How Keyword Research Is Tanking Your ROAS

By , Senior Account Manager at Hanapin Marketing


Keyword research is a vital part of every PPC manager’s task list. Hitting up your organic keyword data, hitting up the search query report, whatever you do: it’s all in the name of expansion, expansion, expansion. You want your account to be huge and always have that conversion and ROAS line creeping up and to the right.

But what does that leave us with? Probably an account with a TON of keywords, right? And, I’m sure you do a good job when optimizing to make note of keywords spending above goal without bringing in revenue to either lower bids or pause them. Hopefully you do this for keywords with a high loss as well. But what about all those keywords you’ve added with hopes of gaining more revenue and conversions that haven’t spent very much, but haven’t converted either? You still have hope for them, right? Maybe they’re long-tail and low volume, but just feel like they should work. What do you do with these guys? How many have you accumulated over the years of managing your account?

Well, I decided to dive in and look into this in my accounts. The results were pretty awful.

I really like using colors in excel, so I apologize for the ambiguous titles. However, the “white keywords” are keywords that have spent less than $20 during this time, and therefore don’t have enough data to make a call either way on.

Here’s the data for another account:

Again, “white” keywords didn’t spend enough to give me enough data to decide if the individual keyword was good or not. But I can see that they’ve had a loss of almost as much as the keywords I decided were awful.

So, what can you do? You want to always be growing your accounts and finding new, revenue generating keywords is a big part of that. My advice is to set aside a little test budget each month for “white” keywords. Don’t just throw keywords in your account in mass uploads and forget to look at how they’re spending as a whole. Set aside a “white keywords” budget. You should be able to figure, mathematically, how much spend you can tolerate without tanking ROAS to test new keywords. The rest of the budget should be spent on keywords with proven track records of revenue generation. Of course, seasonality and market changes will occur that can change keyword performance, but this is the general guideline.


What do you guys think about this? Have you performed this analysis on your eCommerce accounts? Do you already work a budget for these “white keywords” into your strategy?

Twitter Facebook LinkedIn Google+ Email Print More

8 thoughts on “How Keyword Research Is Tanking Your ROAS

  1. Ian Williams

    Hi Amanda,

    How exactly would you enforce a budget on these ‘white keywords’? Separate adgroup with an automated rule to pause them when they exceed a specific budget?


    1. Amanda West-Bookwalter

      Hey Ian! Yeah, Elizabeth pointed out that automated rules are a great tool in managing these keywords, and I totally left out that strategy. Fail! But yeah, using automated rules to manage the bids and make sure they don’t go over goal spends without converting!

    1. Amanda West-Bookwalter

      Yes! Thank you! I should have totally included a paragraph about using automated rules to help manage these. D’oh!

  2. GazPPCMonkey

    Good post though it is worth noting depending on your industry those “white” KWs need to have budgets adjusted for seasonality and should be assessed on a whole years worth of metrics if available, as snapshot optimising can lead to lazy pause heavy strategies.

    1. Amanda West-Bookwalter

      Thanks for taking time out of your day full of eating bananas and flinging poo to comment, PPC Monkey! Yeah, there are a lot of different strategies that should be considered based on the account, any seasonal trends, etc. The main take-away is to keep in mind the spend you’re throwing at these keywords and to make sure you’re constantly analyzing and optimizing them! It’s also good to remember that some keywords that have only spent a but in the past 30 days without converting may have converted the month prior, for a total spend under goal for the lead/revenue generated. So, keeping an eye on new keywords as soon as they’re put in until you can call them either way-bad or good-is always the best strategy!


Leave a Reply

Your email address will not be published. Required fields are marked *