Is seperating broad, Phrase & exact match helpful? Long-tail Keywords v. Root words?

I have two questions:

1) Does separating broad, exact, and phrase match help improve the overall Adwords campaign in terms of higher positioning, lower ad spend, or higher CTR? (using the 3 matching options in the same Ad group with the same ad text).

For example, separating green eggs into green eggs, "green eggs" and [green eggs].

2) If I use long-tail keywords, such as "green eggs and ham" - does this provide any benefit than simply using "green eggs", provided that both are in the same Ad group with the same ad text?

 

of course it's useful to do.

let's say your exact match gets a conversion rate of 3%, but the others are a .5%. wouldn't you want to have that info available to you and adjust your campaign accordingly?

 

No, try to keep only one matching option per keyword per account.

if you keep the 3 matching options you will split your quality score in 3 different cells, but if you keep just 1 you will put it all in just 1 keyword - this rule is also for your statistics (clicks, cost, cost per conversion, etc...)

But if you decide to keep broad match you need to add tons of negative keyword or it will not work.

When you prospect long tail keyword, your are doing well: this give you a better view of your market - this makes your understand better what people is searching in google.

Also during research of new keyword you have the opportunity of to decide to add negative keywords if you are using broad match - for example you can add the broad keyword:

green eggs

and add the negative keyword "-ham" (only if it is not interesting for you) to your campaign or add group, if begin to do it with a lot of negative keywords it put you a step further of your competitors due your "green eggs" keyword has a lot of traffic due it is broad but it has a lot of negative keyword associated to it - to you will push your CTR not showing your ad to non relevant audience.

Also another great advantage of long tail is that you add if you have conversion code installed you can study better the detailed keywords (better segmentation) that convert and the keywords that don't convert after a long time and take conclusions from it - so you will know what keywords worth more to focus more (increase the bid) and the keywords that even do not worth to keep in your account

 

Is the quality score more important than conversions? What if all 3 versions of a keyword convert?

[dog] = 10% conversion (5 sales a day)
"dog" = 7% conversion (3 sales a day)
dog = 1% conversion (1 sale a day)

Assuming that all instances are converting in a profitable fashion why would one want to kill of two and keep the other? Maybe the quality score for the kept word goes up but overall sales could just as easily (and typically do) go down.

Thoughts anyone?

 

I think that is impossible to determine if you are getting any advantage using these method - remember that each one is competing with any other, also do not means that one is getting a better performance compared to another.

And now that you have these numbers what do you thing from them? delete 2 of them and left only [dog]? I think that these numbers do not represent something that can really be used - I think that they are just letting your statistics more difficulty to understand as your stats are being split in 3.

In my option there is only one case that you should go using more then one matching in the same account - when you have extensive broad keywords like "dog" - but you should go using at least 2 AdGroups A and B and use Embedded Match on group A:

Group A - broad keywords:

dog
-[dog] <----- Embedded Match

adGroup B - exact keywords:

[dog]


ok, so in adgroup A you will have ads related with dog accessories as it will no display for who type only dog keyword alone (Embedded Match - using the negative [-dog]) only also on adgroup A you need to ad tons of negative to protect your quality Score and conversion as you're using broad match.

On adgroup B you will create ads for who type directly this specific keyword , you may bid lower for this group as it is very subjective and you may artificially disable content network for it (bid very low CPC value for content). But it has the advantage that you don't need to worry about negatives and it always get good quality score compared to other keywords.

you do not need to add only 1 keyword (dog) for the group, you can add several pairs of keywords for both groups - but you are limited by the group adds.

This model above using Embedded Match is the only one that makes sense to me on having more then one matching option on the same keyword on one account,

in my opinion, if you use matching option like this aleatory: you will split your conversion statistics, making more difficulty to know the keywords that are really converting - for example tell what "type of keywords" better convert in this example:

[dog] = 10% conversion (5 sales a day)
"dog" = 7% conversion (3 sales a day)
dog = 1% conversion (1 sale a day)
[horse] = 18% conversion (9 sales a day)
"horse" 10% conversion (5 sales a day)
horse = 3% conversion (2 sales a day)
[cat] = 12% conversion (6 sales a day)
"cat" = 4% conversion (2 sales a day)
cat = 1% conversion (2 sales a day)

now what would be easier to analyse, if you just decide using broad match + negative keywords on you account instead of using 3 matchs for each keyword?

dog = 10% conversion (9 sales a day)
horse = 18% conversion (16 sales a day)
cat = 12% conversion (10 sales a day)

 

It would be equally difficult/impossible, with the present tools/data available, to determine if you are putting yourself at a disadvantage using this method.

I keep them all as long they profitable. If they convert at a level in which I need them to I don't see a reson to kill them off. I may adjust bids and or creative to increase conversion but is a keyword is producing I keep it.

Honestly, I wouldn't care what "type" of keyword converts, as long as it converts. Broad phrase or exact...I'll use em' all if they are profitable.

I think as long as you have access to the right tools to analyse the data and make good decisions using a combination of matching options works. That said, I see your point and am not discounting your method either. That's one of the beauties of PPC - there are multiple ways/methods to accomplish things in a cost effective fashion.