01The Direct Answer
Why do different agencies produce the same bad result?
Direct Answer
Two suppliers who receive the same input, and who are competent, will converge on similar work. Similar output from different people is evidence about the input, not about the people.
The mechanism is not mysterious. A marketing supplier starts from what the company tells it about the product, the customer and the difference that matters. If that account is the same both times — and it usually is, because it comes from the same people, in the same words, often from the same slide deck — then both suppliers are solving the same problem statement. They will reach for the same channels, describe the product in the same terms, and attract the same kind of inquiry. The second attempt is not a second experiment. It is the first one, run again by someone else.
02What Changed and What Did Not
Make the two lists side by side
This exercise takes fifteen minutes and it usually ends the argument, because it is very hard to look at both columns and keep believing the supplier was the variable.
Column A
What the switch replaced
The people doing the work. The tools. The reporting format and its frequency. The account manager. The tone of the meetings. Possibly the price. All of these are real changes, and none of them touches what the work is aimed at.
Column B
What stayed exactly as it was
Which customer is worth having and which application matters. In which words that buyer describes the problem. What counts as a good lead. Who inside your company answers technical questions, and how fast. Who signs off on copy. What happens to an inquiry after the form is submitted.
Column B is the specification the work is built against. If nothing in it moved between the two contracts, then the experiment held the specification constant and varied the manufacturer. Getting the same part back is the expected outcome, not the surprising one.
03The Context Reset
Why the second start is slower than the first
The first supplier did not stay still. Over the contract it learned things: that a certain segment never answers, that a term you use internally is not the term buyers search for, that one product line generates inquiries you cannot serve, that a specific competitor is always in the comparison. That learning is real value, and almost none of it was ever written down in a place you own.
What the Reset Costs
Rejected hypotheses come back
The new supplier proposes, in good faith, three things the previous one already tried and abandoned. You approve them, because they sound reasonable and nobody remembers they were tried.
→
The vocabulary is relearned
Every industrial account has terms that must be right — norm, alloy, tolerance, process. The first supplier got them wrong for a while and then got them right. The second starts wrong again.
→
Your people pay the bill
The reonboarding is not billed by the agency — it is billed in the hours of your commercial director and your engineer, who have to explain it all again while the month runs.
→
The comparison becomes unfair
You compare the second supplier at the start of its curve with the first at the end of its own. It looks worse than it is, which sets up the third switch.
None of this argues for keeping a bad supplier. It argues that the switch has a hidden bill, and that the bill is proportional to how little of the previous relationship was documented on your side. [EVIDÊNCIA NECESSÁRIA: em uma conta real com histórico de dois fornecedores, quais hipóteses foram retestadas pelo segundo e quanto tempo levou até ele voltar ao nível de acerto de vocabulário técnico que o primeiro tinha no fim. Sem isso, manter como mecanismo.]
04The Briefing Test
Two documents, one answer
This is the diagnostic that separates an execution problem from a structural one, and it does not require anyone to be honest about their own performance.
The Test
Put the material you handed to the second supplier next to the material you handed to the first. Read both.
If it is the same document — or the same deck, or the same conversation reconstructed from memory — then between the two contracts you changed the executor and did not change the specification. That is not a supplier problem and no third supplier will fix it. If, on the other hand, the second document is visibly better than the first, and the result still repeated, you have ruled out the easy explanation and the diagnosis moves downstream, into what happens after the inquiry arrives.
There is a second version of the test for cases where no briefing was ever written. Ask each supplier — the current one now, the previous one from memory — what hypothesis it was working on. If both answers amount to "bring in more leads", then no diagnosis was ever made by anyone, and two suppliers were paid to execute against a goal rather than against a problem.
05Clue, Not Proof
A repeated symptom points upstream — it does not name the cause
It would be convenient to conclude from two failed contracts that the cause is the missing specification. That is the most common cause, and it is still only a hypothesis until you rule out the others. Repetition tells you the cause is upstream of both suppliers. It does not tell you which upstream thing it is.
Three Rival Explanations Worth Ruling Out
The offer itself
Test: does the same difficulty appear in channels no agency ever touched — referrals, trade fairs, long-standing customers? If the referral also stopped converting, the problem is not in demand generation.
→
The response capacity
Test: pull the time between an inquiry arriving and the first contact attempt, for both contracts. If it is the same and it is slow, both agencies delivered into the same bottleneck.
→
The market moment
Test: did the segment you sell to change its own investment behaviour during the period? A halted client industry produces flat pipelines regardless of who is doing the marketing.
Run these three before concluding. If all three come back clean and the symptom still repeated across two suppliers, then the specification hypothesis is no longer the convenient answer — it is the surviving one.
