Keyword Stuffing - Does It Still Work?

Hey,

The heat finally broke here in the UK yesterday, so it's been raining since yesterday afternoon - a welcome break for the Brits!

It's also a very soothing experience listening to rain as you fall asleep :)

Weather is a hot topic for the British, but a hotter topic for Amazon Ads (and a continuation of this Amazon Ads mistakes mini-series) is keyword stuffing...

A few years back, stuffing your campaigns with the full quota of 1,000 keywords was a pretty robust strategy. And the keywords you targeted didn't even need to be that relevant.

One keyword I saw a lot of was simply book book book! And this keyword (book book book) used to generate huge numbers of sales.

In 2019, however, Amazon changed their algorithm to put a much larger emphasis on relevance.

This meant that the keyword stuffing, spaghetti on the wall approach was no longer a robust (or effective) strategy as it once was. It also required more upfront work to find relevant keywords.

You may have witnessed this yourself; if you look at any of your own campaigns that are targeting hundreds of keywords, you'll see that maybe 10%-20% of those keywords will actually have any statistically significant data on them. And maybe 20% of those will have actually generated any sales/page reads.

The remaining keywords may have a few impressions and a handful of clicks, but nothing you can take action on.

Amazon's Algorithm will soon figure out which Keywords it's going to focus your budget on within a campaign based on conversion data and essentially ignore the rest.

Amazon is a business at the end of the day and it wants to make sales. It also wants to provide its customers with an enjoyable (and relevant) shopping experience, by showing them books they are likely to buy/borrow based on what they're searching for and their past shopping behaviour.

If there are keywords you're targeting that just aren't going to convert into sales, Amazon isn't inclined to show your Ads for those keywords - there are plenty of other advertisers it can choose from who have more of a proven track record for making sales from these keywords.

On a more practical level, if you have 500 Keywords in a Campaign and a daily budget of $10, in a perfect world, collecting data on every single one of those keywords would take months.

Seeing as we're not in a perfect world, you may get clicks on 50-100 of those 500 Keywords.

Ideally, you need at least 10 clicks on a Keyword before deciding whether it has the potential to be a winner or not.

Let's say an average Cost Per Click (CPC) is $0.50, you would need to spend $5 on every keyword to figure out if it's going to be a good performer for you.

So, for those 500 Keywords, you would need to spend $2,500 ($5 x 500 Keywords). That's a lot of money to figure out what is going to work! And at $10 per day, that's 250 days - almost a year!

As I say though, we're not in a perfect world and this scenario would never actually play out!

Instead, what I recommend you do is test 10-15 keywords in a single campaign. And make sure every single one of those keywords is highly relevant to your advertised book.

With this approach, you'll save a lot of time, a lot of money and figure out what works within a matter of days, or at most, weeks, rather than months.

To Your Success
- Matt