Pay Per Click is my lifeblood, but as I’ve mentioned before I’m a dabbler.
I just found out one of my ‘active parked’ domains ranks #13 in google for a single word query. It’s not a top tier competitive term, but it’s no slouch either. It’s not even exact match. Never built a link or lifted a finger. Just applied a little dab of secret sauce.
I love this business…
Credit woes?
High gas prices?
Market off 10% since it’s last high?
Housing crisis?
Fuggedaboutit.
We are in the golden age right now folks. Maybe in the 4th quarter, but still in the golden age nonetheless. It won’t always be this way, so enjoy it now while it lasts.
By golden age I mean the growth curve is still relatively strong. It will flatten out at some point. Exactly when depends on who you ask.
Trends that will always continue, however, are those centered around holiday shopping, whether online or off. Performance will spike this time of year whether you are selling hot dogs, furnaces, Elmo dolls, or snow blowers.
On a micro level, I’m seeing some unbelievable performance in my campaigns this week. Here are some examples:
Sports Retailer - 2.5 yrs managing
Historical Average:
Conversion rate: 1.86%
Cost per sale: $22.84
This week:
Conversion rate: 3.42% (all time high)
Cost per sale: $12.42 (all time low)
Those numbers make me warm and fuzzy.
What surprises me, however, is an upward trend in my non-ecommerce campaigns. For example, a lead generation campaign for financial services is performing at alltime highs as well.
Historical Average:
Conversion rate: 3.95%
Cost per lead: $25.98
This week:
Conversion rate: 5.77% (all time high)
Cost per sale: $15.61 (all time low)
I really haven’t been able to come up with a reason. The service is completely unrelated to the holidays or shopping. Anyone care to guess? Seeing anything similar?
The rapid growth of the internet often shields us from the rest of the economy, but every once in a while you will see a change that breaks through that ‘bubble’. Today I received this email:
Dear Affiliate Partner,
In order to stay ahead of the current mortgage cycle and enable E-LOAN to maintain its leadership position in the industry, E-LOAN will begin scaling back some operations/programs to reduce its cost structure and unfortunately removing our affiliate program is one of them.
The lesson, if there is one… Don’t forget we still exist within the greater economy and are ultimately subject to it’s ups and downs. Also - don’t limit yourself to one niche. Diversify and protect yourself.
Unless you’re a turkey….
Question - How will you oversee campaigns while you are traveling?
A. Someone else will take over
B. I’ll check it periodically
C. Let it ride
No, I don’t want you to take that title literally, however, here’s the underlying hypothesis behind the title above:
If you can’t get your site to turn a positive ROI with the most competitive short tail terms in your niche, there is a very good chance that your problem isn’t KEYWORD related.
Meaning, if you can’t play the game with the best, then you will never scale to the major leagues. If you are in the insurance business and someone searches ‘INSURANCE QUOTE’, clicks on your ad, and doesn’t fill out your form, then don’t tell me that person wasn’t qualified. They told you what they wanted, you just didn’t give it to them.
Of course, there’s a lot of room in the land of mediocrity. Competitive webmasters the world over will continue to turn out profits off long tail terms as long as search engines exist. The problem is, the search landscape is not flattening out. It’s going the other way. You need to have the ability to convert with the best of them. You need to aspire to be as good as anyone out there. Complain all you want about the bid prices of broad match competitive keywords. Do you think Lending Tree is losing money on ‘mortgage’ broad matched? I doubt it. (look at these 3 ads by the way - they are TIGHT. look how much information they convey in such a small space)
My point is simply this:
If you are grinding out very little or negative ROI day after day and you can’t for the life of you figure out how to make it work then it’s time to stop and have your site look itself in the proverbial mirror. It’s time to humble up and look at why your site isn’t converting enough to make a decent net profit. If that doesn’t sound like an attractive course of action, don’t worry, you don’t have to do anything.
It’s ok to be mediocre. It’s part of our culture to blame everyone else for your current predicament. Let me help. Here are some things you can blame:
1. It’s Google’s fault. The bids are too high because they are greedy and bad.
2. It’s my competitors fault. They are bad marketers and bid too much for the best keywords. They can’t be making money, can they?
