It happens to the best of us. You work on an SEO campaign with a few carefully chosen keywords for months. But when you finally get to the top of the search engine results…nothing. The traffic you expected doesn’t come in or, even worse, neither does the money. You start to wonder, “What went wrong? Is it that people don’t like the search snippet? Are they finding what they want on the website?”
It’s perhaps the most frustrating thing that can happen to an SEO. But it’s also something you can often avoid completely with a little planning. In this post I’m going to talk about a technique I like for using pay-per-click first to test out my SEO game plan. This way the next time you make it to the top of the search rankings, the traffic and money will start pouring in!
Using the PPC guinea pigs
An important point before we start is to disable the Google “content network” and set up conversion tracking—that’s how we’ll ultimately identify the best keywords. Remember, the best keywords aren’t the ones with the highest number of searches or even the most clicks. They are the ones that bring in the most conversions at the lowest possible cost. Our PPC strategy is simple:
- Start with a large list of relevant keywords using Google AdWords keyword tool.
- Create very specific ad groups and landing pages you want to test.
- Split test everything: ad copy, keyword matching options, and landing pages.
Spend money to make money
Start with a large list of as many relevant keywords as possible. Divide your keywords into small groups so that each can be sent to an appropriate landing page. Set a low budget per day in AdWords so that you don’t spend anymore than necessary. Even if it turns out to be an expensive campaign, you can run it for a limited amount of time just for testing purposes. I think in the vast majority of cases, the ultimate payoff is worth the small investment.
There are some other points worth noting as you go about your testing. Keep in mind that you are finding the best candidates in multiple categories: keywords, ad text, and landing pages. Ad variations will eventually help you title your SEO landing pages and offer meta descriptions that appear in the search results and cause searchers to click on your link. You have to think of an ad as an offer. The searcher is always in the “What’s in it for me?” frame of mind. Your ad text must offer a promise of something specific to the searcher. Your landing page will have to deliver on that promise; if it doesn’t, you’re not going to get a conversion. You can also test the efficacy of different landing pages with Google’s Website Optimizer
Another often-overlooked point is to test variations on the same keyword phrase. AdWords offers you flexibility in terms of a broad match, exact match or phrase match. These are explained here in Google’s documentation. AdWords also has an advanced matching option called embedded match that I can use to help make my split testing more effective. It’s just as important to test these keyword matching variations as it is to test different ads.
You’ve seen the future, now go after it
With Google Analytics, you’ll be able to do a detailed assessment of your campaign when it’s over. You’ll see clearly the exact words people typed into the search that led to conversions. This will allow you to hone your keyword list even more. Based on your PPC conversions, you’ll also be able to estimate how much revenue you can potentially earn and whether your SEO efforts are going to be worth it for these keywords.
Before I leave you, I want to mention that this post was inspired by an answer I read in the LinkedIn forums written by Laura Alter from notebookforums.com. I want to thank her for that, and I hope you’ll let me know what you think of this strategy in the comments.
JDog
May 27, 2008 at 11:01 pm
This has to be one of the best posts I have seen from you in awhile. I know you have been busy with ranksense and I am glad you are back to sharing your knowledge. Keep up the good work.
Paul Montwill
May 28, 2008 at 12:16 am
As PPC is currently my key area, I would like to share my experience and write a few tips that should be a valuable addition to what you have already written, Hamlet. 1. I wouldn't recommend limiting a daily budget in the long term. If you want to stick to the budget, turn off the campaign on some days of the week. If you go for a maximum number of impressions per day, work on maximising your CTR and positions, your CPC will get lower. 2. In campaign settings, make sure delivery method is accelerated and Ad Serving rotate (you don't want the system to pick your best ads, you want to get rid off low performing ads yourself - after a few days/weeks of testing considering all the factors including CPC, conversion etc.). 3. Still in campaign settings, enable position preference. I was able to increase my average positions by even a few places while keeping my CPC at the same level. 4. I also turn off Search Network as they are not always as reliable as Google search. 5. Try aiming on the top 1-3 as they usually bring the highest conversions. 6. Control your Quality Score. Read Google help regarding this topic. Make sure your landing page is relevant. 7. I have recently improved CTR from %16 to %30 by adding one additional word in the displayed URL (one of the keywords). The higher CTR, the higher position and lower CPC (Google always leverage high performance ads/keywords). Test your ads, read, learn. 8. Use PPC stats/tracking to test keywords and use this knowledge for SEO. If a keyword is very popular and competitive in your industry, but doesn't bring high enough conversion in PPC, maybe it is worth focusing your SEO on something else. 9. Make sure you set-up locations and languages for all campaigns. 10. Consider switching off the weekends - especially if your business relies on customer service or direct contact and you are Mon-Fr business. Other reason might be that you have much less clicks than during weekdays and it results in the CPC and Cost per Conversions are higher (or not as low as during weekdays). 11. Benefit from Google Adwords tools, like Campaign Optimizer. Download Google AdWords Editor to see the whole picture of your campaigns and easily import/export ad groups. Use different Reports and set up automatic generation. 12. Use Online Commercial Intention tool from AdCentre if you are in e-commerce. You can have better predictions in terms of conversion rates: <a href="http://adlab.msn.com/Online-Commercial-Intention/oci.aspx" rel="nofollow">http://adlab.msn.com/Online-Commercial-Intention/...</a> (Query) 13. It is not only Google. From my experience the lowest Cost per Conversion comes from MSN first, then Google, then Yahoo. Both MSN and Yahoo platforms are much more difficult to use than Google, but it is worth to have a try. I have read similar comments regarding MSN conversion rate. 14. Learn from your competitors when it comes to ads. They make a lot of mistakes (especially when you type in a keyword and not a single one is included in the text ad) but also may provide you with a lot of valuable clues.
