Online Marketing Blog

SEO Vs PPC – Search Marketing Is Not A Fight! (Another Mini Case Study)

Many Pay Per Click (PPC) management agencies will have you believe that the paid search model for your online marketing campaign is the only way to go. In my previous post on how Pay Per Click (PPC) and Search Engine Optimisation (SEO) should work together rather than as separate entities, I outlined a few reasons how PPC alone can effect your bottom line and have you and your company leaving money on the table. Today I am going to show you an example of how some companies are literally wasting their money by only concentrating on paid search.

We have kindly been given permission from one of our clients to present the following data. Our client is the Carnegie Court Dublin Hotel. The Carnegie Court came to us a while back with a pretty standard request. They wanted to promote the hotel online in the most cost effective manner maintaining a positive ROI. The Carnegie Court had and still has affiliates, however they wanted to generate more bookings through their own website without having to pay any affiliate commissions. RedFly Marketing got to work.

As part of their overall online marketing strategy, we created a Google adwords Pay Per Click campaign to gather conversion data and raw keyword data. With this data we were able to optimize their site for literally hundreds of long tail keywords that converted relatively easy.

For example, we knew from our Google Adwords PPC data that the search term “accommodation swords dublin hotel” although having low volume, converted at 80% over a six month period. Using some simple SEO methods we were able to quickly and easily optimize the site for a number one position in the organic results for that term. That’s great, but as I mentioned, that search term as you can imagine has a very low volume. Do that one hundred times for search terms that you KNOW convert from testing with PPC and you have something pretty amazing. This is the idea behind long tail search engine optimisation.

Next Up -Volume.

Number One

Now that a lot of the hard work for the long tail search terms has been done (It’s actually a never ending process throughout the length of a campaign), we can move on to optimizing for higher volume search terms. After a few months of Organic Search Engine Optimisation we were able to get the site to a number one position for the MUCH higher volume search term “Dublin Hotel”. Organic search results are clicked on up to nine times more than paid results. Because of the volume and conversions the organic results we achieved, the hotel is constantly booked out and they have no need to use paid search any longer. Why pay for leads if you are at your capacity anyway? If however there is a lull, we can simply turn the PPC campaign back on and reach a wider audience through the Google network of sites.

So What Is the Point In This?

Today I decided to do some testing seeing it has been quite a while since we paused the Hotel campaign. I set up a new campaign with the hotel account managed by us to test out the price the competition were paying for the sponsored listings.

Paid Search Bid

With our previous Pay Per Click management on their campaign, we were able to optimize their quality score and get it to the best possible. In this case, the illusive one cent minimum bid. Next, I set the max bid to €30 a click to insure we were placed at the top position of the paid listings. I created an ad and set it live.

Paid Ad

Unfortunately, the €30 bid was not even enough to obtain a paid search spot above the organic/free results. Because the keyword had the best possible score and we had a high max CPC bid, we can safely assume that the ads ranking higher (Above the organic free results) were paying upward of €30 a click. That’s assuming that they too had an excellent quality score. For the purpose of this example, let’s assume it was a nice round €30 a click they too were paying.

Now, looking at the average stats of the Hotel, I can tell you that the hotel receives:

  1. An average of 290 visitors a week or 1160 visitors a month from the search term “Dublin Hotel” alone.
  2. Assuming that the the organic result is clicked nine times more than the paid result, we can assume that the paid ad is CLICKED about 128 times a month.
  3. Again, assuming the paid listing is costing the advertiser €30 a click, those advertisers are spending €3840 a month for that search term alone!

All that while our hotel sits back and enjoys more visits, bookings and ROI from their initial investment in search engine optimisation. Again, this is only ONE search term.

Now I am not saying that paid search is a waste of money. Done correctly it is not. What I am saying is that PPC and SEO should be used to help each other as part of an overall search engine marketing campaign. You want to use your advertising spend wisely so that you can explore other online advertising avenues further down the line.

I hope this has shown the importance of using PPC data for organic SEO and also that paid search is only one part of the overall long term internet marketing strategy. If you place your entire online marketing budget into paid search, I sincerly urge you to think again. Think long term.


  1. avatar
    3rigena Says:
    April 24th, 2007 at 1:18 am

    An quality, eye-opening article Dave.. keep them coming!

    Reply

  2. avatar
    Georgie Says:
    April 24th, 2007 at 1:22 am

    Wow, great post. I really need to play with adwords

    Reply

  3. avatar
    Arnold Majlath Says:
    April 24th, 2007 at 1:34 am

    Another great post Dave.I honestly thought that the minimum bid regardless on adwords was 5 cents.I never sawa one cent bid before.Great article.