06The Record That Stops the Next Loss
Four things to write down while you still remember them
Whether or not you switch again, this record is what makes the next start begin above zero. It is short — a few pages — and it is the only asset in this whole story that belongs to the company rather than to whoever is executing.
The Minimum Record
The customer, in one paragraph
Segment, application, size, who inside that company feels the problem and who signs. Written by you, not by a supplier, and approved by whoever runs sales.
→
The words the buyer uses
Taken from real inquiries and real calls, not from your catalogue. The gap between your vocabulary and the buyer vocabulary is where most industrial demand generation quietly fails.
→
What counts as a good lead
A closed list of criteria that both marketing and sales signed. Without it, quality is an opinion and every contract ends in the same argument.
→
What was tried, and what it produced
Kept as you go, one line per attempt, by someone on your side — not requested from the supplier at the end, when it is a negotiation. A month of this is worth more than a handover document written under notice.
This record is the practical core of the Prebound Marketing method: the base is organized before acquisition work starts, it is approved by the company rather than by the supplier, and it survives the change of whoever executes. That last property is the whole point — it is what makes the second contract an actual second attempt instead of a repeat of the first.
07The Next Question
And if the current agency is genuinely good?
Some readers run the briefing test, find that the second document really was better, and still see the same flat line. At that point the complaint changes shape: there is nothing to accuse anyone of, the work is visibly competent, the reports are honest, and revenue has not moved.
That is a different problem, with a different mechanism, and it is the subject of the next page: how correct indicators coexist with flat sales.
Prebound Marketing is cheaper before.
08FAQ
FAQ
Why do different agencies produce the same bad result?
Because two competent suppliers who receive the same input converge on similar work. Both were told the same thing about the product, the customer and the difference that matters — usually by the same people, in the same words, often from the same deck. So both solved the same problem statement, reached for the same channels and attracted the same kind of inquiry. The second contract was not a second experiment; it was the first one run again by someone else.
What can explain a second agency delivering the same problem?
Four, in descending order of frequency. The specification never changed, so the same brief produced the same work. The learning of the first contract was never written down, so the second supplier retested what had already failed. The bottleneck is downstream — inquiries arrive and nobody answers them fast enough, in both contracts. Or the offer itself is struggling, which shows up as the same difficulty in channels no agency touched, such as referrals and trade fairs. Rule them out in that order.
How do I separate an execution problem from a structural one?
Put the briefing you gave the second supplier next to the one you gave the first. If it is the same document, you changed the executor and not the specification, and that is structural — no third supplier fixes it. If the second document was visibly better and the result still repeated, the easy explanation is ruled out and the diagnosis moves downstream, to what happens after the inquiry arrives. Where no briefing exists, ask each supplier what hypothesis it was working on.
Which data can I pull on my own to test those four causes?
For the specification hypothesis: the two briefings, side by side. For the downstream bottleneck: the time between an inquiry arriving and the first contact attempt, pulled for both contract periods. For the offer: whether the same difficulty appears in referrals, trade fairs and existing customers, which no agency touched. For the market: whether the segment you sell to changed its own investment behaviour in the period. Each of the four is available inside your company without asking any supplier for anything.
Would a third agency solve it?
Only if something other than the supplier changes with it. A third contract with the same brief produces a third version of the same work, plus the cost of a third ramp-up: rejected hypotheses proposed again in good faith, technical vocabulary relearned from scratch, and hours from your commercial director and your engineer spent explaining the business for the third time. If you are going to run the third attempt, change the specification first and the supplier second.
Is bringing it in-house a way out of this loop?
It removes the ramp-up cost of future switches, which is a genuine gain, and it keeps the learning inside the company by default. It does not remove the specification problem: an internal team facing an undefined ideal customer and an undefined good lead produces what an agency produces facing the same void, with the difference that the internal salary continues while you figure that out. Write the specification first — it is useful either way, and it is cheap.
The first agency never handed anything over. Can I still recover the learning?
Partly, and it is worth the afternoon. Most of it is reconstructable from inside your own company: the inquiries that arrived and what they asked for, the reasons deals were lost, the terms buyers actually used on calls, and the campaigns that ran, which are visible in the accounts if they are in your name. What is genuinely lost is the reasoning — why something was stopped. Write down what you can before the people who remember it move on.
When does the next question become "the agency is good and sales are still flat"?
When the briefing test comes back clean. If the second document really was better than the first, if the work can be inspected, if flagged errors became corrections, and revenue still has not moved, then there is nobody left to accuse and the diagnosis has to move. The mechanism at that point is different: each supplier is measured by a number it controls, and nobody is measured on the path between a submitted form and a signed order.
Change the specification before changing the supplier
The record that makes the next start begin above zero — the customer, the words, the criterion for a good lead, what was already tried — is what the Prebound Marketing method produces and leaves with your company, not with whoever executes.
