Here are some stats for one of the campaigns I manage, showing how ROI can differ hugely according to match type. A few points to note which show how reliable these results are:
- almost every keyword has been added for each match type
- the bid amounts are the same across match types
- data comes from approximately 20,000 keywords
- took over a year to accumulate
- has a very mature negative keyword list of over 400 words
The more closely you target your keywords, the higher your ROI. In this campaign, exact match produces 4 times the ROI of broad match, and nearly twice that of phrase match.
Broad match is great for finding new keywords to bid on, but it also finds a whole lot more that won’t produce results for you. This emphasises how important it is to regularly check your server logs & look through the raw search queries for keywords to add to your negative list.
Does broad match go too far when expanding your keywords for related searches? If I didnt have such a good negative keyword list for this campaign, I could very well have made a negative ROI on broad match.