3. It’s the content networks fault. They send me bad traffic because they are greedy and bad. They should filter my traffic for me so I make the most amount of money without ever doing anything myself.
4. It’s my search agency’s fault. The targeted keyword traffic they are sending me isn’t converting.
5. It’s the industry’s fault. I spend hours everyday reading search engine marketing blogs and nothing is happening!
Think about it…what if I gave you these choices:
Scenario 1: You have a team of copywriters and programmers/designers at your disposal, but you can ONLY bid on the word ‘mortgage’ or ‘mortgage quote’ or ‘mortgage rate’.
OR
Scenario 2: You can bid only on phrases that get 10 or less queries per month and you can’t make any changes onsite.
The problem with scenario 2 is, long tail traffic is so disparate and unpredictable that it’s very difficult to optimize for. Meaning, query reinforcement, quality score optimization, etc.
Here are the 3 downsides to focusing on the long tail:
1. Hard to segment due to high percentage of unique queries
2. Hard to optimize for, due to the same reason
3. Low quality score, for the same reason
12 months from now, in order to really optimize your site for a long tail adwords campaign, you’ll need a landing page for every single nuanced search query. That level of customization is not practical for the majority of advertisers out there. Nor is it feasible for the search marketing agencies.
What is realistic then?
To focus heavily on your core competency at the root level. Take your biggest keyword, and optimize the bejusus out of your landing page. Make your landing page better than the biggest player in your industry. Focus on your message, your offer, you price, your product. Everything. If you can convert on that term, you can convert on anything. Think of it like the Sinatra song, ‘If I can make it there, I can make it, anywhere…New York’.
Why not try to hang with the best? Set your sights a little higher. If you can’t make it work, your site will still be better off than the guy who never tried.
Yesterday I was developing a proposal for a client entering their 3rd year with us. My bottom line with this client is that they have been successful, but they’ve never really taken any of my onsite advice in terms of building really good, solid landing pages that convert well. In fact, their landing pages are quite poor. I see this quite a bit. People spending tens of thousands of dollars on advertising, but not investing in the landing pages or website. Penny-wise and pound foolish.
I think what the layperson don’t grasp is how inexpensive it can be to have a really talented designer throw together a very nice looking page. I mean, give me $200 and a Sitepoint contest and I’ll have 20 different people begging you to pick their design by the end of the day. 20% of them will be really good designs. I think people are still scared off by those website design proposals they saw in 1997, when everyone was charging $25000 for a 5 page site with rotating gifs and a ‘dynamic’ contact us form.
Anyhow, with said junky landing page, here is a snapshot of 1 month a couple seasons ago:
The top line is actual data. The subsequent lines are the carrot I’m dangling for them. In line 2 you see that if we can increase the conversion rate a measly 4 tenths of 1%, they would have received 14 more leads (12% better) and reduced their cost per lead $8 (11% better).
As it was the following season, by virtue of some PPC kung fu, I knocked that cost per lead down 30%. Still, imagine the numbers with a really good landing page. Obviously, the campaign works as it is running. The numbers make sense for them. Average sale is in the thousands, so if they convert 1 in 5 leads generated from the PPC campaign they are still doing well.
My gut instinct is that if we had a good cache of landing pages I could get the cost per lead down to $12-15.
Anyhow, if you want to use this spreadsheet as a carrot for your clients, you can download it here.
My most challenging client:
For about 2.5 years I’ve worked with a B2B client in the technology market. Their product, for a non-network admin type like myself, was really abstract and difficult to comprehend. I understand what it is now, obviously, but during the first kick-off call I remember my eyes glazing over as they described their hardware, service, and benefits. Most of my responses we’re variations of the words:
Huh?
Oh.
Hmmm…
Interesting…
Sounds great..
Umm..
It took me 1 entire month to build their campaign.
1 month.
29 of those days were spent scratching at the deep dark depths of every keyword tool I could find. Begging for another keyword, synonym, anything at all that someone would search for that could possibly lead to a sale.
Splitting Hairs
On top of this, semantically this product could easily attract consumers and not network admins. This only compounded our issues. Remember the limitations of the content network years ago.
