Determining demand for keywords

numbers don't match up

[limitup]

I'm trying to figure out the best way to determine the demand for certain keywords. One simple example as to why I'm confused. According to Overture, they had close to 6000 searches last month for my 2 keyword phrase in question. Then I go to Google AdWords, and they say that AdWords averages only 6 clicks per day for this term. So my question is very simple - how is this possible? If Overture had close to 6000 searches for the phrase last month, for the sake of example let's say it's safe to say at least 10,000 people a month search for this phrase. How can AdWords only have 6 clicks per day for this term? What am I missing here?

[Mark_A]

one of the issues is search base or sites included in the stats, you could also use wordtracker which has a further different base of counting ....

Then there are webmasters checking their own positions of course including some using automated software .. ugh

The only way imho that any of these are in any way valid is to suggest that term 1 is more popular than term 2.

[Macro]

you could also use wordtracker which has a further different base of counting

I suspect that WT would be completely useless for this purpose as I have found that it shows exactly the same results as Overture.

[storevalley]

Then I go to Google AdWords, and they say that AdWords averages only 6 clicks per day for this term

A couple of points that might help ...

• Estimated number of clicks received in Google per day is related to the amount you bid. Try ramping this right up to see if it makes any difference

• 6000 searches per month is only 200 per day. And they are split between all of the sites that appear. Google is guessing the number of clicks that your site might get, not all clicks for the term

• Figures do vary from month to month

• Some terms can be very popular at specific times (e.g. weekends, just before a big show, etc)

• The Overture search tool is reporting on Overture's results. They are nothing to do with Google.
  

So this figure doesn't sound so strange to me ...

[fom2001uk]

Mark hit the nail right on the head. These tools should only be used as a guide, to track historical phrase popularity over time. Phrase 1 is more popular than phrase 2 is as accurate as you'll ever get (and is VERY useful).

But forget about the numbers, they are totally useless!

[storevalley]

But forget about the numbers, they are totally useless!

Not always. Forget them ... no. Take them with a pinch of salt ... I'd agree with you.

Phrase 1 is more popular than phrase 2 is as accurate as you'll ever get

This does mean that you'd have to trust the numbers to some degree ... ;)

[limitup]

Ok so I guess the real question is - how can I get some idea of the type of traffic that will be possible for a top position for a particular keyword? When you guys (and gals) have an idea for a new site, how do you determine if there is enough demand to proceed? How do you gauge the potential from a SEO standpoint?

[buckworks]

Another thing that skews the numbers is that Overture's reports lump together more variants than WordTracker's reports do. Singular vs plural, mispellings, apostrophe's, hyphenated-phrases, etc ... Overture reports them all under one term whereas WordTracker reports those useful details separately.

[limitup]

Oh and one other related question, the answer to which may explain a lot of what I'm not understanding. Does anyone have any stats regarding the percentage of clicks on the #1 "organic" search result vs. clicks on the #1 adwords listing? I assume most people see paid listings for what they are, and while paid listings do get lots of clicks I would assume the #1 organic listing should get more clicks than the #1 paid listing. But this is just a guess. Does anyone have any stats on insight into this type of thing?

[storevalley]

how can I get some idea of the type of traffic that will be possible for a top position for a particular keyword?

For Google, the sandbox & Wordtracker probably give about as good a guess as you are going to get without actually going after some of this traffic and analysing your own logs (or chatting up somebody that has already done this)

As already observed, keyword suggestion and analysis tools aren't always accurate for a number of reasons.

When you guys (and gals) have an idea for a new site, how do you determine if there is enough demand to proceed?

That depends on the site's topic. I call this on a case by case basis. It's also worth noting that SEO isn't the only (or even the most obvious) way to market a site in some cases ...

[airpal]

If you want to get accurate numbers, optimize yourself to be #1 in google for a specific keyword phrase, and compare your daily click-throughs to Wordtracker's Predict column for that phrase. This should give you a good idea of clickthrough rates for your specific niche. However, I get the feeling that different categories of keywords might get different click-through rates.

The way I look at it, if I can't make it into the top 3 (3 PR 8 sites or above), I usually don't bother trying to aim for the "first SERP page". Think about how many people are realistically going to browse to listing number 8, if they found what they want at listing number 3! Just my 2 cents, and oh, when you get to number 1, please let us know what stats you came up with! :)

[onedumbear]

The way I look at it, if I can't make it into the top 3 (3 PR 8 sites or above), I usually don't bother trying to aim for the "first SERP page". Think about how many people are realistically going to browse to listing number 8, if they found what they want at listing number 3!

Some people like to shop, some people like to surf.
Some people like to compare prices, and some people like to get second opinions on information, especially on the net.

Depending on your topic you may get a good number of hits for results that show up on the socond or third pages of results. It always makes sense to shoot for #1 though.
I have one particular informational site.
It comes up in the top 3 spots for a lot of targetted keywords. But if i combine the nuber of hits it gets from key words that show up on 2nd or 3rd. pages of serps, the number is very close to the number of hits it gets for it's top 10 keywords combined.
It's all relative. "just depends"

Don't forget that a good title in the #5 position will get a lot more hits than the top 4 positions if there titles do not catch the surfers interest.

[edited by: agerhart at 5:10 pm (utc) on Oct. 24, 2003]

[airpal]

onesmartbear, a very good point indeed, in fact what you wrote was right on the money, but I was too lazy to include it in my post! :)