Your Paid Search Performance Is Relative
Behold the noble savage. This mythical creature, an invention of 18th-century romantic philosophers, coexisted peacefully with his environment, uncorrupted by civilization. My advice to you as a paid search advertiser this week is: be as “corrupted” as possible, if corrupted means recognizing that peaceful coexistence is nearly impossible. You’ll be measured and graded on your performance against other modern economic maximizers. To be blunt, the effort is going to cut into hammock time.
As Jean-Jacques Rousseau conjectured, in the early days of our species, “whoever sang or danced best, whoever was the handsomest, the strongest, the most dexterous, or the most eloquent, came to be of most consideration; and this was the first step towards inequality.”
Later came agriculture. Metallurgy. Fast forward a few millennia, and here we are. Unequal as can be. Especially in terms of our success rates in advertising in that upside-down “L” area on search results pages.
So how’s your paid search campaign doing? Are your ads “good”? Are you appearing in a good spot on the page? What’s a good price to pay for a click? What’s a decent clickthrough rate? And will your ads still be running strong this time next year?
The answers to the first five questions are likely to be dangerously subjective. The answer to the all-important final question depends on how you do relative to competitors in all sorts of ways.
Competitive search auction 1.0
There used to be a pretty simple version of this story. If you were after organic search visibility, then a high rank on a popular search query was a good thing. It might not always have made you money—that depended on what happened after the click—but certainly, you wouldn’t be facing too much risk if you ranked well.
The invention of the straight paid search auction—the more you bid, the higher you appeared on the page—made things tougher, especially as keyword costs rose. It’s worth harkening back to those days (in the case of Yahoo this was still in place as recently as last year in the US) to remind ourselves that the paid search auction introduced a whole new level of complexity to the question of “how your search campaign was doing.” Now, you were risking your money, just like those traditional print and broadcast advertisers had been all along. It was time to grow up! (Or as we say in Canada, welcome to your Carlsberg years.)
In search marketing, because the keyword is the form of targeting being fought over, and a click is what you pay for, the playing field becomes distinctly unequal among businesses of different sizes (and marketers of different skill levels) competing for visibility on the same keywords. If you need to pay $5 for a click and you sell a low-margin product, you are, for all intents and purposes, blocked from advertising on that keyword. Veteran advertisers have been feeling the impact of click inflation for some time.
The quasi-Darwinian logic that threatens to make some kinds of ads extinct doesn’t begin and end with profit margins. Advertisers with higher lifetime customer values or stronger overall business models are willing to pay more (and thus bid higher) to acquire a customer.
A related case, and not to be ignored, is advertisers who may be bidding on the exact same keywords, but in relation to vastly different business types with typical prices several orders of magnitude apart. Advertiser A bids on “power plant” because they’re advertising a guide to local bars and restaurants, and Power Plant is a local club. Advertiser B sells a high-end industry publication dealing with power plants. Advertiser C is a nonprofit seeking to publicize green energy. Advertiser D is “a full service heavy industrial contractor capable of performing most all facets of construction for gas turbine power plants.” Advertiser E is an eBay affiliate bidding on pretty much every word in the dictionary. This illustrates the point that as the ad auction fills up for a given keyword, you can’t assume that advertisers all have an equal shot at a bid price that will keep them profitable.
This process of big-ticket advertisers crowding out the little guys is far from complete. Take solace, at least, in the fact that the guys selling the actual power plants will tend to have much lower CTR’s on broad terms that appeal to advertisers selling more popular products. So in that sense, the ecosystem is forgiving and does allow for diversity.
In any case, a fixed notion of what your “business is” is always dangerous, just as it’s a mistake to believe that getting the “bids right” or the “keyword list built” or the “ads written” is going to be enough to succeed in a tough auction. These things are a good start, but as competition grows, you’ll need to optimize other aspects of your marketing—to the extreme, perhaps, of changing your product emphasis, pricing, or even your whole business model.
AdWords 2.5 and Yahoo Panama: Complexity as opportunity
Fortunately, the newer-generation ad auctions (Google AdWords 2.5 and Yahoo Panama) give you a fair shot at outsmarting your competitors by simply being diligent with your campaign organization and testing at a relatively surface level. By running smart campaigns that rank higher on ad quality measures, which include clickthrough rates, keyword relevancy in relation to ads and landing pages, etc., the diligent advertiser can often climb into higher ad positions at a reasonable cost.
Moreover, the advertiser who tests meticulously can discover ads that actually convert visitors to sales at a better rate, either by filtering some unwanted clicks through clarity of ad copy, or by injecting subtle psychological cues into the process.
You won’t get a great sense of whether your keywords’ clickthrough rates are “good enough” for the Google AdWords ranking algorithm just by looking at them in absolute terms. It’s relative to industry norms. That’s why sometimes your quality score is “Excellent” even with a CTR below 1.0%. And in other cases it’s “Poor” above 2%. Google has also gone on record stating that quality standards are “relaxed” in “less mature markets,” geographically speaking. Relative performance comes into play in a multitude of ways.