In short, the search network traffic was going to be minimal and we were going to be relying on the content network. I knew this from the start. Talk about a challenge..
The cherry on top? The sales cycle could be as long as 12 months. The kiss of internet marketing death.
Fast Forward
Anyhow, to make a long story longer - the client has received great industry praise for their service, have won numerous awards from their leading vertical pubs, and have had an IPO. Great for them.
All the while their little PPC campaign guy has been sitting in the corner generating leads and sending reports. Here’s where it gets interesting…
End to End B2B Tracking
From the very beginning we had to integrate with a high end, robust, but clunky 3rd party analytics tracking program. This required, much to my chagrin, a manually written custom tag on every keyword. The purpose of this tag was so the campaign, adgroup, keyword, and match type could be captured on site, passed through the lead form, and into their sales CRM backend. In short, every lead could be measured back to the PPC campaign by keyword.
Nice huh?
Well, every month for the last 2 years I’ve reported on everything I could, and have made steady progress on the cost per lead, even as we’ve expanded the campaign into 7 new countries.
All the while, no sales data, no matter how nicely I asked.
Out of the blue, about 2 months ago their new marketing analyst sent me some figures on lead to opportunity conversions. The numbers went something like this:
849 leads year to date. 21 opportunities. (No revenue numbers yet).
This came with the caveat that the revenue wasn’t yet attached to the report and we would be working on tying the 2 together. Still, I was enormously disappointed and was questioning all of the work I had done for them.
Finally, last month I received a follow-up email. What did they send me?
Data. Beautiful data.
Year to date:
1119 leads
91 verified opportunities
37 sales
$959,000 in revenue.
I can’t tell you our exact spend but it’s just a small fraction of this number. Much less than 10%. I’m happy again 
What are the lessons for B2B PPC advertisers?
Is this a post about the arbitrage opportunities in the construction equipment market?
Nope, this is just a post about the importance of Quality Control in paid search. Why backhoes? My 20 month old son, who loves backhoes and trucks more than life itself, is obsessed with anything and everything construction related.
While searching for christmas toys I decided to check out the backhoe market on Yahoo. The REAL backhoe market. Why would I do this? I have no idea.
I found a few more ads than I had expected…
One in particular stood out, unfortunately because their landing page was unreachable.
Granted, the backhoe search market is not one likely to run up a quick tab of $3,000/day, but I’m sure this advertiser has burned a few clicks today. Now I know we don’t all have 3 dedicated QC experts on hand at all times to make sure our ads are 100% kosher, but at the very least you should be checking ad creatives 1x week.
Here’s a few quick tips on how to QC your Adwords campaigns:
1. Campaign Management>Tools>Disapproved ads
This is the best feature. Your disapproved ads will show up here automatically. Usually there will be a reason associated with the error. You can fix your ad right here and it will automatically be resubmitted. Insider’s tip: Make sure you update your campaign notification settings to email you when an ad is disapproved. Like so..
2. Campaign Management>Tools>Ads Diagnostic tools
Here you just want to pick a few of your main keywords and run them through the query tool. Make sure your ads are displaying. Simple enough.
3. Manual Options
Lastly, another thing I like to do is go to the campaign summary page and scroll to the very bottom. (if you don’t have to scroll you clearly don’t have enough ad groups and you are leaving money on the table. click here for help)
Usually people will find 3 or 4 adgroups at the bottom that aren’t getting many impressions. Click on the adgroups that fit this description and try to rule out the following:
Bad Quality Score
Low Volume Terms
Low Volume Geo Restrictions
If you’ve ruled those out you need to look at your ad copy for misspellings. (see #1)
There are probably some techniques I’m leaving out, but if you follow these tips on a regular basis you’ll likely save yourself from any embarrassing mishaps.
Google Adwords has bestowed upon us Cost Per Click pricing for their site targeting options. And it’s not called site targeting any longer. It’s called placement targeting. Why? I don’t know. I think site targeting is pretty much an accurate term, no?
Oh well. Maybe it’s those $700 shares and $80 options that are confusing the Adwords marketing team. Everyone is trying to look busy until they vest and retire at age 26.
