Source & Capture
Where content comes from when everyone on your team is already busy — and how one capture day produces a month instead of a post.
"We never know what to post."
- What it actually is
- Nothing is feeding you raw material, so ideas get invented on deadline.
- The fix
- Harvest from work already happening — Part 2.
"We shot all day and got one video."
- What it actually is
- Plenty of raw material, too little finished work out of each session.
- The fix
- Plan the day around output, not footage — Part 3.
Read the full framework
Drop your email and it opens right here. You'll also get the remaining 6 sections as they're built.
Most lean teams don't have a content problem. They have a source problem and a capture problem, and the two get mistaken for each other constantly.
They're different failures with different fixes. Source is about where the raw material comes from. Capture is about how much finished work one session produces. Fix the wrong one and nothing changes.
These are sections 2 and 3 of the Blueprint. Section 1 is the Audit — what you already have. If you skipped it, you can still run these two; the Audit just tells you which one is bleeding worse.
Part 2
Source
Content does not come from ideation. It comes from things your company is already doing and already saying.
The standard failure is treating content as a creative act performed on a deadline. Somebody opens a doc on Monday and tries to think of something. That's the most expensive way to get an idea and it produces the weakest ones, because the only inputs available at 9am on a Monday are whatever the person happened to remember.
The fix isn't more creativity. It's changing where ideas come from. A lean team can't afford to invent — it has to harvest.
You should never sit down to think of content. You should sit down to choose from content that was collected during work you were already doing.
The four seams
Every business has four places raw material is already being generated. Most companies mine zero of them and then wonder why the doc is blank.
Every question a prospect asks on a call is a piece of content that has already been validated by a person with money. The objection they raise is a hook. The thing you had to explain twice is a video.
What actually happens while you do the work. The process, the decisions, the thing that went wrong, the fix. This is the richest seam and the most ignored one, because when you're inside the work it doesn't feel like content — it feels like Tuesday.
The opinions the owner or the senior person already holds and repeats. The things they say in every meeting. Nobody has written them down because everyone internally has heard them a hundred times — which is exactly the signal that it's a real position and not a manufactured take.
The work itself, and what it produced. Before and after, the number that moved, the thing you built. This is the seam that converts, and it's usually the only one companies use — which is why their content reads like a portfolio nobody asked to see.
The Source Ledger
One list. Not a content calendar — a ledger of raw material, captured the day it happens, before it becomes a deadline.
| RawTheir words, not a tidied-up topic | SeamWhich of the four it came from | HeardTimes it has come up | EvidenceWhat proves it | StatusRaw → Chosen → Captured → Shipped |
|---|---|---|---|---|
| "How do I know it's actually working?" | Sales | 11 | 90-day dashboard, Q2 client | Shipped |
| "Can you just do a version for Instagram?" | Delivery | 8 | Reformat log, 14 assets | Chosen |
| "We don't want it to look like an ad." | Sales | 6 | — | Chosen |
| "Nobody outside has seen the Bauen build." | Proof | 4 | Full shoot, unreleased | Raw |
| "Book it like a real job or it slides." | Operator | 3 | Founder, every kickoff | Raw |
Five rows in, the pattern is already visible: Sales is loudest, Proof is untouched.
Frequency is the demand signal you already own — if one seam is 80% of the ledger, you're mining one wall of the mine. And a row with no evidence is an opinion piece, which is fine, as long as you know that going in.
If your next 8 pieces are not already written down somewhere, you do not have a source. You have a deadline, and something is going to get made in a hurry on Thursday.
Choosing from the ledger
Once the ledger is full the job stops being "what should we make" and becomes a selection problem, which is a much easier problem. Three filters, in order:
- Heard the mostFrequency first. The question 6 prospects asked beats the clever idea one person had.
- Has evidence attachedAnything you can prove outranks anything you can only assert. Proof is what makes a small brand credible before it has an audience.
- Only you can say itIf a competitor could publish the identical piece, it doesn't build anything for you. Cut it or find the angle only your business has.
What survives all three goes to Capture.
Part 3
Capture
A shoot is not a day that makes a video. It's a system that should make a campaign.
Here's the math that keeps lean teams from ever producing anything: a capture day costs a full day of everyone's time and returns one asset. At that exchange rate you can never justify the day — so it keeps getting pushed, and it should, because it's a bad trade.
