You dropped a safelist from your rotation last month. Before that it was a subject line style you had used for weeks, and before that a step in the funnel that seemed to slow people down. Each one felt weak, so each one went. The numbers did not improve after any of them, and now you could not say what weak meant if someone asked. If you are searching for what to cut when campaigns are not working, the answer starts before anything gets cut.
Why a Cut Without Criteria Leaves Nothing Behind
A feeling is not a measurement, and a campaign always has something that looks weak. Safelist traffic moves in waves. Credit mailer clicks rise and fall with how active the members are that week. Open rates drift for reasons that have nothing to do with your subject line. When there is no line set in advance, the piece that gets cut is whichever one looks worst on the day your patience runs out. That is a decision about timing, not about performance.
A cut made on feel can only end two ways, and both leave you with nothing to use. If the numbers improve afterward, you cannot tell whether the cut caused it or the next week was simply better. If the numbers stay flat, you cannot tell whether the cut was wrong, whether it was right but not enough, or whether the real problem was somewhere you never looked. You have nothing to defend if the result is questioned and nothing to repeat if the result was good.
You cut the traffic source because it felt weak. The numbers did not improve after it was gone. Now you cannot say what weak meant, whether the source was the problem, or what you would do differently next time. The cut happened, and nothing was learned from it.
The cuts also stack. A traffic source goes, then a subject line style, then a funnel step, each one in response to the last disappointment. After three changes the campaign is different from the original in three ways, and there is no way to separate what each change did. Even if the numbers finally move, you are looking at a result with three possible causes and no record of which one mattered.
What to Cut When Campaigns Are Not Working Depends on What You Defined First
Before anything gets cut, three things need to be written down: the one number that piece is supposed to move, the line that number has to cross, and how long the piece gets to cross it.
The number has to belong to the piece being judged. For a traffic source, raw clicks tell you very little, because a source can send plenty of clicks that never do anything. The number is closer to how many of those visitors opt in. For a subject line style, the number is how that style performs against your other emails sent in the same stretch, not against a week from last spring. For a funnel step, the number is how many people who reach that step move on to the next one. If you cannot name the number a piece is responsible for, you are not ready to cut it.
The line has to be a specific figure, and it has to be set before you look at the results. A line drawn after the results are in will almost always land just above wherever the piece happened to finish, because you already know the outcome and it is easy to set the bar where the feeling already was. A line written down in advance cannot be moved to match a mood.
The window has to be long enough that one bad stretch does not decide the outcome. On safelists and credit mailers that usually means counting in visits and sends rather than days, since a slow week can hold half the traffic of a normal one. Set the amount of traffic the piece gets before it is judged, and do not judge it before that amount arrives.
What Changes When the Rule Comes First
With a number, a line, and a window in place, you cut one thing at a time, and when something goes you know exactly why. If the piece that replaces it also misses its line, that tells you something real about where the problem sits. If the replacement clears it, you know what to keep and you can do the same thing again on the next campaign. The decision becomes one you can explain to yourself six months later.
Sometimes the rule tells you not to cut at all. A piece that felt weak turns out to clear its line, and the problem is somewhere else in the campaign. That result is often the most useful one, because it stops a cut that would have changed nothing and points your attention at the part that is actually failing.
This does not make a weak offer convert, and it does not promise the next test will work. It will not tell you which number matters most in every campaign you run. What it does is make sure the next cut produces an answer you can use instead of another feeling you have to live with.
Write the Line Down Before You Cut Anything
Pick the piece of your campaign that feels weakest right now and leave it alone for the moment. Write down the number it is responsible for, the line it has to cross, and the amount of traffic it gets before you judge it. Then let it run. That is what to cut when campaigns are not working: only the piece that missed a line you set before you knew the result.
If you want the rest of what happens between the click and the sale laid out plainly, start at acclaimedjames.com/traffic-system.
