.

The sequence you're violating

WHY / HOW / WHAT: WHY is the pain point and the values behind it — why this matters, why it should be solved. HOW is one piece of your actual process shown in public — proof you can do the thing. WHAT is your product, your offer, the ask itself.

Here's the sequence, and the stage of trust each one builds:

  • WHY builds Know. A stranger decides you understand something true about their life.
  • HOW builds Like. Someone who already suspects you care about their problem gets to watch you actually solve a piece of it.
  • WHAT builds Trust... no. WHAT spends Trust. It only works on someone who already has a balance of it.

Most people build their content calendar backwards. They lead with WHAT because WHAT is tangible, WHAT is what you're proud of, WHAT is the thing you can point at and say "look what I built." Nobody wakes up excited to explain their pain point to a stranger. Explaining your product feels like progress. It isn't, not yet, not to someone who hasn't decided you're worth listening to.

Most of what you should be publishing is not about your product at all. It's about a pain point someone hasn't fully named for themselves yet.

The volume split is deliberate, and it's not a suggestion:

70% WHY · 15% HOW · 15% WHAT Not roughly even. Not "a good mix." Seven out of every ten pieces you publish should never mention your product.

That 70% names one pain point, tied to one of three things people actually move for: relationships, status, or money. The ask on a WHY piece is small: raise your hand, answer one question, self-identify. You are not selling here. You are being recognized.

Trust: the only currency that lets a WHAT-piece land. You don't request it, you spend down a balance built by WHY and HOW, and you can't skip the deposit and still make the withdrawal.

Where the WHY actually comes from

Here's the part that stings. Most people think they know their audience's pain point. They're often wrong, and they only find out by asking.

A business coach the agency worked with started with no customers. Her personas were hypotheses. She built two years of educational content around the problems she assumed her audience had. Then, with real customers across three tiers, she finally ran fifteen-minute interviews with her best ones.

They told her they'd come to her for a problem she'd never thought was her subject. They'd only bought once she happened to talk about it. She went back and checked her own video analytics. The videos on that "off-topic" problem outperformed the ones she believed were her core content, by a margin she hadn't noticed because she wasn't looking for it.

Two years of "correct" content, built on a guess that was never checked, aimed at a pain point that was never the actual reason anyone hired her. Not wrong content, exactly. Content answering a question nobody was asking.

Notice the sequence: interviews first, calendar second. If you have zero customers yet, you have no one to interview, so build on your best hypothesis and treat it as temporary. But the moment you have three real customers, that hypothesis has an expiration date. Most people never go back and check, because checking means finding out you were wrong.

The HOW bucket: show one computation, not the job

Once someone believes you understand their pain (WHY), they start liking you enough to wonder if you can actually solve it. This is where HOW lives, and it's the bucket most people get backwards too. They try to demonstrate the whole job, every step, the full process, in the mistaken belief that thoroughness equals authority.

It doesn't. It equals noise. The move is smaller: show one computation inside the job that the audience didn't know was there. One decision, one method, one thing you do that they've never seen anyone show. That's it. HOW isn't the demo, it's the proof of one specific capability, placed after WHY has already done the work of making them care.

Cutting HOW down this small will feel wrong. It'll feel like you're hiding your work, like you built something with forty moving parts and you're showing three of them and calling it honest. That instinct to over-show is the same instinct that makes people over-explain their product before anyone's asked. Resist it here too. Authority is built by showing a small piece of the work well, not by showing all of it. This is only 15% of your output for a reason: it's not where trust is built, it's where the liking already built by WHY gets confirmed into something closer to trust.

Inconsistent vs. consistent

Leading with WHAT (most calendars)Sequenced Why → How → What
Opens with the product or the pitchOpens with one pain point, no mention of the offer
Assumes the audience's problem, never checkedInterviews best customers, updates the WHY based on their words
HOW content tries to show the entire processHOW content shows one computation, one proof point
Volume split is even or WHAT-heavy70% WHY / 15% HOW / 15% WHAT
CTA asks for the sale immediatelyCTA on WHY content is a raised hand, nothing more
Team briefed on tasks, not on what the venture actually isWhole team reads and confirms the "what this is" doc before producing anything

The diagram

What this actually costs you

Here's the resistance, named plainly, because pretending it isn't there is how people quietly revert after two weeks.

