Google keyword match type tournament: Place your bets on Exact

December 3, 2018 keywordana

The latest exact match update from Google looks promising for search marketers hoping to capture more relevant traffic but hesitant to switch away from phrase or broad match keywords.

Deciding which match type works per keyword is an age-old Google Ads question. After the latest exact match update, it’s in every paid search marketer’s interest to answer it. So we did.

On Sept. 6, Google informed the search community that the meaning of “exact match” would be open for interpretation from search engines. Ginny Marvin reported on the update explaining that aside from close variants like plurals and misspellings, exact match keywords would now also include synonyms and paraphrases. Google’s decision was reacting in part to the fact that people can search for something as simple as deodorant 150,000 unique ways.

These changes are intended to be positive for advertisers who can have their ads show up for more terms without having to perform exhaustive keyword research (although it puts greater pressure on campaign managers to audit search terms reports). The extent of how this update has impacted Google search campaigns at large has yet to be measured.

But with exact match keywords now showing up for so many paraphrases and related queries, why bother with other match types at all? Before doing away with broad and phrase match however, we as a search community needed to actually sit down and prove it to ourselves. We set out to measure and analyze exactly how the same keyword with two different match types would perform. The battle unfolds below, but first let’s visit how this complex analysis went down.

What and how were match types measured?

Currently there are four different match types offered inside Google Ads:

To measure the impact of the exact match relative to other match types we had to put them all under a microscope for the analysis. Aside from match types, we also focused our research on accounts in our database using Acquisio Turing machine learning (my employer) since its campaign level bidding data is unbiased to bids set for different match types of the same keyword.

Between January and October of 2018 there were 978,358 keywords that had at least 2 different match types among the machine learning-managed accounts.* (see research notes below) To make a valid comparison between different match types of the same keyword, they had to be used in pairs on the same day with the exact same bid. Filtering for this constraint reduced the number of candidates to 477,935 keywords.

Since a typical keyword has very little traffic on any given day, many of the keyword match type pairs had statistics on multiple days, which were combined to get a more robust measure. If the same keyword (e.g. “dog”) appeared with different match types in multiple different campaigns, those were considered as individual entries because their bids and other settings could have been very different.

It was observed generally that across the 470k+ keywords, people used mostly exact, phrase and modified broad match types. Broad match was used much less frequently on accounts inside our database.

Match-type showdown

Choosing a match type is the subject for PPC debates nationwide, leading us to this very showdown. Here to finally put some numbers to the conversation, in the midst of industry changes, is the match-type battle. First, we’ll announce the match type teams:

Place your bets and let the games begin!

Round 1: Broad vs. Modified Broad

There was a total of 19,182 keywords that had both broad and modified broad match types assigned (again pairs with the same bid on the same day).

Broad match keywords had a very similar average position to modified broad match
Broad match keywords had a 16 percent lower CTR than modified broad match
They had practically identical CPCs
Broad match keywords had 23 percent worse CVR
Broad match keywords had 33 percent higher CPA
Round 2: Phrase vs. Modified Broad

There was a total of 146,698 keywords that had both phrases and modified broad match types assigned. Here’s how they performed in the SERPs:
- Phrase match keywords had a similar average position as modified broad match
- Phrase match keywords had an 8 percent higher CTR than modified broad match
- They had practically identical CPCs
- Phrase match keywords had 6 percent higher CVR
- Phrase match keywords had 9 percent lower CPA
- Round 3: Exact vs. Modified Broad
  
  There was a total of 149,355 keywords that had both exact and modified broad match types assigned. Here’s how they performed in the SERPs:
  - Exact match keywords had a similar average position to modified broad match
  - Exact match keywords had an 18 percent higher CTR than modified broad match
  - They had very similar CPCs
  - Exact match keywords had 10% higher CVR
  - Exact matched keywords had 22 percent lower CPA
  - Semi-final conclusions
    After three intense match type battles we concluded that the only time MBM wins is when it’s up against broad. Otherwise, phrase match or exact match win, knocking MBM out of the competition.
    
