Back from a most yummy lunch and it’s time to talk about some amazing PPC tactics. This session is power-packed with awesome with Matt Van Wagner moderating speakers Addie Conner, Ryan Lash, Dan Soha, and Dan Thies.
My witty banter has been replaced with a full tummy (cannolis at lunch!), so let’s just skip to it. And by “skip” I mean “I can haz nap now”? Please? No. Jerks.
Matt says this session is modeled after the Give It Up sessions of SMXs past where panelists were challenged to give the most leading information – whether or not it’s “legal”. So, um, enjoy.
Up first is Dan Thies.
There are many ways to suck. Time sucks. Space sucks. Gravity sucks (but that’s an old joke). Your profits may suck too. Let’s start with time!
Day parting and DOW Effects
Some hours of the day are more profitable than others. Some days of the week are too. Experts warn that day parting is really “advanced” but it’s just 5th grade math. He shows us a graph from Google Analytics showing conversion rates by hour. The order value differs based on the time of day. He says:
Revenue Per Visit = Conversion Rate (X) Average Order
If you do leads, assign a value.
Figure out where you break even. Assume you make a 30 percent profit. Then 30% x 2.03 = 61 cents per click. Breaking even may be okay for some, but it may not be for you. Find where makes you happy.
Google lets you control days and times and networks. Try a separate Night Owl special to reclaim profitable night hours and get the profits you’re missing. You can literally slice this geographically by time zone so you that know exactly when it’s midnight. Instead of turning off unprofitable terms at night, try to find what you’re not getting and go after them.
Day Of The Week Effects: Different ad copy can perform differently by time of day and day of week. He recommends separate campaigns for weekdays and weekends. This can break ties between ad variations into simple truths.
Time Sucks. Space Sucks.
The book says that the sweet spot for lowest CPC is the #5-8 positions on the right side. That may not be right for you. You have to find what your industry “sweet spot” is. Don’t trust eye tracking either. You can prove anything with science. Do math.
Some markets are different but test at the top. If you win at the top you get more sales, more traffic, and more profits. You also get market dominance! You want to understand how Google’s smart pricing works. It’s called a second price auction. There’s an auction where the winner plays what the second place person bid. Or something. I may have ruined that. I’m not so good at PPC. And my brain is melting. I can feel it.
Locking Phasers: Match Types
If you can’t make a profit when you know EXACT-ly what the searcher typed, good luck with broad match. Phrase is underused. He gives away a free SEO book on his Web site but search queries like [free seo book] as an exact match have an extremely low volume. And most of them are looking for Aaron Wall (we’re all looking for our own Aaron Wall, aren’t we?). Instead, he does a phrase match for [free seo book] and it brings much more search volume.
Did you know you can make multiple copies of any ad that you want to? Because it seems a lot of people don’t. You can. Try it.
Mind Meld with the Market
Clicks are not money. Just because you get more clicks doesn’t mean you’ll make more money. He talks about the MSN AdCenter Labs OCI Tool and say its really cool. You can find out the likelihood that someone has commercial intent.
Scanning the System
Ad Creative and Landing Page are combined in the database so they’ll get the same landing page quality scores. You can use that to scan pages to get their relative quality score. You can also scan it for ad creatives. The payoff can be big. He shows some actual results where the company makes more profit from AdWords than they do being number one in organic results. Nice.
Ryan Lash is up next.
Your paid search campaign isn’t performing as well as it used to. You need to fix it. Your boss wants answers. They want to know how you can grow the campaign. How you can turn this around and what’s the solution?
If you’re running Google, you can run the Google Conversion Optimizer, you can implement a third party system, or you can optimize it yourself. The problem is you don’t have a lot of time.
What would MacGyver/SearchGuyver do?
Commonly used office tools
- Web Browser
- Spreadsheet
- AdWords Editor
- Items you’ve picked up – 3rd math and your analytical mind.
What you want to do is fire up browser, download your Google reports, open your spreadsheet, do some math, look at the analytics, drop it in the editor and you got your new stuff. [Is that all?]
Rules of Engagement
Calculating Ideal CPC: Look at existing costs by keyword. Your idea cost is the (conversions*Target Cost Per Conversion). Your Ideal CPC is the Ideal Costs/Clicks. Your CPC sweet spot is closer to or effectively half net margin. They’re based on actual performance.
Ryan is throwing out a whole bunch of numbers. My brain has officially kicked me for hurting it.
In terms of CPC floors and ceiling, its easy to accidentally jack the bid too high. What you want to do is figure out the max CPC so you don’t jack something up too high. You want to take your Cost Per Action and multiply it by the Average Conversion rate.
Additional Considerations
- Avg position
- Avg CPC
- Conversion Volume
- Not increase bids if the avg position is $1 or if the avg CPC is less than the max CPC.
Ryan’s presentation was totally over my head. He said he’ll be posting the slides up on his blog, so look for those. kthxisux.
