The worst mistake you make is the one that looks right.
This mistake is common in computer modelling and other programs.
The inputs all look OK, and the output seems to be in the proper range, but the results are wrong.
Just because something appears to be correct doesn't mean it is.
That is the problem with computer models.
The computer does all the calculations so much faster than a human can that we tend to trust the output.
One famous case that proves the point is the case of the crashed lander.
In September 1999, the Mars Climate Observer crashed because one team wrote the software for controlling the thrusters using Imperial units while another team took that data and assumed it was Metric units.
Here's the kicker though, the computer modelling that tested the software didn't pick up the error and declared the software good to go because the outputs looked right.
Fortunately, no people were on board, and it only totalled a $140m program.
Another computer modelling problem is when too many variables are ranges and unclear values.
This comes back to solving equations such as you might have done in school.
When you have two variables in an equation, you need to know the value of one of those variables to be able to solve it.
If you don't have one of the variables, you need two different equations to solve for either variable.
When one of the variables is a range of values, your answer is also a range of values.
The more variables you have, the more fixed single values you need to solve the equation, and the results are only as accurate as the least accurate input.
I no longer trust computer models to predict anything.
Remember that just because a computer gives a result doesn't mean the result is correct.
Regards,
Brent.
P.S. How does that relate to marketing?
It's simple, whenever a keyword tool says a particular keyword has no searches, it's wrong.
If there were no searches, the keyword wouldn't be listed.
There is a simple, possibly costly way to test the search volume for a keyword.
Set up a campaign in Bing for the keywords in question.
Make sure the ad goes to a place where you can recoup some of the cost.
Run your campaign with the highest click cost in their suggested range and watch the campaign like a hawk.
You don't want clicks.
You want ad exposure.
The number of times your ad is shown will tell you how much search volume there is.
Pull the ad as soon as you have sufficient data.
When you're looking for a quick product for a lead magnet or low-cost front-end offer, this is a great place to start https://go.wm-tips.com/rrweekly.
~ end ~