Talking about pain points instead of your product feels like you're not selling. It feels indulgent, soft, like you're stalling. Every instinct you have as someone who built something says: show them the thing, tell them what it does, prove the credentials. That instinct is the one that produced two years of the wrong content for the business coach above, and it's the one that produced dropped tasks and off-target assets on my own team. It's not a stupid instinct. It's just aimed at the wrong step in the sequence.

The real cost is the interview. Sitting across from your best customer and asking why they actually bought risks an answer you don't want: that the thing you're proud of isn't the thing that moved them. That's a small ego injury for a large correction. Most people avoid the conversation entirely and keep producing content into the void, confident, wrong, and unaware of both.

What to do Monday

Rank matters here, so don't treat these as equal, and don't expect this to feel clean.

  1. Most important. Pull your last ten published pieces. Sort them into WHY, HOW, WHAT. If WHAT is anywhere near half, you've found the actual problem, and it's not your product, it's your sequence.
  2. Second, and this is the one you'll want to skip. If you have real customers, interview three of your best ones this week. Ask what problem actually brought them to you. This step will produce an answer you don't want, most likely that the thing you've been publishing about isn't the reason anyone bought. Do it anyway. The coach's two wasted years exist because someone else skipped this exact conversation.
  3. Third. Before anyone on your team produces another piece of content for this venture, write down in plain language what it actually is and what it's for. Confirm every single person working on it has read it, not just the people you happen to talk to most. I found out what happens when you assume that part is already handled.
  4. Minor, do this last. Look at your HOW content and check whether it's trying to show the whole job. Cut it down to one computation. It will feel like you're holding back. Publish it anyway.

Skip step one and none of the rest matters. Skip step two because the answer scares you, and you'll keep polishing content nobody needed. Skip step three and you'll be cleaning up dropped work in a month, the way I was.

And if you have no customers yet: build your WHY on your best hypothesis, ship it, and put a date on your calendar to run the interviews the moment you have three real buyers. The coach's mistake wasn't guessing. It was guessing for two years without ever checking.

Questions

What people ask next

Once I've spent someone's trust by showing them the product, does that balance reset to zero, or is there a way to keep some of it for the next ask?
It isn't a one-time reset — think of it as a running balance that WHY and HOW keep replenishing at a 70/15 clip while WHAT only ever draws down. If you go straight into another WHAT ask without depositing more WHY or HOW first, you're withdrawing from an account you never refilled, and that's when the pitch stops landing.
If I don't have any real customers yet to interview, how do I avoid building two years of content on a guess like the coach did, since there's nobody to check the hypothesis against?
You can't avoid guessing at zero customers — that's the honest starting point, so build on your best hypothesis and treat it as explicitly temporary, not settled. The discipline isn't in getting it right on day one, it's in going back the moment you have three real customers and checking the guess against what they actually say, instead of letting two years pass without ever looking.
How do I actually know when a stranger has moved from Know to Like to Trust, or am I just supposed to trust the 70/15/15 ratio and hope the timing works out for each person?
You don't get a clean signal per person, which is exactly why the ratio exists — it's a structural bet, not a tracking system. The closest thing to a readout is the small ask itself: a raised hand or a self-identification on WHY content tells you someone's at Know, and only after that pattern shows up repeatedly should HOW and then WHAT follow.
Can WHY and WHAT ever share the same piece of content, or does even mentioning the offer inside a pain-point post undo the sequencing?
The sequence works because WHY content asks for nothing more than recognition — a raised hand, a self-identification — and the moment you slide the offer into that same piece, you've turned it into a WHAT piece wearing a WHY costume. Keep them separate; the whole point of the 70% is that seven out of ten pieces never mention the product at all.
Once the interviews show my real value is something different from what I've been publishing about, do I scrap the old content entirely or just fold it in as a smaller piece going forward?
The article doesn't say to throw the old content out — the coach's move was to go back, check the analytics, and notice the off-topic content was already outperforming what she believed was her core material. The correction is in what you publish next: update the WHY you're building the calendar around, and let the real pain point take the 70% slot going forward.
How do you get an entire team to actually absorb a 'why' document instead of skimming it and quietly filling gaps with guesses the same way mine did?
Don't just distribute the document — have the whole team read it and confirm they understand what the venture actually is before anyone produces a single asset. The failure wasn't that the document didn't exist, it's that confirmation was assumed instead of checked, and a guess got built on top of a gap nobody flagged.