We have been developing Beacon, the digital marketing intelligence dashboard, for a few years now. It began as a tool to help digital marketing agencies (and ourselves – a team of marketers, creatives and digital designers) better insights into the social media and PPC campaigns we were running for our clients. We didn’t intentionally set out to create a tool to tackle ad fraud and bot clicks, but we discovered it was a huge problem that needed solving!

The bots are coming

This soon developed into something a lot bigger, and early on we realised the impact that fake click and bot activity was having, not only on skewing our data, but on our campaign spend. We realised that a good portion of our paid for clicks were not actually generating human website visitors – we only discovered this when we designed and implemented our Bot Click Indicator tool.

Our digital marketing experiment

We recently ran an experiment of our own campaign, as part of the 2019 Leeds Digital Festival, across various marketing campaigns including Facebook, Twitter, Google Ads, LinkedIn and Reddit. And the results were interesting…

“Half the money I spend on advertising is wasted; the trouble is I don’t know which half.”John Wanamaker

As you can see from our chart below, the various platforms to which we assigned PPC spend, reported their Cost Per Click (show in orange), and our Beacon platform reported the real Cost Per Visit (shown in blue). We calculated this by simply removing the bot clicks and fake website visits that Beacon identified from the overall spend, meaning that we actually ended up spending more per click, if our goal was to drive real human visitors to the website… which, of course it was.

PPC spend, reported Cost Per Click vs Cost Per Visit

These pie charts below show which platform was the most effective in driving real human visitors to our website. In the first chart you can see where most of our campaign budget was spent. Looking at the second chart, you can see where our visitors for the campaign came from and you can see that Google Ads delivered most of our website traffic (33.7%), followed surprisingly by Reddit (25%).

Our platform, Beacon, also analyses your website visitor actions and maps their user journeys across your website. This allows us to determine if your website visitors (once identified as real humans) are actually engaged in your content and interact with your webpages. We call these engaged visitors.

So looking at the 3rd chart, we can see that the clear winners at driving engaged human visitors are Google Ads and Reddit.

Human visitors vs bot clicks

The following chart is a breakdown of the clicks, human visitors, engaged visitors, and the percentage of visitors to clicks. Looking at the clicks (blue) and comparing them to the human visits (red) you can clearly see the difference between what these ad platforms are telling you when it comes to delivering genuine human website traffic.

A breakdown of the clicks, human visitors, engaged visitors.

I understand that every campaign is different, and there is no one advertising strategy or standardised audience, but the figures of our experiment campaign speak for itself. We worked out that the average cost per click during our campaign was £1.01, but the average cost per real human visit was £2.02, which meant half of our clicks during our campaign were from bots!

Our campaign was a big victim to ad fraud and we have learnt a lot from this by tweaking our future PPC campaigns and adapting our tactics to maximise our spend.

Conclusion

So, we have learned that measuring clicks is not a very good metric if you really want to know if your campaigns are driving real people to your websites, you need to focus on measuring human visits otherwise you will end up wasting half your marketing budget to ad fraud.