"Random acts of marketing" is a phrase people use to describe a business doing a bit of everything with no throughline. A blog nobody reads. A rebrand two years ago. Some ads running that nobody has looked at in months. A social account posting into silence.
It sounds like an accusation of laziness. It almost never is. In my experience the businesses in this situation are the ones trying hardest — they said yes to every reasonable marketing suggestion they were given. That is exactly how they got here.
Every layer inherits from the one above it
Marketing is a stack, and it runs in one direction.
Ad copy inherits from the website. The website inherits from the message. The message inherits from what the business has actually decided about itself: who it is for, what it is genuinely better at, what it is deliberately not.
When that stack is built top-down, every decision below has something to be right about. A writer asking "should this ad emphasise speed or reliability?" has an answer, because the message already settled it. The decision takes ten minutes and it is consistent with fifty other decisions made by other people on other days.
When the stack is built bottom-up — which is what happens when a business starts with tactics because tactics are what agencies sell first — each layer has to invent the thing above it in order to proceed.
The ad writer needs a positioning angle, does not have one, and makes a sensible guess. The web designer needs a message, does not have one, and makes a different sensible guess. The person writing the email sequence makes a third. None of them did anything wrong. Each guess was defensible given what they had. And the business now has three positioning statements it never approved, running in three channels, contradicting each other in ways nobody will ever notice directly.
The failure is invisible at the level where it happens
This is what makes the problem so durable.
Look at any individual piece and it holds up. The ad is well-written. The site looks current. The email sequence follows best practice. Bring in a specialist to review any one of them and they will find small improvements — a better subject line, a stronger call to action — and those improvements will produce small, real, temporary gains.
What no channel review catches is that the pieces are not reinforcing each other. A prospect who sees the ad, then the site, then the email, encounters a slightly different business three times. They will not consciously notice the inconsistency. They will just fail to develop the feeling of recognition that makes someone finally call you.
Marketing that compounds does so because each impression deposits into the same account. Five channels making five slightly different arguments do not compound. They average.
Why the diagnosis usually lands on the wrong thing
Because the symptom shows up in a channel, the channel gets blamed.
The ads underperform, so ads "don't work for our industry." The blog does nothing, so content "isn't worth it for a business like ours." Each conclusion is drawn from real evidence, and each one is wrong in the same way: it attributes a problem to the last place it was visible rather than the place it was created.
The expensive part is that skipping the upper layers does not announce itself when you skip them. You discover it three quarters later, as underperformance nobody can quite attribute, in a channel that gets blamed for something decided long before it ran.
What to do about it, in order
The instinct is to fix the visible thing. Resist it — not forever, just long enough to answer one question.
Ask: if all of this worked perfectly, what specific thing would more people understand about us?
If your team can answer that in a sentence and roughly agree on the sentence, your stack is fine and your problem genuinely is in the channel. Go fix the channel with a clear conscience.
If you get four different answers, or a sentence so general that a competitor could say it truthfully, you have found the actual work. And it is upstream of every tactic currently running, which means fixing it improves all of them at once rather than one at a time.
That is the argument for order. Not tidiness. Leverage. A decision made at the top of the stack is inherited by everything below it, which means it is the only kind of decision that improves work you have not commissioned yet.
Doing this in sequence — what's true, then what a stranger hears, then where they meet it — is the whole of the Loqui Framework, and the reason we won't scale tactics before the message is settled.
