How do you bid on keywords that you don’t have statistics for?
I have seen many consultants just delete the long tail keywords from their Adwords campaigns. They didn’t even look if they performed well as a group.
I will make some suggestions on how to handle long tail keywords in PPC. Lets first look at the different keyword categories after numbers of searchers.
When you sort keywords after the amount of searchers you get:
1. Head keywords
2. Mainstream keywords
3. Long tail keywords
If you have keywords that get lots of clicks you want to make sure they are profitable. If competitions are driving click prices to high, you want to pause or put a low bid for these head keywords. On the other hand you want to use some time to optimized ads, landing pages, segmenting/filtering the traffic, derived organic traffic etc. if they are profitable or you can see them profitable in the future, because even small improvements can pay for your effort.
Is where most online marketing professionals use their time. You can analyze each keyword and see how it has been performing and plan further improvements.
Long tail keywords
Is keywords that have few or no clicks to built a statistic conclusive on.
How do you decide if a keyword that never had a user click on it is a converting keyword?
If your adgroup convert with an average of 7 % of the clicks and you have:
Keyword 1: One click and one conversion
Keyword 2: 15 clicks and no conversion
Keyword 3: 40 clicks and no conversion, but you just changes the ad(s) and landing page
Keyword 4: No clicks
Keyword 5: 3 clicks and 3 conversions
You can’t use any statistics rule, to say if any of these keywords, can be used in the future based on statistics.
The first keyword might just start with a conversion and never see a conversion again.
The second keywords might still convert more that 5% doing the next clicks.
The third keyword should have converted 2 times to keep the average, but it is still very thing ground to judge and with the new ads and landing page, you never know.
The fourths keyword has no clicks and therefore no statistic, but you might have nailed the keyword and therefore a 100% of all clicks will convert, even there isn’t many.
The fifths keywords is a keeper even you don’t know if you just been lucky.
As a group the five keywords convert 7 % of the time.
How I tread long tail keywords in Adword
When you don’t have enough statistics to base a decision on, I turn to gut feeling and use the following rules:
1. If I think the searchers, using the keyword, is looking for my product and the statistic isn’t conclusive showing otherwise, I keep the keyword
2. If I think the searchers, using the keyword, isn’t looking for my product. I delete the keyword.
It is to time consuming to think too much about keywords with few clicks. Wait until you can say something conclusive. Long tail keywords have often low click price and are very target. It makes it likely, that you want to keep them.
On the other hand it takes to much time and brainpower to handle long tail keywords where the searchers are searching for something that isn’t right on target. If it was head and mainstream keywords or you have a special strategy, it can make sense, but not in a long tail keyword group, where keywords are target to your product.
If you can’t make decisions on a keyword level, you might want to group long tail keywords and make decisions on an adgroup level. That way you can tread a group of longtail keywords as a head-keyword.
I like to keep my target tight, so I don’t end up doing to much work sorting keywords and my ads are right on the problem people are search a solution for. I do that by not using broad match, but have all keywords in both exact and phrase match. I use modified broad match to target a large ranges of keywords that I would otherwise had to generate by word combination. I go though all clicked keywords to maintain a high average quality score and get most of my clicks though exact march after some time.
What is your strategy for longtail keyword management?