    Final competitor rounds
    
    With MBM out of the competition, the final competitor rounds focused on unmodified match types.
    
    Round 4: Broad vs. Exact
    
    There was a total of 23,012 keywords that had both broad and exact match types assigned (again pairs with same bid on the same day). Here’s how they performed in the SERPs:
    - Broad match keywords were placed in 5 percent better position than exact match
    - Broad match keywords had a 40 percent lower CTR than exact match
    - Broad match keywords had slightly higher CPC
    - Broad match keywords had 15 percent lower CVR
    - Broad match keywords had 40 percent higher CPA
    When Broad match went up against Exact match, it lost big time! While broad match keywords had better ad placement overall, this is likely an artifact of the bidding algorithm because having the same bid for both match types can put the less expensive match type in a better position.
    
    Aside from ad position though, broad match keywords were more expensive and performed worse than exact match keywords.
  - Round 5: Broad vs. Phrase
    
    There was a total of 18,325 keywords that had both broad and phrase match types assigned. Here’s how they performed in the SERPs:
    - Broad match keywords had very similar average position to phrase match
    - Broad match keywords had a 23 percent worse CTR than phrase match
    - CPC between broad and phrase match was very similar
    - Broad match keywords had 17 percent lower CVR
    - Broad match keywords had 27 percent higher CPA
    When broad match went up against phrase match it also lost pretty bad. While average position and CPC was similar, phrase match came out on top with better CTR and conversion rates at lower prices.
  - Round 6: Phrase vs. Exact
    
    There was a total of 121,363 keywords that had both phrase and exact match types assigned. Here’s how they performed in the SERPs:
    - Phrase match keywords had very similar average position to exact match
    - Phrase match keywords had a 9 percent lower CTR than exact matched
    - Phrase match keywords had slightly higher CPC
    - Phrase match keywords had 4 percent lower CVR
    - Phrase match keywords had 20 percent higher CPA
    While exact match and phrase match won their individual battles against broad match, when they went head-to-head exact match was a clear winner. With lower costs to advertise and better performance exact match is the way to go.
  - After three intense match type battles we concluded that the only time MBM wins is when it’s up against broad. Otherwise, phrase match or exact match win, knocking MBM out of the competition.
    After the semi-final and final match type battles, there was a clear winner that will be announced in the next section. However, the tournament also surfaced another conclusion. It’s safe to say broad match is the least used match type and the lowest performer. However broad match can be good for discovery, effectively populating search query reports with potential keyword opportunities.
    
    While the broad match type team lost, there was one match type that stood out above the rest in this tournament.
  - se exact match for best performance at lowest cost
    
    The data is in, the bets have been settled. According to our findings, the cheapest and best-performing match type is exact every time!
    
    The latest exact match update from Google looks promising for search marketers who are hoping to capture more relevant traffic, but hesitant to switch away from phrase or broad match keywords. But if the update hasn’t inspired you to get more granular on match-types than hopefully the data presented here has. Even before the update, exact match keywords were performing better than every other match type and costing campaign managers less
    
    The one caveat with exact match is low volume, despite impressive ROI. It can be hard (i.e., impossible) to scale your campaigns when you’re cherry picking only the very best queries using exact match.
    
    While broad match should be used for discovery, phrase match and MBM can also generate decent quality results – and yet exact match came out on top in this analysis based on costs and performance. The key lies in understanding the role of each match type and using them strategically to scale and maintain profitability.
    
    TLDR summary: Exact match FTW!
  - All graphics made by Acquisio.
    *Researcher Notes: The results are based upon the median value of the comparison statistics. The comparison for different metrics narrows the initial numbers further. Average position only requires impressions, CTR and CPC further requires having clicks, and CPA and CVR conversions as well.