Selecting The Right Keywords: Your Input Needed

[Planet13]

Hi everyone:

My adwords campaigns ALWAYS fail miserably. :(

So I ask, what are the best practices for researching the RIGHT keywords?

What sorts of things do you look at BEFORE you spend a dime on advertising?

And if you research your competitors first (before choosing your keywords), what sorts of opportunities / warning signs do you look for FIRST before bidding on keywords?

Thanks in advance.

[netmeg]

Well first I look at analytics to see what people are coming in on now (both paid and organic)

The two biggest problems I see when I take over a client account is poor organization and keywords that are too broad. Start out with really precise keywords, even though you won't get as many searches, and build out (rather than going broad and working in)

[buckworks]

Please give us more detail about why you say your campaigns fail. Not enough impressions? Traffic too expensive? Clicks but no conversions? Etc etc. Where are the weak points - search or content?

One of the biggest things I try to figure out is "Is this a product that people are actually searching for, or do I need to figure out other ways to reach good eyeballs?"

[netmeg]

Yah, good point. If people are already searching for what you got, then search ads are a good choice.

But if people aren't searching because they don't know you (or your product) exists, then you probably need to go all Display Network on them to make them aware.

[alexsel]

I would recommend creating tightly themed campaigns with exact matched keywords.

If you want to grow the account and really find out what people are searching for the keyword tool is a good start but don't stop there. Add a few broad matched keywords or use the broad match modifier if you're on a tight budget. Then pull a search query report to see what keywords pertain to you the most. You can then apply exact matches to these keywords and bid more aggressively.

You won't know the RIGHT keywords until you start the campaign. Even with organic keywords data, that's just data from keywords that you rank for plus you don't know the cost associated with those keywords. They may not produce the ROI you're looking for.

The toughest thing for paid search advertisers is understanding that paid search positive ROI doesn't generally happen overnight, unless your product is unique and awesome. You'll have to be ready to lose a bit upfront in the effort to find the right solution for you. Continue testing, continue trying new things and be patient.

[Planet13]

Hi there, Netmeg and Buckworks:

Please give us more detail about why you say your campaigns fail.

Traffic too expensive?
Yes! Number 1 Reason why I say they fail. for example, it costs me $50 in clicks to sell a $50 product.

Not enough impressions?
Sometimes. When I try to get a very narrow focus with my keywords (so I don't break the bank on clicks that don't convert), then I only get like 10 impressions a week and only two clicks a week. Those clicks don't convert very well, either.

Clicks but no conversions?
Yup. When I try to broaden my impressions to get more clicks, then I get clicks with no conversions, hence the $50 in clicks for a $50 product, as I mentioned.

Because I am in California, and SOME of the items I sell are a bit heavy (about 6 to 15 pounds), I have also tried to limit my ads for heavier items to showing ONLY California visitors in the past, too. But that didn't work either. Again, very few impressions, and the impressions that I did get didn't generate any conversions.

You'll have to be ready to lose a bit upfront in the effort to find the right solution for you. Continue testing, continue trying new things and be patient.

Well, that's kind of the frustrating thing. I have been TRYING to learn from seeing what doesn't work, and I don't seem to learn anything from it. I guess I am learning a lot about what DOESN'T work, but not so much about what DOES work.

One Other Thing: what I have definitely USED to be a niche product, but more and more of the ebay's, amazon's and target's of the world have been moving in. Heck, even Donna Karen NY has sort of moved into the space, and since it is NOT a product related to fashion in any way, I was totally surprised by that.

I know I can't really compete in price on generic terms. For instance, there are lots of places that can sell widgets for less than I can, but the don't sell the same widgets I do because widgets is such a broad term and there are all sorts of styles.

The terms I usually try to bid on are similar to "sporting goods" or "french dresses" or something like that, but a LOT lower search volume. They are non branded terms, and there aren't any big brands that are by default associated with them (the way that nike or reebok might be associated with the keywords "running shoes")

thanks in advance.

[Planet13]

One of the biggest things I try to figure out is "Is this a product that people are actually searching for, or do I need to figure out other ways to reach good eyeballs?"

That brings up an important point.

The other companies (and they are small businesses, except for the amazons and ebays that bid on adwords), don't tend to use adwords much. I see them advertise more on journal-type sites (usually web sites that are built around a printed magazine that has been around for a while).