    Reply

  4. avatar
    Dave Davis Says:
    April 24th, 2007 at 1:35 am

    You guys are up way too late. :) I take it you are not recovering from Barcamp Dublin either?

    Thanks Arnold. Yes, 1 cent is still the minimum bid. The next blog post is actually about that. Stay tuned!

    Reply

  5. avatar
    Daniel Jepsen Says:
    April 24th, 2007 at 8:57 am

    Nice post Dave.. Keep up the good work!

    Reply

  6. avatar
    BlueBobbo Says:
    April 24th, 2007 at 8:57 am

    Thanks for the reply to my link Dave!

    You’ve already been in my RSS reader ;-), all I have to say is keep bringing on the awesome posts… You add tons of value to the industry–priceless.

    Reply

  7. avatar
    CustardMite Says:
    April 24th, 2007 at 10:18 am

    Have to agree with everything you’ve said there.

    We recently took over a campaign that had previously been run by another agency, and over 90% of the traffic that they’d generated came from terms where they topped the natural search results, and 80% from terms on which they were the only bidder as well.

    As a result, they were wasting the majority of their PPC spend on traffic that they’d probably have got for free.

    We’ve redirected their cash into terms that they aren’t high on the search results for, and they’ve seen a significant increase in traffic to their site.

    On the other hand, as our agency does SEO and PPC, I’ve had to stop bidding on many of the best terms in some campaigns, as the websites are optimised for them. Great for the companies in question, great for the SEO people, not so great on the weekly reports I produce!

    Reply

  8. avatar
    Cormac Moylan Says:
    April 24th, 2007 at 11:00 am

    Dave, it’s good to know that your RSS feed has subscribers! The swarm affect. :)
    I’ll digest your post later this evening.

    Reply

  9. avatar
    Jeroen de Koning | J8 Zoekmachine Optimalisatie Says:
    April 24th, 2007 at 6:01 pm

    Interesting case. Thans for sharing!

    Reply

  10. avatar
    Iain Says:
    April 24th, 2007 at 6:12 pm

    Yes, another very well written and informative post. I too am a subscriber :0)

    Reply

  11. avatar
    Dave Davis Says:
    April 24th, 2007 at 6:42 pm

    @Daniel & Bobbo – Thanks

    @CustardMite – I hear you. A large chunk of my time is spent going through and managing reports. In fairness, Adwords and Analytics offers some pretty decent number crunching options. I don’t fully agree about removing a paid listing in all cases of a number one organic position however. In this example, the business was al full capacity and didn’t NEED the added cost of more bookings.

    If you have an unlimited inventory (Software for example) or a massive inventory, I still think the more screen real estate you occupy the better. As I have mentioned before, as long as the visitor has a value higher than what you pay for the paid listing, it’s still a profit. I’d never take the risk of “Hoping” a visitor clicks on the number one ogranic listing because as I am sure you know, sometimes a user will not click it. It depends on the meta description or snippets call to action.

    Reply

  12. avatar
    Dave Davis Says:
    April 24th, 2007 at 6:44 pm

    Ian, thanks for stopping by. I absolutely LOVE your website. It is one of the cleanest, most usable websites in the industry in my opinion. You have whitespace usage down to an art!

    Reply

  13. avatar
    Jim Rucker Says:
    April 25th, 2007 at 5:54 am

    Very nice piece. I’ve been very surprised at how much PPC is affecting my traffic. I pulled back on it because one of my sites was placing well in the natural search, but without the PPC, my clicks and conversions have dropped.

    Reply

  14. avatar
    Dave Davis Says:
    April 25th, 2007 at 6:22 am

    Jim, your site is cannot be found in the organic results for EVERY search term you want it to be found for. The point of the article was to use PPC and find out EXACTLY what search terms convert best, then optimize for those terms.

    Also, with all due respect, you have a nice aged domain but a terrible landing page. Have you run any tests on the usability and conversion rate of the usedcarsonly site? I mean, you don’t even have a “Real” title or meta information on the page.

    Reply

  15. avatar
    CustardMite Says:
    April 25th, 2007 at 1:42 pm

    Hi Dave,

    Absolutely agree that there are circumstances where you would want both – but in this case, it was variations of the brand name, and nobody else was bidding on the terms, so people were almost certainly going to click on the free listing had the paid listing not been there – splitting the cost amongst the additional traffic would have led to very expensive clicks in this case.

    But those clicks would have been lost to competitors, had the term been something else, and there been others bidding on the term.

