Bidding Strategies: A Pair of Case Studies

By Kristina McLane | @Hanapin | Account Manager at Hanapin Marketing

Sometimes, I think there is one single mantra that runs through the mind of every PPCer, “Let’s test that.” Ad testing and conversion rate optimization get a lot of the testing hype, but what about bidding strategies? At intervals, experimenting with bidding strategies gets sprinkled into the testing mix, but not enough.


How do we spend money in paid search? The answer is the click, which is influenced by ad copy, but also our average position. Average position is determined by how much we are willing to pay per click, how much our competitors want to pay, and by quality score. The point is the amount we decide to bid is just as important as the other paid search bits. Bids determine how much we spend and the strategy makes sure that we are bidding wisely.


More Bidding Strategies Out There Than Can Be Listed!


There are tons of options of bidding strategies to try in our accounts. Below is just a short list of a few broad examples:



However, when you get into it, there are tons of other options that make up the subcomponents of that list:



With so many options, how do you choose which bidding strategy is best for your account? If there were ever a party filled with PPCers, this question would be a great conversation starter. There is much to consider before making that decision:


  • Is your account eCommerce or lead generation?
  • What is your goal? Should you be looking to for consistency, efficiency or growth? This generally ties into the goal of the account: volume, CPL, ROAS
  • How consistent is your account? How dependable your keywords are ties with how frequently the inventory varies on the website.
  • What is your typical conversion volume? Conversion volume requirements can prevent use of certain bidding strategies, such as conversion optimizer.


Case Study #1 – Goal of Volume


In this first example, our goal was to increase conversion volume and spend in this campaign. This was a lead generation account with conversion volume in the 100s. With lead generation, it is very rare that you have to worry about conversion consistency, because the website’s services generally do not vary. Taking all of this into consideration, Google Conversion Optimizer is one of my favorite bidding strategies to test on high volume, lead generation accounts.


We turned Conversion Optimizer on the first week of January at 15% higher than Google’s recommended “Target CPA”. Again, we wanted more conversions and spend. An increase in cost-per-conversion was an acceptable tradeoff for fast growth.


Screen Shot 2015-03-16 at 12.34.34 PM


At that time, we saw January’s conversions grow by 136% and cost increase by 189%. Even January to February, saw increases of 53% and 29%. After 60 days of testing this bidding strategy, CPA Bidding proved to be a winning bid strategy for this account. Spend and volume increased, which was the goal.


Case Study #2 – Goal of Consistency


For my second example, I wanted to focus on an eCommerce account that has high inconsistencies in conversion volume each week. This account offers events that vary week-to-week by location. Keywords and campaigns that had amazing ROAS one week were terrible the week after. In this situation, our goal was to increase the consistency of the amount of revenue that came in each month, even with the varying products that were available!


For this account, we set up two types of automated bid rules that were turned on and off based on the upcoming offers over the next 6 weeks. We knew that we always wanted to be showing for any upcoming events, but people were more likely to actually purchase a pass within 6 weeks of the event.


These automated rules were setup in Google. The first automated rule favored keywords with high assisted conversion metrics when there were no events in the upcoming 6 weeks. Overall, bids were reduced to put more of our budget towards locations with upcoming events. The second bid rule favored keywords that typically saw direct conversions in the last 90 days and were used for campaigns with upcoming events.


3.17 pic2


By tailoring the bidding strategy to mirror the actual behavior of the business, we will be able to change the variance in revenue from 97% month-over-month changes in the previous 90 days, to changes as small as 2-3% in the following 90 days. While we no longer see huge spikes in revenue one month, we also no longer see huge dips as well. Consistency in revenue helped us plan strategy for the month ahead.


The key factor to remember about automated rules is that they still require regular maintenance. In my example above, constant monitoring of our offers and inventory on website was necessary to change bid rules based on upcoming events. Automated rules should never be put in place and forgotten about.


Final Points


Changing a bidding strategy can make a huge difference in achieving paid search goals. It does not come to a surprise to anyone who works with multiple paid search accounts that all paid search accounts are different. What works for one account could tank another. Looking at the typical behavior of the account should help choose the correct strategy to improve performance.


When trying a new bidding strategy, the key to success is to revisit after 30 days or 60 days the strategy and determine if it worked. It would be amazing if every single test worked, but winning tests every time would probably lose most of us our jobs. My examples just highlighted bidding strategies that ended up being successful. However, in many situations, adjustments to bidding strategy need to be made to optimize results to goal. Bidding strategies are important to test and to always be changing to reflect the ever-changing goals.