With Google’s systems for detecting invalid clicks becoming more sophisticated every day it’s unlikely you’ll ever run into them, but it’s still worth knowing about them to prepare for those scary blog posts you’ll read occasionally.
Although Google’s detection systems are becoming sophisticated enough that invalid clicks within paid search rarely even show up in your data, it’s still a good idea to have a basic understanding of what they are and how to deal with them.
Invalid clicks, as defined by Google, are clicks that are illegitimate, unintentional, or resulting from malicious software. The most common examples of this are:
- Clicks by bots or other automated clicking tools
- Manual clicks, used to increase your costs or the profits of websites hosting ads
- Clicks that provide no value (i.e. accidental double clicks)
- Impressions intended to artificially lower your CTR (not technically clicks but work on the same principle)
Identifying Invalid Clicks
The good news is that Google has two separate systems in place to detect these, and, in all likelihood, you’ll never have to do anything about invalid clicks yourself. The first system automatically analyzes every single click on several data points, such as IP address, time of interaction, duplicate interactions, and various other interaction patterns.
If Google identifies a click to be invalid it is removed before ever even showing up in your reports. This means that the vast majority of time, invalid clicks can be safely ignored (if you’re curious, you can add a column for invalid clicks, invalid click rate, invalid interactions, or invalid interaction rates at the campaign level or within the dimensions tab).
Occasionally, some invalid clicks that were missed on the first check will make it into your data. However, Google has a second system which diagnoses click patterns a few days after they occur, and if it detects invalid clicks then these will be removed and their cost automatically credited to your account (you can see these in the billing section of your account, labelled ‘invalid activity’).
When Invalid Clicks Slip Through The Cracks
On very rare occasions, invalid clicks will make it through both systems. When this happens, it’s unlikely to make any noticeable difference to your data or costs, since it’s almost impossible for large amounts of invalid clicks to make it through. However, if you suspect that you’ve been charged for fraudulent clicks that weren’t spotted by Google, send them an email with your reasoning and evidence (i.e. estimated spike time and dates, web logs, etc.) and they will assign an engineer to run a manual review of your account.
Of course, you will hear stories about advertisers with very small budgets and an unscrupulous but crafty competitor who, while constantly changing IP address, will click over and over – quickly consuming said advertiser’s limited daily budget. While these do eventually get refunded, it can still make it hard to run campaigns effectively when the budget is being constantly depleted.
Most of these stories are relatively old, as Google’s systems for preventing this are constantly improving. However, if this does happen to you and emailing Google doesn’t seem to work, the best solution may be to contact the competitor directly and try to see if there’s a mutually beneficial arrangement you can make. If necessary, you can also make use of third party click monitoring tools, although it’s important to choose wisely as there are a number of ‘snake oil cures’ out there. However, Google’s systems are already extremely advanced and improving every day, and we believe they’re more than enough protection against this kind of activity.
So, should you worry about invalid clicks? The short answer is no – chances are you’ll never even notice them. It’s very rare that they will ever even make it past Google’s systems, and if they do, Google will likely sort it out pretty quickly.
Here’s our position on invalid clicks:
At Merkle, we do our utmost to ensure that there is minimum loss of client spend to fraudulent activity. We do this in the following ways:
- Working with technology providers to ensure there is a layer of protection against non-human traffic. This includes the removal of invalid impressions and clicks from all reporting inclusive of billing.
- Where additional fraud is detected by the agency, there is a provision from technology partners to provide additional credit for the loss of media spend.
- As an agency, we continually pull reports and manually identify any suspicious sites to block from all activity across our campaigns.