New Google Research: Can The Long Tail Be Covered Without Broad Match?

Wondering about the benefits of using broad match? Columnist and Googler Matt Lawson got his hands on real Google data to put the debate to rest.

Chat with SearchBot

google-gears-brain1-ss-1920

Google talks all the time about the necessity of broad match.  As a representative of Google, I’ve argued the importance of this match type myself. I believe in broad match, but talking to a lot of advertisers, it’s clear that many aren’t believers.

The common argument we hear is that heavily built-out exact, phrase or broad match modifier keyword lists are enough to cover the long tail.

Luckily, Google is a data-driven company. The product management team for match types actually wonders similar things as you do. Does broad match help you cover the long tail as intended? And can the long tail be covered without broad match?

A few months back, the team conducted a study on broad match performance, and I was lucky enough to get my hands on it. Even better, they were willing to letting me write up the results and share them with you.

The Methodology

Here’s some basic info about the stats I’m about to present:

  • A “long-tail query” in this context means a query that appeared at most 10 times in the time period the data covers (28 days during Q2 2015).
  • The stats I’m presenting here are relatively stable over time (based on studies in 2013, 2014 and 2015).
  • The stats within this study only include advertisers that spend more than 20 percent of their total spend in broad match. (Those that don’t use broad match at all would clearly have zero percent of their spend coming from broad match.)
  • These stats come from advertisers that are tracking conversions in AdWords.
  • All performance numbers are based on last-click conversions, which might not be the friendliest way to measure broad match keywords.

The Findings

What does broad match actually do in the real world? A lot.

1. Accounts that invest more than 20 percent of their spend in broad match keywords drive a larger percentage of their conversions (about twice as many) from long-tail queries.

I think this is a big deal. Accounts that rely solely on exact and phrase match keywords tend to see a smaller proportion of long-tail conversions.

In my mind, this is all the evidence that I would need to at least consider using broad match in my account. Long-tail conversions deliver a lot of value to an account; if you’re serious about capturing them, then you need to identify a workable strategy for broad match. (We’ll get into that below.)

If you use broad match in a significant way, your account is much more likely to uncover, reach and drive conversions from those long-tail searches.

2. More than 85 percent of all long-tail conversions come from broad match keywords.

As it turns out, broad match isn’t merely raking in long-tail searches that other keywords could gobble up. Long-tail conversions are predominantly captured by broad match.

Broad tends to be focused on coverage and reach, while exact is more focused on precision and performance. You should evaluate these two match types differently.

Broad match can help you be efficient with managing campaigns — you can cover long-tail searches with a broad keyword instead of using multiple exact/phrase (or, as we call them, syntactic) keywords. Exact and phrase will let you specifically manage queries that happen in significant volumes.

Now, this statistic and the ones above include both broad match and broad match modifier (BMM) together, so I wanted to dive a bit deeper into the performance of broad match versus BMM.

3. Broad match is the single best match type at covering the long-tail.

We looked at stats across eight different verticals and found that in each case, broad match did the best job of capturing long-tail conversions, handily beating exact, phrase and broad match modifier.

Broad match is the winner and still champion of covering the long-tail. In general terms, broad match accounts for about one-half of the long tail conversions, BMM covers one-third, and other match types make up the remaining one-sixth of conversions. (Did I just impress you with my fractions?)

Whether it’s Jobs & Education, Law & Government, Travel, Finance or something else, plain old broad match is the leader in terms of driving long-tail conversions.

One of our engineers even pulled a cross-section of keywords to see if BMM could cover these searches with some modifications to existing keywords, and in more than half of the cases, the answer was “no.” Broad match really is the ultimate way to capture conversions on long-tail queries. BMM isn’t enough.

Using Broad Match Smartly

If you want long-tail conversions (which, in my mind, is akin to saying “if you breathe air”), it’s time to start using broad match keywords. I hope the above numbers have convinced you to give broad match another shot.

Whether you’re finally giving broad another go or trying to optimize the broad keywords you’re already using, here are some best practices to follow as you enable it in your account:

  1. Stay on top of your Search terms report, and add negatives for poorly performing queries.
  2. Combine broad match with automated bidding. Target CPA is my favorite, but there are a wealth of other options. AdWords automated bidding sets query-specific bids for each auction. It’s not just based on your keyword — it goes all the way down to the query (and a ton of other stuff).
  3. Combine RLSA with broad match keywords that might not have worked before. If you’re still unsure about broad, you can dip your toe into the long-tail water by reaching those who have already visited your site.

I’m not saying that you need to transition your entire account to broad match or anything like that. It’s good to have a combo of broad match and other more restrictive targeting. Cover the long-tail with broad, but still give yourself a chance to pay special attention to your most valuable queries with those syntactic match types.

And before I sign off, I should mention that Dynamic Search Ads (DSA) are another phenomenal way to reach long-tail queries. It’s especially useful if you have a large inventory but don’t have the time to keep up with new campaigns for all of those products, services and other stuff that you offer. Maybe I can get my hands on the stats for DSA some point in the future.

Conclusion

Broad match is supposed to cover the long-tail, and it does exactly that. Based on the research here, I (along with the product managers for broad match) feel confident in our guidance to use this match type in your account. There are significant gains to be had.


Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.


About the author

Matt Lawson
Contributor
Matt Lawson is Vice President of Ads Marketing for Google, responsible for a broad portfolio of ads products including search, shopping, display, and analytics.

Get the must-read newsletter for search marketers.