This is one of the most promising marketing startups I've seen in awhile: Predicative.
We all have the same problem: How can you make ads show in more relevant places? Predicative has developed a "location intelligence engine" that helps you to target ads in genuinely new ways.
Also worth noting:
The Predicative dataset contains billions of global geolocations, painstakingly generated by data mining a variety of extremely large datasets. Our proprietary indexing technique is lightning fast, making Predicative ideal for programmatic buying.
I can't wait to get my hands on predicative!
This is part 2 of a 2-part series about advanced location targeting in AdWords. See also part 1: Demographic (Household Income) targeting in AdWords.
Google has quietly launched the ability to target central commercial areas, airports, and universities within AdWords. They call it "places of interest" targeting. Here's how it works:
It's been possible to target these areas manually for a long time, using radius targeting. This is a nice upgrade that makes things easier. Or does it?
Does places of interest targeting work?
One of my clients sells into commercial areas, so I gave that feature a spin. In my own testing, only the large commercial areas are included. Downtown areas in smaller cities were missed. Unfortunately, Google doesn't provide any background or documentation on this feature. Like its cousin household income, this feature is somewhat half-baked. Still, it's nice to see Google heading in this direction.
I haven't tested University targeting yet. Although, it's easy to do this manually.
Why would I use central commercial targeting?
As with demographic targeting, it's not hard to imagine how this feature is useful. If your business offers services or products that are of interest to white-collar suits downtown, or students, or travellers, "places of interest" targeting is worth exploring.
Google has quietly launched demographic targeting for search campaigns in AdWords.
Here's how it works:
Quite simply, you can select the household income of people who will see your ad.
It's not hard to imagine the possibilities. If your company sells low-end laminate flooring, you might gain an advantage by advertising in the "Lower 50%" income bracket. If you sell private jets, it might make sense to limit your searches to the top 10% of household income.
Okay, does this really work?!
Last time I checked Google hadn't even announced this feature, much less describe how it works. How do you know that the top 20% is really what it appears to be? Does this include business parks? How large are the areas?
Keeping complexity out of AdWords
As Google adds more advanced features to AdWords, it's clear that they face a dilemma. What's the balance between user-friendliness and fine control? Demographic/household income targeting appears to be a case where Google hasn't decided which way to go.
And Google isn't the only player in this space. My friends at Predicative are developing an advanced location intelligence engine, which could likely be more useful than Google's version, which inevitably has "user friendliness" as a more important requirement.
Places of Interest
In my next post, I will talk about places of interest targeting. This includes targeting central commercial areas, airports, and universities.
Recently I was reviewing Dynamic Search Ads, a new AdWords feature, and wondering how Google's motivations play into their feature announcements.
Clearly, Google is a financial superpower. As one of the largest companies on earth, they make the majority of their money from AdWords. And like all public corporations, they are self-interested. I'm not a lawyer, but my understanding is that public companies are obligated to act in this way, and doing otherwise could result in investor lawsuits.
I've been an AdWords professional since 2002. In those early years, Google was clearly trying to build a first-rate product. That was their priority. Over the years, their feature announcements gradually became more self-motivated. This isn't a problem, but as an AdWords user, it's important to understand how decisions get made.
Ultimately, Google is trying to balance several issues:
So basically, it comes down to profit. And that's okay.
Evaluate new AdWords features for YOUR needs.
Some people have the same reaction every time a new AdWords change is made: "It's all about Google's stock price!" I don't think that way. The questions I ask are:
It's about the ROI! Your ROI!
Dynamic Search Ads is a feature for Google Innovations that lets you target relevant searches with ads generated dynamically from your site. Google keeps a database of your content, and uses this to target your ads. When a user search happens, Google generates a custom ad based on that query (using your predefined parameters). Dynamic Search Ads enter the ad auction just like normal PPC ads.
Benefits of Dynamic Search Ads
The principal benefit of Dynamic Search Ads is convenience. If you run an active website, with new products arriving daily, or frequent changes to information, you may benefit from Dynamic Search Ads.
When evaluating any new AdWords feature, it's always useful to ask "What's in it for Google?" In this case, I think Google is simply trying to expand their keyword inventory, and spread the money more evenly across un-monetized areas of that keyword inventory. I don't smell anything fishy.
Are you missing relevant searches?
Anyone actively managing an AdWords account knows that new searches are important. Keyword tools can scan your website to find these words. Dynamic Search Ads is in some ways similar to that, except it happens without your involvement.
Dynamic Search Ads and Low Search Volume
I wonder how Dynamic Search Ads will play together with Low Search Volume keywords. On the sites I manage, much of the content would be considered LSV. Is that stuff included as triggers for Dynamic Search Ads? Could this be a way to unlock the power of the long tail. Is this effectively an end to Low Search Volume? I hope so.
Adwords isn't a meaningless game of math, stats, A/B testing, rules, scripts. First and foremost, it's a marketplace of ideas, culture, and people.
It's easy to think AdWords is like math. There are so many numbers to think about. Thousands of keywords, ads, each having their own stats. You can get multi-dimensional about everything and pimp out your pivot tables until you drive yourself crazy. (I know, I sometimes dream of my AdWords stats).