    I try to think in terms of whether the additional traffic is worth the total spend – and that’s something that has to be evaluated on a case by case basis, as you can’t really tell with any certainty how many clicks are being cannibalised – it’ll depend on what other links are on the site, the quality of the other adverts, etc etc…

    But there’s no doubt at all in my mind that it’s certainly something that people should be thinking about, and agencies that deal exclusively in PPC have no real incentive to ask the question…

    Reply

  16. avatar
    Dave Davis Says:
    April 25th, 2007 at 5:48 pm

    Some good points CustardMite. In the case of no other paid results being shown, that’s fine. However a broad match campaign could generate some additional trafic on never before searched for term combinations.

    I agree, agencies that deal only in PPC have absolutely no incentive to deal with this. In fact, it is in their best interests not to deal with it. That’s why it’s important to deal with an agency that is not only proficient in both but also works with the client to suggest additional work for the SEO team (Whoever that may be). It is also important for agencies to offer a performance based cap and important for clients to ask for it.

    Reply

  17. avatar
    CustardMite Says:
    April 30th, 2007 at 3:01 pm

    Not sure that I’m that keen on performance-based charging.

    From my perspective, I’d feel more cautious about trying new ideas, if I know that they’re going to cost my business money if they don’t work (I’m happy enough to take risks with other people’s money!).

    Also, from the client’s perspective, if we start performing far better than they were expecting, they may start suspecting our original estimates and forecasts. And let’s face it, if it’s a new campaign starting from scratch, they are a bit hit-and-miss to begin with.

    Unscrupulous agencies could deliberately overstate the expected cost per click (or conversion) in order to get more money from their clients. We use a fixed fee, and as somebody who used to be a client, I’d be more comfortable with this, particularly if I only had a restrictive budget.

    It’s always possible to ‘cheat’ to improve a statistic in isolation – cpc or cpa can be fiddled by reducing the bids, and getting less traffic but more cheaply. Clicks can be inflated with less relevant keywords or vague adverts…

    Reply

  18. avatar
    Dave Davis Says:
    May 2nd, 2007 at 5:46 pm

    Of course isolated metrics can be manipulated but what client hires a SEM firm to increase a single metric? The only one that matters is ROI or ROAS.

    We use a very common hybrid pricing model for larger accounts. That being a small % of advertising spend increasing at set milestone ROI increments. This has worked amazingly well as it reduces the risk for the client initially and as I said, incentivises the agency to achieve goals rather than just “Monitor” the account until the end of the contract. This model also insures that the client is NEVER experiencing a ROI smaller than what they had before they hired the agency. Even WITH the agency fees.

    Inflating statistics serves neither client or agency and any agency basing their fees or performance on this should be…. well, shut down. It is important for clients to use due diligence and understand this concept.

    Regarding trying new ideas with risk to your own agency revenue, that is completely up to you of course. But we all know that those that do not innovate get left behind. Also, if you try something completely new and outside the box and it works well, you can apply the technique to other clients campaigns. If you are working on a performance basis again, everyone wins.

    Reply

  19. avatar
    CustardMite Says:
    May 4th, 2007 at 10:12 am

    I think the difference is in that my Agency are quite new, and in the past the majority of our clients have been quite small. Frequently, they have had no prior history on Adwords, or what they are expecting or willing to accept in terms of performance – in some instances not even knowing what their breakeven CPA would be (in the case of new companies).

    This is changing over time, and perhaps we should consider offering different types of contract to different customers, though it’s not a decision for me to make.

    I think that flexibility is probably important in our position, where we have clients with budgets ranging from £10 per day up, and if our competitors offer alternatives that we don’t (or vice-versa), it’s quite important…

    Reply

  20. avatar
    Dave Davis Says:
    May 4th, 2007 at 3:31 pm

    Absolutely flexibility is important. But as too is showing results and having the motivation to show results on the agencies part. We have clients from the absolute minimum level to some pretty top tier spenders.

    It also depends on their exact requirements. For example, a client may come specifically for search engine optimisation and not require PPC management, or already have a firm grip on their SEO and need some prefessional pay per click campaign management. Then there are those that require a full service online marketing campaign encapsulating everything. Sometimes, all they need is a little advice to send them on the right path.

    Search marketing is a service, not a product. And it needs to be treated that way. It needs to be priced differently for each client according to their exact needs and goals and to insure they get the best possible return on their inventment.

    Experimenting with pricing at the end of the day is really the only way you can know for sure though. Those who entered the search engine marketing industry early have this to their advantage.

    Reply

  21. avatar
    CustardMite Says:
    May 4th, 2007 at 3:46 pm

    At the moment, we send out a weekly report showing the trend of the major metrics over the last four weeks overall, and by adgroup. We also send out a monthly report with a more complete analysis of what’s been happening, what we’ve been doing, and what we’re planning on doing – I know that when I was on the other side of the fence, working for a retailer, this was what I wanted.

