Many AdWords strategists have been suggesting a strategy of running exact-only campaigns. In this strategy, you have search campaigns which have exclusively exact match keywords. I have strong reservations about this strategy. Not only these campaigns are costly, they also miss a tremendous amount of opportunity in AdWords. Let us take apart their arguments one by one.
Exact-only Keywords offer high conversion rates
There is some element of data backing to this argument. Mostly exact keywords have a slightly higher conversion ratio than other match types. This is however not applicable to all accounts and campaigns and is certainly not a rule. There could be many keywords for which other match types might have a better conversion ratio.
Even if we assume that exact match keywords have higher conversion ratio, it would be a bad move to pause other match types. It is always better to change the bids so as to get conversions at same cost per conversion
Exact match will have lower cost
This argument has no logical backing. It is definitely a best practice to have a keyword in exact, phrase and broad match. This best practice only ensures that by doing so one can target keywords more accurately and will help bring down the cost. This cost reduction will only be visible when comparing a phrase match and exact match of the same keyword. In other words, it is cheaper to target a search query from exact match keyword when compared to a phrase match keyword.
There are some strong counter arguments against this strategy. Let us look at them.
Exact-only match searches can cover only a small percentage
Humans are extremely diverse and it would be extremely difficult to define every single keyword user might search for in google. Google sees at least 16-20% new searches every day. Exact match keywords can only cover at max 50% of your target searches. if you are not showing ads for half your customers, you are doing it wrong
For the same cost, one can get more conversions if you target all match types
Assume you have only exact match keywords and your campaign creates 100 conversions at 50$ per conversion which results in 500$ spent.
If we assume that other match types can get you an equal amount of clicks and conversions, we can split the budget among exact and other match types. We already know that if we reduce conversions, cost per conversion will also reduce exponentially. This should ideally result in the following situation or something similar.
250$ spend on exact and 250$ spend on other match types. Since cost is reduced for the same number of searches, your cost per conversion will reduce. let us say this number is 40 from the earlier number of 50. Which means conversions of 125 in total. Essentially we have generated 25 more conversions at the same cost
In conclusions never exclude traffic based on silly assumptions. Remember there are no non-converting keywords, only keywords which convert lesser or higher.