Skip to content

Take a look at what’s working

What’s the best approach for getting the best results in something? Is it to work as hard as we can, and be the best we can? Or is it to find a way to beat the system?

I’ve always believed the best approach is to go straight down the middle: discover exactly what’s needed, and focus on doing just that, to the best of my ability. I learned this long ago from a maths student at university, who explained that there were 10 topics on his course, and the best exam result he could probably get in each, if he worked hard, would be 50%. But there were only going to be 5 questions in the exam, and he thought he could guarantee predicting what 3 of them would be. If he revised just those 3 topics, instead of all 10, he predicted he could score 100% in each. Even if he scored 0% in the other 2, it gave him an overall score of 60%, which he reckoned made it the logical approach.

He was right.

When my son was in the sixth-form, he had to do a year-long project called an EPQ. He discovered that the ‘marking scheme’ was published online, and that it consisted of a large number of items for the assessor to evaluate, each typically only worth 2 or 3 marks, but adding up to 100. When the assessor can only score an element 0, 1 or 2, it turns out to be pretty easy to score 2. We gleefully went through every marked element together, ensuring that the work he presented couldn’t be given less than 2 out of 2 for each one. His final score? 100%. It wasn’t ‘cheating’, it was just allocating resources efficiently to what truly needed to be done.

Reverse engineer the existing results

This all came back to me when I saw the same advice being given, over and over again, in the new year’s bunch of “how to do SEO” articles from experts. The advice? Take a look at what’s working well now, and do it better. Don’t try to cover every base; there are hundreds of factors that Google is supposed to take into account, but some have fallen by the wayside, some were never a thing, and even more turn out to be rather insignificant.

If we want to rank for a specific term, the starting point is trying to reverse engineer the existing results, seeing how they score on the most important factors. If the common thread is that they’re all owned by much bigger businesses, or that they’ve all been around much longer than us, maybe that term shouldn’t be a priority for our efforts after all.