Hamlet Batista
May 28, 2008 at 7:39 am
Thanks for your comment, JDog. Paul - Thanks a lot for your input. Those are very useful tips. I tried to avoid getting into too many details about PPC because the purpose of the post was to explain why SEOs should use it to improve their keyword targeting. Sadly, most SEOs hate the word PPC. There is definitely a lot to learn about PPC. I'm looking forward to Dan Thies new free book PPC fast start. Check it out when he releases it at <a href="http://www.seofaststart.com" rel="nofollow">www.seofaststart.com</a>
Andy Beard
May 28, 2008 at 10:26 am
I think Dan would say forget about CTR and CPC as they don't matter, just focus on conversions and $$$
Scott
May 28, 2008 at 2:14 pm
Not really a comment on PPC, but adsense. I trawled last week through all the websites we host / own and found 30 odd sites that no longer had an owner, business defunct etc. I decided to install adsense on all of them as they all have some decent traffic. I installed the adsense code on Tuesday, and blow me over, my adsense account is at $232 today.
Paul Montwill
May 28, 2008 at 11:23 pm
Sure, Hamlet. I thought that some people may be hungry for more tips :)
Paul Montwill
May 29, 2008 at 12:32 am
Hamlet, it is a shame that so many SEOs hate the word PPC. SEO & PPC are usually complementary and on high competitive keywords where three Adwords ads appears at the top of the search result, sometimes being strong in organic is not good enough. On the other hand, people think SEO is free and that is why it is better than PPC. It is not. Sometimes it takes a lot of effort to get to the first page in Google and we shouldn't really consider our time on link building as free.
Hamlet Batista
May 29, 2008 at 10:24 am
Hi Andy, You are right on the money. Paul - That is correct. SEO is not really free. It is more cost effective long term, but it is also more risky and you don't have the kind of targeting power you have on PPC.
Jon Dunn
June 2, 2008 at 12:47 pm
Hey nice post, and nice idea about testing conversion rates for keywords. However, With a bit of good keyword research "The traffic you expected doesn’t come in" this should never happen. Also, excellent PPC tips from Paul Montwill up there! :)
Levon
June 3, 2008 at 1:49 pm
I guess I should start doing this sort of stuff, but it doesn't fit into my strategy. I've had one real stinker before. Having it rank well didn't actually make a difference to sales. My experience is that the rankings will generally bring you enough to justify the costs, although my costs are less than most SEOs. If anyone would like a tip, I find that the best niches are anything to do with home and garden. This surprised me, but they always seem to do well. Another tip is to avoid male beauty products :P
Rex
June 6, 2008 at 7:26 am
My comment is not completely topical but I think your readers will find it interesting. I just ran across an article entitled: "How To Whip The Competition In Organic Search Results"at a UK site called "Connected Internet." The author claims to have increased his organic search traffic by a factor of three with one simple change: using all caps in the Title and Meta Tags. I haven't tried it yet but I'll probably start experimenting with it this afternoon. Here's the link to the article: <a href="http://www.connectedinternet.co.uk/2008/04/22/how-to-whip-the-competition-in-organic-search-results/" rel="nofollow">http://www.connectedinternet.co.uk/2008/04/22/how...</a>