Next up is Dan Soha.
Headlines: Putting your headline in all caps will get you a higher click through rate but it’s also not legal. Google may catch you but it’s worth it even if its just for a short period of time. Just saying.
Rules about Yahoo Ad Group: The quality index is at the ad group level. You benefit from having highly targeted ad groups. Keywords should be similar to each other, as well as the ad. There is no broad match. Advanced Match is NOT the same as Broad Match
- Dilemma: You don’t have Broad Match. How do you come up with all the terms?
- Solution: Let Yahoo do the work. Use the Yahoo Keyword Tool. If the term is [polarized sunglasses], list it in all three fields in the Yahoo Keyword Tool. Keep iterating until no more keywords can be generated. The tool will generate only new keywords (ones not in your ad group or keywords list). Occasionally, when you generate the keywords on the second iteration it will tell you that you already have that word in the list. When this occurs, take all the keywords that you generated in the right box, put them back in the left box and it’ll sort itself out.
- Dilemma: I think my adgroups are well targeted but the quality index still isn’t good. What do I do?
- Solution: Divide and conquer. Quality Index is at the Ad Group level. Analyze keywords CTR performance in an ad group. Pause all the CTR keywords. Duplicate the adgroup. In the new AdGroup, delete the active keywords and unpause the remainder.
Yahoo Dynamic Keyword Insertion
Why use it? Yahoo gives you extra credit when you use Dynamic Keyword Insertion (!). He shows an example. My fingers are yelling at me.
Google Ad Competition
- Situation: I have 1 adgroup with 10 keywords and 2 ads.
- Problem: I think the first ad is performing better for some of the keywords and I think the second is performing better for the rest. How do I optimize without analyzing my web logs?
Let Google do the work for you.
- Match sure bids are targeting current metrics of success.
- Duplicate the ad group and run 1 ad in each.
- Let ads run until they have ample conversion history and
- rebid to your current metrics of success – ads will eventually cannibalize.
Google: The #2 Dilemma
He shows two ads. For the most part, the ads are identical. Ad 2 is not compelling and is prob penalized for a “normalized” CTR. Ad 1 probably got there first. Ad 2 is probably paying a higher CPC.
- Create a second ad with a different headline
- Edit Campaign Settings -> Ad Serving: Optimize
- Think about how glad you are you don’t use an auto bid management tool
Result: Number two now poses a threat to number one. It will increase their CPC due to an improve CTR rate for number two. If advertiser number 1 is using an automated portfolio bid management tool than advertiser 2 is in luck and they’ll probably pop into the number one spot. Now that you’re in the number one position, Google Bid Optimizer will switch back to your first ad and he’ll have to play the game (if he knows how). He does this at the end of the day so when he wakes up everything is awesome. Heh.
Addie Conner is next. And we’re running out of time. And I’m going to cry.
Yahoo MinCPC
Another revenue blocker put in by Yahoo to further reduce their market share (hee!). It’s similar to Google MinCPC that was recently replaced by MinFirstPaid Bid. Prevents keywords from showing up at all. Has hit larger percentage of keywords within an account. How do you fight it?
If you get MinCPC’d then…
Quality Score in Yahoo mainly lives on the ad/adgroup level. Ask Yahoo to create a new account unlinked to your current one. Unlike Google, Yahoo won’t port the data if they’re unlinked. Move the keywords over, watch the MinCPC disappear. If it reappears, repeat action but change creative.
Gaining Competitive Insight
Google data is saved in their system and ports from account if variables remain unchanged. This is a key feature for reorganizing accounts because you don’t lose valuable history. It’s also a great way to peer into competitive metrics.
Disclaimer: These business practices are purely theoretical and not recommended for actual business use.
Google a keyword you want to gain some insight into. Capture exact ad, click on the ad to get the destination URL. Watch for any appended dynamic URL parameters. Strip those if you don’t know how to revert them to parameter form. Upload via AdWords Editor into new Google account. Observe.
The QS and FirstPageMinCPC should transfer. Sometimes there are slight differences possible due to account level quality score. You can do this with your own ads too to see if a new account might be beter. If you were to actually let the ads run, you could build bid to click and click to position models.
Knowing competitors can do this, keep an eye on your referring URLs and look for URLs that don’t use your exact structure. Watch the discrepancy between clicks and arrivals. Look for any ads that you did not write. Investigate any seemingly unprovoked Quality Score drops.
Dominating the Page
This is also known as double serving. A lot of companies are doing this right now to take over entire SERPs.
It’s also legal. The engines realize that a single company may wish to run multiple sites to micro target different business segments and optimize user experience. General rule is no great than > 80 percent overlap in user experience. Targeted URLs and landing experience can often lead to better performance.
Oh, my goodness, I hope some of that was useful. I gotta run. Quick break between sessions this time. Have mercy on me.