If I remember correctly, Netmeg has brought this up in other places before (sorry if I am paraphrasing), and that is that you should ask yourself, "Why should the customer buy from me?"

Unfortunately, I don't know if I have an answer to that question yet...

[netmeg]

You need to answer that first. Because if you don't know yourself why you deserve the sale, you sure can't convince the user.

[Planet13]

You need to answer that first. Because if you don't know yourself why you deserve the sale, you sure can't convince the user.

Thanks, netmeg. I know you have touched upon it before, but do you have a systematic approach for determining your value proposition?

And, if all else fails, are the people who work for google in the adwords department helpful at all when it comes to keyword suggestions? I know I get emails from google employees that say, "your adwords are not being displayed, contact Sheila to start getting more impressions."

Or are they just going to try and get you to bid on every keyword they can possibly think of?

[alexsel]

Or are they just going to try and get you to bid on every keyword they can possibly think of?

I wouldn't recommend relying on Google for keywords advice. Their goal and your goal isn't aligned.

[alexsel]

Also, why customers should purchase from you is beyond any AdWords advice I can give you. That's a business strategy question which we can't answer. Ad copy is obviously important in setting the right expectations, but if people are landing on your page and you see little to no conversions, AdWords may not be the problem.

[netmeg]

Thanks, netmeg. I know you have touched upon it before, but do you have a systematic approach for determining your value proposition?

Well hopefully I know it before I begin. You have to think about what people want (and if you don't know, then ask a bunch of people but don't ask people who do the same thing you do, ask people like your family and your non-industry friends and so forth, because they're probably a better reflection of your users) Then go look at your competition, and be honest with yourself and write down everything they're doing better than you in one column, and everywhere you think you could do something better in the other column, and then look at both columns. That should tell you pretty quickly. And to be *brutally* honest, if you can't figure out a way to do it better, then start looking around for another space where you can. Because otherwise you're just beating your head against the wall.

(and yes, this is beyond the scope of AdWords)

Google will offer you suggestions for keywords in their 'opportunities' section of AdWords. These are machine generated and know nothing about what your business is really about or who you are. But, I do still peruse them from time to time, because now and then they do come up with something I might want to try out. But you never just take them and dump them into an ad group, pick a few, create an ad group and write a killer ad specifically for those keywords alone.

[bwnbwn]

Looks like your bidding on generic terms that might be related to the widget but not what the user is looking for hence clicks but no sales.

I would pause the accounts build campaigns specific to each product. Bid exact with the product name. If your told not enough searches, then look for very related searches to the product. Not how much you spend but conversions. This will reduce the spend but increase conversions.

Continue to add specific related terms that apply to this campaign. Build a campaign for each product if there is enough differences that you can target different keywords. When you start showing a profit from the lower spend but targeted ads you can expand to see if broader keyword phrases will help generate more sales.

[Planet13]

Thanks again, Everyone:

@ alexsel:

I wouldn't recommend relying on Google for keywords advice. Their goal and your goal isn't aligned.

Thank you. I suspected as much.

@ netmeg:

Again, a million thanks. I can certainly write down everything and try to be honest. Just gets a little frustrating because I REALLY should have done this 6 years ago. So it's just depressing when the column with 'what they can do better" is full, and the column with "what I can do better" is blank...

@bwnbwn:

I would pause the accounts build campaigns specific to each product. Bid exact with the product name. If your told not enough searches, then look for very related searches to the product...

The one problem with that is there isn't really a product name, nor a specific manufacturer who dominates the industry. Imagine if someone were bidding on "koi fish." It is rather generic.

also, at this time, there isn't a retailer associated with it. So it's not like "books" where amazon or borders or barnes and nobel are almost immediately associated with it.

[netmeg]

So it's just depressing when the column with 'what they can do better" is full, and the column with "what I can do better" is blank...

Oh I know; I didn't come to this realization overnight. But it's also not going to get less depressing the longer you wait to do the exercise.

[Planet13]

Oh I know; I didn't come to this realization overnight. But it's also not going to get less depressing the longer you wait to do the exercise.

yeah, I'm working at it in small steps. Right now I am trying to muster up the courage to look at my weekly rankings reports from RavenTools. The last couple of weeks have not been so good.

I would like to have a cigarette and a shot of tequila on hand just in case there is more bad news...