After the click
We’re not done yet. Conversion improvement experts like Bryan Eisenberg have long believed that it’s the website itself that is often the weakest link in your economic chain. Your paid search campaign economics can improve dramatically if you meticulously plan and test your user’s experience post-click, to the point where (let’s say) the same set of clicks converts at double or triple the rate. I wish those first-generation online media brokers back during Bubble 1.0 had educated the marketplace about the folly of sending paid traffic to a nonconverting website, but unfortunately, their incentive was to cloud the truth. So today, there’s plenty of work left to do. Often, a complete website overhaul is needed. In other cases, some judicious tweaking may suffice for rapid improvement, making it worth at least spending some money on clicks.
Take my client, who for argument’s sake I’ll say sells paint and wallpaper. Only 20% of their search traffic is paid because the paid search campaign economics are not currently favorable—they are lucky they rank high organically on some core terms, but the lack of traction on the paid side is hindering growth. New visitors convert to sales at a paltry 0.8%; returning visitors 3.6%. The real action is happening around that 0.8% number, because the majority of visitors are new. By testing 24 versions of the homepage, I intend to discover which factors are influencing conversion rates. This is a painful first step that interrupts the flow of normal business, but it may be essential for survival.
Businesses like this are often every bit as competitive as they need to be in spirit. But their outlet for that sentiment is often wrong.
Being hyper-competitive should not mean time-wasting gamesmanship with bidding, for example, or watching your competitor’s ads float up and down the page all day long. It shouldn’t mean worrying that your competitor’s tagline is taking business away from you, so you need to place your (superior?) tagline in the user’s face. As Guy Kawasaki might say, “get better reality” in order to improve your standing in the auction, as opposed literally jousting with competitors.
Back to bidding basics (and then beyond)
Returning to the basics of managing a paid search campaign: Let’s talk bids. Short-term, you’ll see all kinds of bold behavior by competitors in the auction. Sometimes, you’re lucky enough to be sitting in relatively uncompetitive spaces. Other factors being equal, the only way you can win long term is if comparable competitors are either too bold or too timid in their bidding on many of the keywords you’re scrapping over.
If everyone else is bidding too little or have yet to enter the auction (hello, Canada!), then your most rational course of action (assuming a sound campaign in the first place) is to increase your company’s annual paid search budget as high as management will permit, to make hay while the sun shines. I suppose it’s “good enough” to give your competitors a fighting chance to catch you by deferring your plans to grow market share, but if you’re a ruthless type, you should lobby your organization to increase its overall click budget.
Bidding to ROI objectives sounds straightforward, but it generally isn’t—even in consumer retail. Only the densest, highest volume campaigns seem to be optimal for the kinds of automated bid adjustments provided by third-party bid management tools. In large accounts with moderate to high (but not very high) volumes, I find that hundreds of difficult judgment calls are made every week or two when incremental adjustments are made to respond to ROI data on ads and keywords. Have the wrong person making hundreds of judgments? That’s hundreds of incorrect judgments, potentially.
To offer a simple example: an inexperienced campaign manager, Luke, builds out keyword groups for numerous categories of camping equipment. One of the highest-volume ad groups (let’s say it’s the Coleman Tents group) has the highest cost per order number of all the groups—over $100—so he simply “bid manages it” (with an automated tool or manually, doesn’t matter) into oblivion (12th ad position). The overall campaign ROI hits its number, but at suboptimal, “safe” volumes.
Main competitor Wendy is making so much money, meanwhile, that she can afford to invest money back into improving various elements of her business; as a result of this overall re-engineering, her target cost-per-order numbers are relaxed because a customer is worth more. Luke continues to nibble away, out of sight, out of mind, while Wendy continues to gain market share. One of Luke’s biggest mistakes (in this very hypothetical example) was turning down that high cost-per-order ad group bid, rather than recognizing that the broad match terms in it were triggering irrelevant clicks from people seeking household appliances, not camping equipment. By using negative matching, matching options with intelligent bids for different match types, and experimenting with the ad copy, Luke might have saved that ad group and opened the door to growth. Carry that expertise gap over to other parts of these competing accounts, and the performance gap just continues to widen.
If you think bid management still means “turning off the nonperformers,” as it did in 2002, then you’re behind the times. If your targets aren’t being met, is the answer to “nibble” in 12th ad position, basically giving up on the idea of entering the auction in earnest, or is it to address fundamentals or subtleties that you may be missing? Throw in the towel only when absolutely necessary.
Be ruthlessly intelligent, or intelligently ruthless
Your ads may be “good enough,” your campaign manager “smart enough,” and goshdarn it, people may like your website and products (sort of), but that’s little consolation if your competitor is better, smarter, makes more money from a click than you do, and gets to keep their paid ad above the fold as a result of their superior overall economics.
The research department here at Team Goodman tells me that dogs don’t really eat other dogs. But classic economic maximizers jockeying for best placement in the upside-down “L” arena, on the other hand? That gnawing feeling halfway up your mouse-using arm isn’t your imagination: It’s a competitor making mincemeat of your “pretty good” ads, unreflective bidding tactics, me-too landing pages, undifferentiated products and value proposition, and weak business model.
As more of the world’s “real” businesses move into the paid search space, the bounteous primordial wetland of low bids and clueless competitors will dry up. “Lifestyle”-based business models, such as click arbitrage, will be available to fewer and fewer players. This prediction has very little to do with editorial rules and quality-based bidding formulas; it has much more to the inevitable laws of supply and demand. If you’re thinking about career options and wardrobe choices, try 2007-era business casual at least one day a week. Save the loincloth for special occasions.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.