    On the plus side, it means that there’s really nowhere to hide – if we drop the ball, or even fail to make improvements, they know almost as soon as we do!

    Of course, there are clear downsides to this – random (for want of a better word) noise can lead to some interesting phone calls, particularly if my contact isn’t particularly numerate. And it limits the number of adgroups that I can have without losing reporting functionality.

    On the whole, it’s a good thing, as it means that we have to keep working on their campaigns (it’s quite easy to spend 99% of your time working on the new campaigns, where improvements are easy to come by), and it means that our clients know that we’re giving them something for their money.

    Reply

  22. avatar
    Mike McMinn Says:
    May 18th, 2007 at 1:10 pm

    Nice post, we also fell foul of this during us first two months on adwords promoting PDA and mobile phone devices. We got to spending £450 a week with little return so we’ve halted the campaigns until we could focus more on specific keywords that convert. Of course our strategy has now shifted toward organic SEO. Great work please keep it up.

    Reply

  23. avatar
    Dave Davis Says:
    May 31st, 2007 at 3:15 pm

    You’re welcome Mike. I suggest not only moving focus to organic but using them both as a one two punch approach.

    Reply

  24. avatar
    CustardMite Says:
    May 31st, 2007 at 3:22 pm

    I agree – there’s no way (?) to appear at the top of natural search for hundreds or thousands of search terms – but if you can get the more generic ones, and pick up the low-traffic terms with PPC, you can get the best of both worlds…

    Reply

  25. avatar
    Dave Davis Says:
    May 31st, 2007 at 3:24 pm

    Well, with enough work you can rank number one for a hundred search terms if they are “Long Tail Enough” :) But it’s a lot easier to rank for the less generic terms and pay for the more generic terms.

    Reply

  26. avatar
    CustardMite Says:
    May 31st, 2007 at 3:32 pm

    On the other hand, you’ll generally get better PPC results for the less generic terms, and more impressions (and hence more savings) on the more generic terms…

    But, as you say, trying to rank number one naturally for ‘mobile phones’ or ‘PDAs’ could be ‘challenging’…

    Reply

  27. avatar
    Dan P. Says:
    August 1st, 2007 at 10:13 am

    A common reason why an ad may not appear above the organic listings… the “ribbon ads” is that the ad has not yet been manually reviewed… ads MUST be manually reviewed in order to be eligible to appear on Google partner sites… as well as these “ribbon ads”.

    Reply

  28. avatar
    Universos Virtuais Says:
    August 4th, 2007 at 11:27 pm

    Thank you for the nice tips Dave. I am trying to learn about CEO, but in portuguese (i´m portuguese) there isn´t almost nothing about CEO, so, i´m glad i find you blog.
    Have a great weekend!

    Reply

  29. avatar
    Dave Davis Says:
    September 13th, 2007 at 2:01 pm

    @Dan P, Ads do not need to be manually reviewed to appear as sponsored listings above the organic results (top spot as apposed to right hand column).

    You are right about the partner sites though. Also, ads must be reviewed manually before going live on the content network (for obvious reasons).

    Reply

  30. avatar
    Dan Perach Says:
    September 13th, 2007 at 7:20 pm

    Dave, I just double checked on the top placements ads… i was just now told that they do indeed to first be manually reviewed before being eligible to appear.

    Reply

  31. avatar
    Dave Davis Says:
    September 13th, 2007 at 7:22 pm

    Hi Dan, I have just started a new campaign for a client. Brand new, new keywords, new adgroup, new URL and new ad text. Within minutes it was displaying in the number one position above the organic results.

    Who told you it needs to be approved?

    Reply

  32. avatar
    Essex Web Site Designer Says:
    September 20th, 2007 at 10:55 pm

    Dan and Dave, it most definitely does not need to be approved because I also (not so long ago) ran a brand new campaign, for a new online client and had it above the organic results in no time.

    Unless something has changed?

    Reply

  33. avatar
    Dan Perach Says:
    September 21st, 2007 at 7:25 am

    It is possible that your ads were manually reviewed very fast, which would explain this. I have verified this issue 3 times, so I am 99% confident on this.

    Reply

  34. avatar
    Dave Davis Says:
    September 21st, 2007 at 7:33 am

    Although AdWordsAdvisor did not comment, please have a read of this thread on Webmasterworld:

    http://www.webmasterworld.com/google_adwords/3383118.htm

    Reply

  35. avatar
    Marc Says:
    September 21st, 2007 at 7:59 am

    Very interesting thread Dave… looks like this has been discussed elsewhere and that Dan’s content/keywords may have been flagged by Google.

