Here's the question: What is the best way to estimate the total number of clicks I can buy for a specific set of keywords? Here's why I ask. I have a client who is trying to build a sales model and wants to use Adwords for PPC. We got into a conversation where he simply said this.. "I put in my 40 KW's (rocker widget, roller widget, blaster widget, etc..) and used phrase match to get the total traffic I can potentially purchase. There is no more traffic to be had, because based on these phrase match KW's, every other possible combination should be represented in the numbers the tool gave me." Now, I started to think... "is that true?" let's see, so... would "big green rocker widget" & "rocker widget in bristol" be included in the potential traffic he had generated? I think so. How about the really long tail stuff? I guess as long as phrase "rocker widget" was there it would already be counted... How about exact match traffic? Would the traffic for the exact match KW "killer cool large rocker widget", be included in the traffic estimate for the phrase match KW "rocker widget"? If it is, then actual click inventory is something far less than anything I've been calculating. Thoughts?
While having the phrase match "rocker widget" for example in your campaign will trigger for any search containing that phrase, I never gave a thought of the way the keyword tool shows its information. Thinking about it, I *believe* it is the number of all searches containing that phrase. It makes sense but I could be wrong. Therefore, you can use the phrase match numbers to estimate the number of monthly searches for that phrase. However, you don't want to bid on just a two-word keyword or just the phrase match. You want to bid on the longer-tail and the exact matches. The more precise your keyword to the search the better. Knowing exactly how people do their searches can go a long way in optimizing your campaign.
Lucid, Yep... I'm running into the problem of exactly how much traffic there is to purchase. For instance... if you yse the adwords tool and look for "rocker widgets", [rocker widgets] and rocker widgets, you get huge buying potential. If you just use "rocker widgets" and [rocker widgets] you get a different lesser number (no broad match... makes sense). However, it seems you can't simply add the traffic from "rocker widgets" and [rocker widgets] to come up with a potential traffic number because with "rocker widgets" you should be getting the [rocker widgets] traffic. Using "rocker widgets" you should also get all of the long tail stuff... just at a higher price. Leveraging Adwords to drive the cost down is a goal for later, and will involve hundreds of long tail KW' and KWP's once I discover what they are... The real question is... is phrase match the best way to get an idea of all of the traffic that CAN BE purchased? This has been baking my noodle and I've asked this question on several forums without resolution. One of the issues is that Adwords looks at exact match first, then phrase, then broad... if you're paying for all three... I'm used to using all three and getting clicks for all three, which seems like far more traffic that just setting up a phrase match campaign. whew!
Personally - I would recommend split testing phrase and exact match, make sure you track what converts.. then narrow down after that