Bid Management Automation Basics: No Excuses!

Here, in the Paid Search column, there’s been a lot of talk of bid management tools and strategies for using them. Some of the uses of these tools can get pretty advanced. But in the real world you can do better with pretty basic bid automation strategies.

Many people are afraid or skeptical of campaign automation rules, in part because of misconceptions about how tough they are to use. Some are afraid that ceding control to an automated system will make a campaign worse, not better.

These are valid concerns, but there is a way to climb up the learning curve gently, assuming your software gives you plenty of options.

I’ve jumped on the bandwagon precisely because there is no need to get fancy in all cases. There’s almost always a benefit to using some of the simplest bid rules. So no excuses!

If you’re new to bid management, here some things worth noting:

1. Bid management rules can certainly make your life easier, especially if accounts are large, margins are narrow or the areas you’re advertising in are extremely competitive. Automation reduces the need to manually sift through accounts.
2. Say goodbye to the days of the automated tools forcing you to make frequent small bid tweaks across a whole account. As noted previously by Andrew Goodman, bid changes through the AdWords API costs money, which is why some legacy automation platforms feel the need to charge too much. Later-generation software typically works more surgically with your priorities to make more selective, less frequent bid changes.
3. With bid management, you can execute bid rules either automatically or manually (via suggestion mode). So, if you’re nervous about implementation, you can opt to receive email notifications of recommended changes and can make changes at your own discretion. The best platforms let you run simulations before enabling the rules. They also have a way of warning you when you’re making major changes.

To start, there’s no need to be super aggressive with bid changes. Instead, tackle the biggest problem areas, or gently massage segments of an account. Here are a few tips to get you started:

Pause or re-bid underperforming terms

Consider setting rules to pause terms if they haven’t reached a predetermined account goal (like CPA, cost, etc). For example, in one of our accounts, the client’s margins are slim and terms become less profitable if the CPA is over $10. As a result, we take a couple of very basic approaches to bidding here:

• Pause any keyword that has zero conversions on over, say, 500 clicks
• For all other keywords, if the CPA is above the target, lower the bid by 20%. Perform this check every week.
• Alert by email for all of these changes.

Even in these simple examples, you can see how today’s tools allow you to set a number of parameters and customize campaigns to achieve simple outcomes.

You can follow the exact same rules manually, but wouldn’t you rather save the time?

This approach can still lead to excessive caution, however, if you don’t use a long enough date range or click history, or if you fail to account for randomness and fluctuations in performance. As you get more sophisticated you’ll probably want to experiment with some more nuanced ways of bidding to CPA targets. Sometimes your instincts or other information may cause you to keep some keywords on a longer leash than a pure rule might dictate.

It’s also worth noting that bids aren’t your only variable. When campaigns are underperforming, lowering bids is a defensive maneuver. This shouldn’t let you off the hook in terms of coming up with creative solutions to building a great campaign, or in fixing other aspects of a campaign.

Adjust bids by geographic region

Take note of geographic areas that are performing well for your company and boost bids in those specific areas. Conversely, take note of areas that are not performing well and decrease bids in those areas. Take a look at the following example:

NYC

• Ad spend = $1000
• Conversions = 2
• CPA = $500

Brooklyn

• Ad spend = $1000
• Conversions = 100
• CPA = $10

In the example above, it would make sense to ramp up bids in Brooklyn and decrease bids in NYC. Note: the above example is for illustrative purposes. It’s important to ensure data is statistically significant before making changes in an account.

Consider dayparting

If you’ve been collecting data for a while, try to determine when conversion events occur. For one of our clients, we noticed very few conversions (in this case sales) in the early morning hours from 1 to 4am. We reallocated the budget to peak conversion times were able to improve the overall account efficiency. Take a look at the following data (for a one week time period):

Before September 2009

• Spend = $10K
• Conversions = 758
• CPA = $13.19

After September 2009

• Spend = 10K
• Conversions = 832
• CPA = $12.02

By running ads only during peak conversion times, this account saw a 10% increase in conversion with a 9% decrease in cost per acquisition.

Dayparting is more refinement than some advertisers want or need, but it’s an awesome strategy for competitive terms (like car insurance).