Hi Sam

I'm not a search engine expert but we have done a fair bit of this ourselves (both 'organic' / 'algorhythmic' / 'editorial' / 'free' listing and pay-per-click) and my few tips would be:

- you need to work out what people will actually search for that is relevant to your proposition. How do you know this? The cheap way is a bit of common sense and asking your mates / existing customers.

- you can then use tools like Wordtracker (http://www.wordtracker.com) to check against your own list: are there others being searched for that are similar and should be on your list? How many searches are there for your chosen terms? This will also help give you an idea of the kinds of volumes of traffic you might expect to be able to drive

- given your size, budget, proposition I'd suggest going for pretty specific search phrases (not keywords). Bear in mind the following (recent research from Overture) on search term distribution:

36% people search using 2 terms
28% search using 1 term
22% search using 3 terms
9% search using 4 terms
5% search using 5 or more terms

Complement this with research from Espotting which shows that despite the above, most PPC (pay per click) advertisers compete on 1 word and I think the message for smaller players is - buy 2 or 3 word phrases. Why? 2 main reasons:

1. Because it is more targeted so the traffic is more likely to convert (lowers your cost of customer acquisition)

2. Because there is less competition you'll pay less per click

The downside? Only that there may not be high volumes of traffic available that match your chosen phrases. That's more of a problem for brands going after metrics like reach and frequency, but less of an issue for smaller, niche players.

So all you need to do is choose your phrases and get going. You can use tracking tools with Google (assuming you can insert their code into your pages) to check on your conversions and return on investment too.

Personally I don't think skiers will be thinking hot tub as the key differentiator in what they are looking for... It's a nice/great to have but not what you'd search for...? I agree with your points about generic terms being hotly contested but combinations of them may not be e.g. "chalets in chamonix"? "skiing short breaks"? "privately owned chalets"? "DIY skiing holiday"? You get the idea...

Ashley makes some good suggestions. especially using the trial versio nor forking out a few bucks for a more detailed analysis at wordtracker.com. Do be sure to cross reference the results with those from say Overture (http://inventory.overture.com/d/searchinventory/suggestion/) , Espotting (http://www.espotting.com/popups/keywordgenbox.asp) plus of course Googles own keyword sugestion tools (https://adwords.google.com/select/main?cmd=KeywordSandbox). Once the brain storming is out of the way the best thing to do is really to give them a try. Run a campaign of what you regard given the suggestions from the tools for a week or two and then make an excel table of the conversion rates/traffic. Then change them completely to the next set of say 5 keywords. At the end you will find those that increase your bottom line which it is all about. It is not just about levels of traffic but ROI. Consider also 'buying' keywords such a 'reservation', 'booking(s)', 'cheap' etc.
I would not personally bring 'hot tub' into it. I really cant imagine anyone searching for it. You should include it in your description but not as a keyword.

Also consider terms that are relevant such as 'winter holiday', ski accomodation', 'winter breaks', '[town] accomodation'.
try and think linear. Not all will be accepted but you might get lucky. The key though is in the keyword research and testing. You never really know about roi and traffic unless you try things out. Also as already mention try and get the keyword sin the body text a few times. Not only for normal SEO but because Google editors are going to expect to see it.

Hope that helps.
Alan

> 'Chamonix' and 'chalet' will be most in demand and hence expensive.

I agree. They'll also have a poor click-through & conversion rate compared to highly targeted terms (& therefore give you a lower return on investment).

Your ROI is all that counts, NOT traffic.

I'll post some tips shortly...

Rusell

Russell comes up with some excellent actions and I totally agree that tracking results is the secret to success.

However, Google's system is a bit like an onion and often you need to peel it back to see what is really going on.

Let me explain.

There are 4 types of keywords you can have in Google Adwords :

1. Broad match - your ad is shown if the words you are bidding on show anywhere in your list.
2. Phrase match - your ad is shown if the phrase you are bidding on is shown within the context of the search made
3. Exact match - your ad is shown if the phrase you are bidding on is shown EXACTLY as the search made.
4. Negative match - your ad is NOT shown if one of the stop words you choose is used.

You could add in extended broad match, where similes will also generate traffic, but it gets a bit complicated at that point.

Now you know the matches, let's look at the strengths/weaknesses of each type.

1. Broad match -

Strength - allows you to be a lot less accurate with keyword selection. Find one phrase and get all the variations, e.g. your keyword is bank loan, your ad will also show for, cheap bank loan, bad credit bank loan, loan from the bank etc.. so lots of traffic for one phrase.
Weakness - you end up with a lot of visitors that are not what you want and the actual results make you believe you are doing OK, the CTR seems OK and you are making sales, but all those wasted clicks erode your potential results.

2. Phrase match -

Strength - allows you to refine what you are looking for more acurately. Avoids the jumbling effect, only searches with bank loan in that order would be shown. Avoids you receiving extended broad matches.
Weakness - Can significantly reduce traffic levels and if you have insufficient phrase matches you will get a significant drop in traffic levels.

3. Exact match -

Strength - allows you to appear higher in the listings as generally gives a better CTR as less "chaff".
Weakness - can eliminate clicks where the intention is obviously what you are selling/offering, due to extra characters in the search term.

4. Negative match -

Strength - allows you to omit obvious instances of irrelevance, e.g. you don't do anything for free, so you use -free as a negative keyword.
Weakness - We've seen plenty of people buy where they used free in their search, so what they really meant was "cheap/inexpensive/discount"

We use a tracking tool that enables us to track each click and where a referral string is presented we can pick that up and marry it back to the visitor and the actual word that was used.

So it may be we were bidding on the keyword "chamonix ski" and we were getting a lot of visitors where they actually typed chamonix ski chalet, my phrase match would pick it up, but I'd get much better results keeping the "chamonix ski" keyword but making [chamonix ski chalet] and exact match because it would get much better results and higher CTR and thereby a lower CPC.

The order for search is :

exact
phrase
broad
negative

So Google look for an exact match in your list (better relevance), if it finds one it shows it, if it doesn't it goes to the next level, (relevant but not quite so), if it finds one it shows it, if it doesn't it goes broad (less relevant but still slightly), and then the sanity check is to look if there is anything you don't want to be shown for (as that makes it less relevant to your offer).

Track the visitors and you'll be able to get great results, and don't assume that your campaign is working at optimum level just because Google's system says you are getting a 5% CTR.

Hope that helps.

Jim Banks
Web Diversity