I wanted to analyze the impact of negating bad search terms, so I conducted a test.
The idea was to flag 'bad traffic'—search terms or targeted products defined as keywords or targets with no sales and costs exceeding the allowance over the flagging period.
We monitored spend on the flagged traffic in the future.
The test involved flagging periods of 2 weeks and 4 weeks. We conducted these over a year and analyzed the weeks following each flagging period.
The results show that AI quickly adjusts by reducing spend on bad traffic. In the mid to long term, the tool achieves a similar ACoS to blacklisting but generates double the sales and spend.
From the example in the graph, this account spent an average of €45,185 monthly.
If we negate bad search terms, then in a 360-day timeline, these would be the stats:
Sales: Almost €600,000
ACoS: Almost 32.47%
However, if we don't negate the bad search terms, then in a 360-day timeline, these would be the stats:
Sales: Almost €910,000
ACoS: Almost 32.52%
See the graph for the actual difference figures.
This indicates that if we don't negate the bad search terms from our PPC campaign, we can generate more sales by just increasing ACoS by 0.17%, which is crucial while optimizing our PPC campaigns.
This is just 1 account's stats. I've tested more accounts—stay tuned for the other account stats. (2/5)
Disclaimer: The attached graphs were generated using randomly selected accounts; they weren't chosen to highlight favorable results.
Our system never negates bad search terms outright. If a keyword did not perform well in the past, the AI will reduce the bid for the keyword until it stops spending but keeps it active on the off chance that the keyword starts to shoot up in searches, becomes more relevant, and may actually get some really cheap sales.
Want to give this a try? Let's talk, and I can show you how our AI system is helping Amazon Sellers scale further! https://hubs.ly/Q02Dm-ht0
#Signalytics #PPCAI #NeverNegate #AmazonPPC