The trade only changes when the day returns 6 or 7 pieces instead of 1. And that doesn't happen in the edit. It happens because you decided the outputs before you decided the shoot.
Write the asset list first. Build the capture day backward from it. Never forward from "let's get some footage."
Where the day is actually won
Almost everyone assumes the shoot is the hard part. It's the third hardest.
The day is won or lost in the hour before it, not during it.
The Capture Plan
One page. Written before anyone books anything. In this order — the order is the whole point.
- The asset listName every finished piece this day will produce, before you know how you'll shoot any of it. Main piece, behind the scenes, the team, the lessons, teasers, stills for the site, material for sales calls. If the list has one item on it, stop and rebuild the day.
- One message per assetEach piece says one thing. Written in a sentence. If two assets have the same message, you have one asset and a duplicate.
- Shot list, built from the asset listEvery shot exists because a named asset needs it. This is the single highest-return document in production — with a real shot list a team can clear 27 shots in 90 minutes. Without one, the same team gets a fraction of that and finds out in the edit.
- Coverage rulesThe four things you always roll on regardless of the shot list: the setup, the mistake, the debrief, and the work itself. These are what turn one shoot into a story instead of a folder of clips.
- People3 is the floor: one in front of the camera, one on the camera, one picking up everything else. A producer or PM sits on top of that. Below 3, the third job doesn't disappear — it just gets done badly by whoever is least busy.
- LogisticsLocations, wardrobe, props, gear, permits, model releases, call sheet with a timeline on it. Boring, and the only reason the day runs.
- Expected output, written downWhat you expect to come out, committed to paper before you go. This is the step almost everybody skips, and it's the only one that makes the day measurable afterward.
The setup — people arriving, gear going up, the plan being said out loud. Free, and it's the opening of the behind-the-scenes piece.
The mistake — the take that burns, the thing that breaks. Roll through it. It is the most watchable footage you will get all day and you cannot recreate it.
The debrief — 60 seconds at the end where people say what they'd do differently, tired and unguarded. Consistently better than anything the same people would say to camera on purpose.
The work — the actual thing being made, uninterrupted, from a second angle if you have one. This is your proof seam feeding itself.
The Capture Multiple
One number tells you whether you have a supply chain or a video habit.
Finished assets shipped ÷ capture days spent.
Track it per capture day, not per month. Averaging hides the day that produced nothing.
The part nobody warns you about
Work you make for your own company is always the first thing to slide, because a client is waiting on everything else and only you are waiting on this. There is no process fix for that. There is only one thing that works:
Put it on the calendar like a real job and treat it like one. Same call sheet, same start time, same standard. The teams that produce their own content aren't the ones who found spare time. They're the ones who booked it as work.
Run it
The one-page version
Print this. Everything above collapses into 12 lines.
Source — this week
- Pull the sales seam. Last 10 calls, every question written down verbatim.
- Pull the delivery seam. What did we solve this month that we'd solved before?
- Pull the operator seam. What does the founder repeat in every meeting?
- Pull the proof seam. What shipped in 90 days that nobody outside saw?
- Open the Source Ledger. Raw · Seam · Heard · Evidence · Status.
- Choose 8. Heard most → has evidence → only you can say it.
Capture — before the day
- Write the asset list first. Every finished piece, named. More than 1.
- One message per asset. One sentence each.
- Build the shot list from the asset list. Every shot serves a named asset.
- Lock coverage rules. Setup · mistake · debrief · the work.
- Staff it to 3 minimum. On camera, on the camera, on everything else.
- Write the expected output down before you leave, and check it after.
Score yourself — 5 questions
- Can you name your next 8 pieces right now?No → source problem.
- Do more than 2 of the 4 seams appear in your last 20 pieces?No → you're mining one wall.
- Does a shot list exist before your capture days?No → this is the cheapest fix on the list.
- What was your Capture Multiple on the last shoot?1 → capture problem, not a content problem.
- Is your own company's content on the calendar as booked work?No → it will keep sliding, and no system fixes that.
Want someone to run this against your actual business?
Book a teardown. We look at what you're publishing now, find where the content is losing momentum, and map what a capture day would need to produce for it to be worth booking. About 30 minutes, no commitment, and you keep the plan either way.
Book a teardown →Building the marketing team of the future. Execution gets cheaper. Judgment gets more valuable.