::::Unrelated mumbo jumbo:::: (if you’re busy move on)
Q4 is starting to feel like, well, Q4. I agreed to teach a paid search class at an advertising agency next Wednesday and am speaking at a conference in January. It’s 4 hours of me talking so I am just now realizing how much material I will have to prepare. As I compile all of my paid search knowledge (and not much else) that is knocking around in my head I’m realizing this is a nice refresher for me, to write these things out. I’m also working on my Pay Per Click Book that should launch next month. The two projects are mutually beneficial so I’ve got that going for me.
In addition to that I have 5 proposals due yesterday, 2 major campaign expansions, and I haven’t even finished invoicing from last month. On top of that, because I’m overly friendly to strangers I spent a total of 3 hours in the last 10 days giving free consulting advice on the phone. Or maybe I just enjoy hearing my own opinions. Argh! Any interns out there?? Seriously - if you want to learn from me while getting a heavy duty dose of the paid search business behind the scenes contact me.
What does that have to do with your paid search campaigns? Nothing, except to give you an idea of why paid search services are hard to scale as a business model. When was the last time you came across a huge, exclusively paid search company? They don’t exist for the most part. Because it’s specialized knowledge that is hard to transfer. Anyhow, more on that later.
Hasta.
I get this question often so I’m writing an answer that I can just point people to. Here are some possible reasons that your Google Analytics and Google Adwords reporting clicks have a major discrepancy:
These are 2 very different metrics. For example, a user clicks your adwords ad, visits your page, then goes back to their google search results page, doesn’t see anything else they like, and then clicks your ad again. Voila - 2 clicks, 1 visit (session).
Again, same scenario, multiple clicks on the URL, but 1 visitor.
Your site could be slow, or the Adwords visitor changes their mind, and before your analytics code renders, the user hits ‘back’ or ’stop’.
Google also addresses discrepancies between Adwords and other 3rd party tracking solutions:
We’ve found that most discrepancies between web logs or third-party tracking software and AdWords reports occur due to the fact that third-party tracking methods are unable to detect all the clicks that your ads receive. There are several reasons for this:
* Google Network statistics: Google displays ads on a growing network of search and content sites and products. Typically, web tracking software is not able to recognize clicks from Google Network sites as being affiliated with Google. These clicks are generally labeled only with the third-party site name. If your ads are currently, or have ever been, distributed to Google Network sites such as About.com, AOL, Netscape, etc., we recommend using referrer headers and tracking URLs to monitor where your website visits originate.
* Browser limitations: There are limitations to your third-party tracking software’s ability to track website visits with referrer headers. Most recent Internet browsers automatically pass along a referrer header when a user clicks your ad and is brought to your site. This referrer header indicates what site the user came from. For example, if a user sees your ad on a Google search result page, clicks it, and is brought to your site, the referrer header would most likely read ‘http://www.google.com/…’ However, some older versions of Internet Explorer and Netscape do not support this functionality and will not send this information. Also, some proxies and corporate firewalls strip out the referrer headers.
* JavaScript not enabled: If your third-party tracking software uses cookies to record referrer headers, ad clicks that occur in a browser without JavaScript enabled will not be tracked. Your AdWords account, on the other hand, will record clicks that occur in browsers with or without JavaScript enabled.
* Repeat visits: Users may click your ad several times in a relatively short period when, for instance, comparison shopping or conducting research. Your third-party tracking software may not count these repeat visits to your site. As long as the click patterns do not fit a profile of abuse or invalid activity, however, they will be counted in your AdWords account statistics.
* Shared IP address: Some Internet Service Providers (ISPs) assign the same IP address to multiple computers. Your AdWords account statistics will reflect multiple clicks that occur from the same IP address, while your third-party tracking software may not.
Google analytics will also differ from other 3rd party tracking solutions.
At the end of the day, all your tracking mechanisms will tell you a different story. My advice is to use a combination of relatively trustworthy solutions and have redundancies in place. Adwords tracking, Google analytics, and then something to act as an antidote to all that Google data - maybe awstats or a paid solution like indextools. Lastly, there are your trusty log files which can be tedious to review but relatively reliable.
