Our method
Our method is simple. We measure first, so we know what is working. We map how things run before we build anything. A person checks what is automated. We test everything against real cases, then hand it over so you can keep it going without us. If we are not the right fit, we say so.
How we think
Six principles
- 01
Measure first
Before a campaign runs or a site launches, we set up the tracking. You cannot improve what you cannot see, and the ad platforms learn from the events you send them.
- 02
Map before you build
We trace how things really run today, including the odd cases nobody wrote down. The map is yours whether or not you build anything. Two of the last engagements we scoped ended with a process change and no build.
- 03
A person checks
Every piece of content passes a human editor, and nothing publishes unread. Every automated workflow has a route to a named person and an off switch.
- 04
Test, then hand over
Everything is checked against real cases before launch. We hand over what we built, so you can check it yourself and run it without us.
- 05
Say it plainly
If something is not working, you hear it from us before you spot it. When a month is flat, the report says it is flat. You will not find invented statistics on this site.
- 06
Keep it reversible
You own the code, the accounts and the content. We would rather you stay because the last release was good than because leaving is hard.
How it runs
Four steps, stop after the second
- 01
A 30-minute call
You tell us what you are building, what you are spending and what has not worked. We tell you what we would do first and whether it is worth doing at all. Two of the last engagements we scoped ended with a process change and no build.
- 02
We map what is there
We trace how things really run today, including the odd cases nobody wrote down. The map is yours whether or not you build anything.
- 03
We build in short steps
You see the work running as we go, and you hear from us before you spot a problem.
- 04
We test and hand over
Everything is checked against real cases before launch. After that every run is logged, problems raise an alert, and anything automated has an off switch.
Just as important
What we will not do
- 01
Automate a broken process
Automating a broken process gets you the same bad result faster. Mapping first is partly there to catch this before you pay for a build.
- 02
Ship unattended work
We do not ship software that acts without a logged decision, a route to a person and an off switch. We do not publish generated content that no person has read.
- 03
Take some kinds of work
We do not take crypto or Web3 work, brand identity design or cold outbound sequences.
- 04
Dress things up
We do not report on impressions when you needed leads, or send a report that makes a flat month look good. We do not publish client logos or testimonials we cannot stand behind.
Ready when you are
Start with a call
Thirty minutes. You tell us what you are building, what you are spending and what has not worked. We tell you what we would do first and whether it is worth doing at all.