The Long Tail Keyword Myth | Hacker News Reader

The Long Tail Keyword Myth

clicteq.com

25 pointsClicteq14 comments

The optimization of time is by far the best argument. That said, this leaves out that there is a relatively easy way to generate long tail keywords in a part of your workflow that absolutely should exist: mining the search query report. You should be doing so anyway to make sure you are using negatives effectively, and it's very easy to add long tail queries you notice from there without adding a bunch of time.

That is correct, however to improve performance you must also write adverts that are tailored to your long tail keywords. Otherwise you will end up with the same results as you would have done provided that you had just used shorter tail keywords picking these up.

This is true to a degree. Simply bidding on the long-tail keywords will improve your performance, although probably not enough to be meaningful. But going through the SQR shouldn't yield enough long-term queries for this to pose much of a problem, particularly given that you should be periodically rewriting ads anyway.

> mining the search query report

You haven't even got to do that!

It takes very little time, in many industries, to build 100,000s of keywords. It is just string concatenation:

{modifier} {Size} {Colour} {Clothing item}

Adjective: buy, cheap, buy cheap, cheapest, best, quality Size: large, small, extra large, petite Colour: .... Clothing item: t-shirt, tshirt, hoodie

That's hundreds of keywords quickly.

Keyword research is something I think people overdo. You know the sorts of keywords that exist, you need to get them all into the system, and then optimise en masse.

If you are going to use keyword multipliers or concatenation then I would suggest that you go into Google Keyword planner and then use the find search volume option. Once you've done this use the filters and set it to monthly impressions greater than 10 and this should weed out all of the low search volume keywords that clutter your account.

This is a start, but the underlying data is something like $200k of SEM spend across what looks like a couple of campaigns. How confident are you that this is not specific to the campaigns/industry/clients/etc?

Also, Wesley, the scrolling on the site seems to be broken/hijacked.

Hey, you're correct about the data being from a couple of account with a few hundred thousand in spend. We have however run this on several accounts with several million in spend and found very similar results in varying different industries. By all means test this on one of your own accounts and see the results for yourself, Id be interested to see what you find!

His results are very general. They depend only only on the fact that keywords with more words have lower search volume. For any industry for which this is true (i.e. any), you will find qualitatively similar results.

If you think about it, it's a trivial result. In any other industry if I tell you "I'm going to focus on increasing the ROI of 10% of our sales", you'd laugh. Why not focus on increasing the ROI of the other 90% of the sales?

However, it does depend on the account. On an account where you have the same number of 2 word keywords as 5 word keywords, of course the 5 word keywords will have a tiny tiny percentage of the impressions (say 1%). But obviously a marketer who goes for the long term strategy will tell you that you need 100x more 5 word keywords, in which case the 5 word keywords would have the same number of impressions as the 2 word keywords. But in comes the author's argument that you can do virtually nothing to manage that many keywords (we'd be talking millions here).

It is certainly correct that in any other industry you would focus on the 90% of sales not the 10% however when it comes to Adwords many marketer seems to completely overlook this in my experience.

I would disagree with your second point though. If you take a look at the search query data you will see that over 80% of the searches come from where users have typed less than 2 words in length. Only a small number of impressions are coming from searches that are 5 words or more in length so even if you had 1 million 5+ word keywords you would never be able to get the same number of impressions as you would with 2 word keywords if that makes sense.

By all means it is possible to manage accounts with huge numbers of keywords and we regally do at Clicteq through using things tools such as Adwords Editor and Adwords Scripts, however I would argue that for doing this with 5+ words keywords is probably not an effective use of your time

> I would disagree with your second point though. If you take a look at the search query data you will see that over 80% of the searches come from where users have typed less than 2 words in length. Only a small number of impressions are coming from searches that are 5 words or more in length so even if you had 1 million 5+ word keywords you would never be able to get the same number of impressions as you would with 2 word keywords if that makes sense.

No, it doesn't make sense.

10k 2-word keywords: 1M impressions/day

10k 5-word keywords: 100k impressions/days

If I now increase the number of 5-word keywords to 100k, what do you expect will happen? I expect to get abobut 1M impressions/day

If you look at our data that shows how long each search that users are making is you will see that 94% of searches made by users are 4 or less words long.

Short tail keywords will still catch long tail phrases with BMM for example the two word keywords +ppc +agency will still catch a search like "ppc management agency prices in york" which is along tail search query.

If you add 1000000000 million long tail keywords to capture that 6% of long tail searches that will never = the 94% of searchers that are shorter than 4 words in length.

So 100K 5-word phrases = 6K impressions and 10K 2-word phrases = 94K impressions

No mention of brand v. generic in the entire article. If you're going to analyze keywords and ctr, you separate brand and generic no matter what.

Obviously two/three word keywords bring in more volume and have a higher ctr.. They include your brand name. Brand ctr is always 25% or above, and non-brand fathead keywords have much much lower ctr and impressions.

IMO all the data is suspect if you leave brand in their. As it stands its a better comparison of brand v. generic rather than fathead v. longtail.

Great point about Brand v Non brand. This is a relevantly new company with very very few brand searches (less than 10 per month) so is unlikely to have skewed the results. I may add that as a point within the articles if thats okay?

Yeah I definitely would, that lends a lot more reliability to the data.

Reply on news.ycombinator.com