September 12, 2017
Our instinct as marketers can be to chase what’s working and max out performance everywhere possible. When our optimizations are going swimmingly and we see conversion volume is growing and CPA is shrinking like a kid in a Rick Moranis film, we get to pat ourselves on the back and push more budget, energy and time into our top performers. This is not a bad instinct, but I want to remind everyone that the inverse can be just as powerful especially if you’re facing budget constraints, so if you can eradicate erroneous clicks, you can have more of that gravy to push towards productive initiatives.
How to Use Negative Indicators to Improve Ad Performance
Pull a location from the Dimensions Tab to find out if there are regions where your products or services are in highest demand and conversely, where you are spending marketing dollars but not meeting conversion volume or CPL goals. Place Negative bid modifiers on under-performers or add high-click, low converting locations as negative locations.
If it’s not already established in your account, layer in Income based targeting from the Advanced Settings Section of the Locations menu. Add all income levels and give the account a few weeks to collect data. These levels are frighteningly broad, and it’s likely that Google’s algorithms aren’t 100% accurate in determining who makes what, so you don’t want to necessarily eradicate an entire group. However, there is ample opportunity to use negative bid modifiers on income groups that are not performing well in your accounts or that have high CPA and low conversions. If you sell a luxury product it might make sense to bid down on lower tiers, but check the data first before placing your modifiers.
Even if you’ve been in the PPC biz for a while, don’t take this for granted. Pull an SQR once a month and see what’s spending and not performing or worse, what’s not at all relevant. I once had a new client who was bidding on keywords for a product called “weave” and the keyword was in the account as broad. They were spending upwards of $500/month on searches related to hair weaves. We added “hair” as a negative and dropped CPL in that campaign significantly.
Again, depend on the data here to guide you, even if it’s counter-intuitive. Use Analytics or Audience Insights Data from your “All Converters” remarketing list in Adwords to see what age groups and genders are supporting your marketing efforts and which ones are ineffective. Apply negative bid modifiers to low converters or remove entirely. It’s easy to get boxed into traditional beliefs about demographics. Don’t do that. Let the data guide you. For example, it seems intuitive that gaming products would perform best with a younger demographic. It might surprise you to learn that over 45% of gaming searches are conducted by individuals 35 years old and up. (Source: Mobile Search and Video Behavior Analysis, Millward-Brown Digital, US, 2015)
For your display efforts, pull a placement report and see if there are websites that are strangling your budget without providing any value. Add them as negatives immediately. Also, under the +Targeting Option, go to the Campaign Exclusion options to exclude specific site categories you don’t want associated with your brand. Use this liberally.
Pull device reports regularly and see how performance stacks across desktop, mobile, and tablets for your various conversion types. I primarily run lead gen accounts where I depend heavily on this level of data. Getting someone to fill out a contact form is usually pretty easy from a mobile device, however it gets harder when my conversion also requires that someone provide a full on data drop with attachments. In an analysis performed for a staffing firm client, we used back end data (client’s CRM) paired with front end data from our accounts and discovered that leads coming in from Mobile don’t place in jobs as well as leads originating from desktop. Now, the data didn’t say why that is, but armed with this knowledge we immediately dropped budgets on mobile specific campaigns in favor of desktop even though on the front end mobile leads had higher volume and lower CPL. We know we are being throttled ever-more into a mobile first world, but if your data tells you to stick with desktop to meet your objectives then do it! However, first eliminate mobile site speed from the equation. Google offers a great tool for quickly testing mobile site speed. It identifies and creates a great looking report to show clients weaknesses in their mobile game. Additionally, the report offers actionable insights on how to improve mobile speed.
While it feels intuitive to latch on to what’s working , use negative indicators to go a step further in your campaign optimizations. Don’t trust your instincts, look at your data and if it’s not working take the advice of Liz Lemmon and “Shut it down.”