Structuring Automated Bid Strategies

You may have been advised that the less data buckets the better when implementing automated bid strategies - this is true. It’s important to give the technology enough ability to make decisions for it to yield progressively improved results. Although bigger data buckets allow the algorithm to reach it’s configured goal faster, merely achieving that goal doesn’t mean that it delivers the business goal.

Brands need to achieve an ROAS or CPL goal with their media spend - let’s call this a digital goal. Digital goals often reflect the final objective that a brand wants to achieve, which is after visibility, site traffic and CPCs among others. This creates a paradox for paid search practitioners as the strategies to grow various metrics or KPIs are different for each.

This is where structured automated bid strategies meet proper campaign segmentation. Simply put, campaigns are structured to support specific products or services. Group campaigns supporting the same product or service together, leverage the same automated bid strategy and this creates levers to optimize each product or service to a digital goal.

While this sounds intuitive, it still does not incorporate the multiple objectives brands need to manage before hitting that ROAS or CPL. Hitting an ROAS goal of $5 is only noteworthy if the clicks generated in this period were at similar levels with previous periods. Otherwise, you may be giving up long term sustainability for immediate gains.

Within each product line or service line, there will be campaigns that support different levels in the funnel - nonbranded product terms, nonbranded product specification terms, branded product terms and so on. This is where complexity starts to kick in but the levers created by grouping campaigns pay off. Structured automated bid strategies need to consider campaigns with similar objectives (not just a digital goal) to create the data buckets recommended and provide levers to make ongoing optimizations.

Here are some useful questions to answer in building out the right framework for automated bidding:

• Should campaigns be grouped together by keyword type - brand, nonbrand or competitor?

• Should campaigns be grouped together by point in the user journey - upper funnel, consideration, specifications, branded, etc?

• How should ROAS or CPL goals be determined for each campaign grouping?

The answer to these questions will differ depending on the volume of traffic coming through to an advertiser’s account - that is, how much the brand is spending on paid search.

Nonetheless, keep in mind that each grouping does not have to be a mirrored image of each other. Grouping select campaigns differently is usually the key in building the right automated bid strategy infrastructure.

Subscribe or follow Quisit on social media for more thoughts around the changing landscape of digital marketing.