    This makes sense to me, as I am pretty sure they follow a similar approach to reviewing websites, generally automated/spidered and then when something weird or a flagged item pops up they go in and manually review it.

    Reply

  36. avatar
    James Farrell Says:
    October 4th, 2007 at 11:33 am

    Hey Dave, thanks for the interesting article

    I find it interesting that you can rank higher for “dublin hotel” than “hotel dublin”

    I didn’t realise that this makes a difference before now.

    Overture doesn’t seem to discriminate on which is searched for more.

    Did you consider this idea when you ran this campaign?

    Reply

  37. avatar
    Mick Wright Says:
    January 9th, 2008 at 11:36 am

    Dave,

    Interesting enough… the problem with long tails is though that the time you spend on research trawling through the incoming and working out conversions only gives the client a tiny proportion of sales. In the case of this example I got my client right under the hotel for the NAME search… which surely is more important? no?

    Reply

  38. avatar
    Dan - PPC Pro Says:
    July 22nd, 2008 at 6:24 pm

    Dave, thanks for that url. Ok, seems that their is automation involved in the review process, and a trust factor involved. Drinking coffee is preferable to manually reviewing all new ads, very true indeed.

    Reply

  39. avatar
    PPC Bid Management Says:
    August 6th, 2008 at 2:27 pm

    Many people view SEO and PPC as 2 sides of coin when in fact we can use them both at the same time. And the best way of doing this is to bid on the same keywords you rank highly for and send PPC traffic to a review’ site. Having traffic sent to your main site through a ‘neutral review’ site greatly boosts sales.

    Reply

  40. avatar
    milan Says:
    June 2nd, 2009 at 4:49 pm

    Dave,

    I’m new to the adwords thing, and I think they’ re fooling me. the adcampaign metrics are at 0 impressions and all 0, yet my domain host web-statistics give me 40 hits a day…

    Reply

  41. avatar
    SEO Salisbury Says:
    August 26th, 2009 at 4:03 pm

    Hi Dave,
    Fab post, I totally agree that using SEO and PPC to work together is the best way to secure long term results. Running a PPC campaign for a few weeks, months if necessary gives us webmasters a good solid idea of what keyword/phrases work and apply them to the SEO campaigns. Thanks very much for sharing. :)

    Reply

  42. avatar
    IDSTC Says:
    November 30th, 2009 at 8:48 am

    Hey Dave, thanks for the interesting article.I was just now told that they do indeed to first be manually reviewed before being eligible to appear.

    Reply

  43. avatar
    diseño web Says:
    January 30th, 2010 at 8:06 pm

    Both they are very necessary and they must go in line together like part of a marketing strategy. At the outset he is recommendable to use PPC to obtain immediate results, while the positioning in finders works.

    Reply

  44. avatar
    steel supplier Says:
    March 2nd, 2010 at 1:22 pm

    Great post. PPC data is gold dust for SEO in terms of working out what search times convert and which ones really work in terms of ROI. I would always recomend running a couple of months worth of PPC with a large set of keywords to get as much data as you can.

    It is always worth looking at the intent of the search in terms of ROI. Is it an information search of one with intent to buy. It there any confusion in the terms shingles or shutters mean different things to different searches for example. Get your keyword research correct and use your PPC data.

    Reply

  45. avatar
    Daryl James Says:
    March 3rd, 2010 at 9:24 pm

    Thanks for the case studies. SEO and PPC are simple yet intensive concepts and things like case studies help newbies like me to better absorb the information. I’d love to see more such articles on your site.

    Reply

  46. avatar
    business IT support Says:
    February 25th, 2011 at 10:33 am

    I am just starting to dip my toe into PPC and SEO and this is really useful. I guess its worth running a large set of keywords with a big enough budget to start with so that you can track which keywords are sending the best traffic and then feed this into your SEO plan. I hadn’t thought of that so thanks for sharing.

    Reply

  47. avatar
    Flash Says:
    May 13th, 2011 at 1:30 pm

    On the other hand, you’ll generally get better PPC results for the less generic terms, and more impressions (and hence more savings) on the more generic terms…

    But, as you say, trying to rank number one naturally for ‘mobile phones’ or ‘PDAs’ could be ‘challenging’…

    Reply

  48. avatar
    Ünlülerin Güzellik Sırları Says:
    September 4th, 2011 at 2:14 pm

    Another great post Dave.I honestly thought that the minimum bid regardless on adwords was 5 cents.I never sawa one cent bid before.Great article.

    Reply