But taking a step back, it's not about that. It's about people doing searches to find what they're interested in. A lady looking to hire a web designer for a small business. A man wondering how to fix his aching back. A kid gathering research for a science project.
Math doesn't care about you. People do.
Optimize for real people. Use the AdWords stats to help you. Don't fall into the trap of optimizing for the sake of the stats. The math doesn't care. The people care.
I read something about Quality Score at least once per day. This is a hot topic. But recently someone asked me, "Does quality score even exist? How important is it?"
I assure you, Quality Score exists. If you want to succeed in AdWords, you need to care about Quality Score. But how far do you take that?
I like to tell people this: Forget about Quality Score (QS), and spend your time worrying about quality, period. Use proper AdWords hygiene for Quality Score, but put your best efforts into real quality. Why? Because Google isn't human, you are.
You are the judge of quality.
You are a human, living in (likely) North America, speaking English. You've immersed yourself in this culture for a long, long time. Your BS-detector can sense a Nigerian scam before you even open an email. You are very well equipped to judge your own AdWords for quality.
Look at your queries. Are they whole searches or partial ones. Fully on-target, or near-misses? Some industries are so crowded, that you absolutely must have every keyword be fully-targeted. For example "buy books about AdWords" is more focused than "books about AdWords". The word "buy" adds a tremendous amount of meaning to the search.
Quality Score isn't everything.
Your industry is different, and nuanced. You understand it, likely better than anyone else. Make sure your own feelings about keywords and queries are included in decisions about quality. AdWords isn't a meaningless game of math. It's a complex cultural marketplace.
The importance of Quality Score (QS) has risen consistently over the past few years. When first introduced, it made only a slight difference whether you had a high quality score or a low quality score. More recently, quality score can make or break your campaign.
Knowing your keyword-level quality score isn't enough anymore. It's generally acknowledged that Google has an account-level quality score, campaign-level quality score, ad group-level quality score, and keyword quality score. Only the keyword quality score is published. The rest is guesswork.
Or is it? As someone with a programming background, I sometimes find exceptional uses for AdWords Scripts. I have one such script that aggregates quality score for ads groups, campaigns, and account-level. In addition, it keeps a historical record of quality scores, so that I can see them trending up or down. If you'd like help with AdWords Scripts to track your quality score, please contact me.
Back to the burning question: What is a good quality score? What is a bad quality score? This is actually quite simple to answer. (A whole lot easier than what is a good CTR.) Here's the rule of thumb I use: Aim for 8, settle for 6, delete keywords at 4 or less.
Of course before you delete keywords, check if they are profitable. I've sometimes had odd situations where 2-QS keywords are profitable. Weigh the pros and cons. A 2-QS keyword is almost certainly dragging down your account. Is it worth it? It depends. But in most situations, your 2-QS keywords aren't making you money anyway. Fire them and don't hire them back!
Automated Rules and Quality Score
Quality score is calculated every time your keyword is triggered, so keeping up can be challenging. I make it a habit to set up automated rules to help with quality score management. In the simplest scenario, I would pause keywords that are 1-4, and make sure the keywords with quality score 5-10 stay enabled. This isn't fool-proof, but running this rule automatically each day will increase your group, campaign, and account-level quality scores.
Group-Level, Campaign-Level, and Account-Level Quality Scores
You can't see these quality scores in your account, but beware: What you can't see can sometimes hurt you. It's good AdWords practice to maintain mediocre-to-high range of keyword quality scores. Keep the garbage out, because it'll bite you when you're least expecting it.
Achieving a high CTR is key to winning with AdWords. CTR is the biggest single metric behind quality score, and you need good quality scores to succeed.
About a year ago, I asked a Google agency rep what a good CTR is. He shrugged his shoulders and gave the standard "it depends on a number of factors" answer. True, it does depend on many factors. It also depends on your industry, Google search vs. search partners, and a host of other things that may or may not be within your control.
So, what *is* a good CTR? Should you aim for 0.5%? 1.0%? 6.0%? I've heard pros say anything under 1.5% is considered horrible by Google. I've had a Google rep (not the one mentioned above) tell me that 1% is "actually doing pretty good".
I think it's safe to say that 0.5% CTR is not "doing pretty good" under any circumstances. At half a percent, you need 200 impressions for one click. Think about that: Imagine you're in a school gym with 200 people in it. Now imagine that only one of those people clicked. From a relevance standpoint 0.5% isn't defensible.
(Aside: It's sad that Google's search query report puts so many searches in the "Other search terms" row, because seeing those would make problems much more visible. There are some ways around this, but I'll share that with you another day.)
Now let's talk about 1%. I've managed successful campaigns that average 1% CTR, and struggle to get above that. Hey, if the business is successful, it's working just fine.
I've also seen campaigns that went poorly despite 4% CTR. Getting people to click isn't necessarily a measure of success. For example, there might be a misunderstanding caused by your keywords/ads that make users think they're clicking on something different than you do.
In many cases, it's best to focus less on CTR and more on old-fashioned common sense. Are your queries relevant? Are your keywords bringing in strong intent, or weak intent? Are you making money?
For those who didn't bother to read the above article, here's a false summary. 2% is the precise CTR you need in all cases. Just kidding. Go back and read the article!