A report on the "Natural Born Clickers" ("NBCs" from here on out) was released today by Starcom, Tacoda, & Hitwise. The data points to some head-scratchingly bizarre figures about the people who click on paid display advertisements (emphasis mine):
...heavy clickers represent just 6% of the online population yet account for 50% of all display ad clicks. While many online media companies use click-through rate as an ad negotiation currency, the study shows that heavy clickers are not representative of the general public. In fact, heavy clickers skew towards Internet users between the ages of 25-44 and households with an income under $40,000. Heavy clickers behave very differently online than the typical Internet user, and while they spend four times more time online than non-clickers, their spending does not proportionately reflect this very heavy Internet usage. Heavy clickers are also relatively more likely to visit auctions, gambling, and career services sites – a markedly different surfing pattern than non-clickers.
If you're worried about the online ad market taking the news badly, don't despair yet. It would appear that there's good news mixed in. Apparently, display ads, much like TV, radio, and print ads, don't demand immediate action to produce results.
...data suggests no correlation between display ad clicks and brand metrics, and show no connection between measured attitude towards a brand and the number of times an ad for that brand was clicked. The research presentation suggests that when digital campaigns have a branding objective, optimizing for high click rates does not necessarily improve campaign performance.
OK, so if as long as we're not targeting clicks, but rather attempting to garner attention and brand recognition with our display ads, we're in the clear. Winners in this scenario are probably big brand advertisers like Dell, Motorola, Apple, American Express, and Macy's, while losers might include that lovable brand behind the "punch the monkey" banners. In essence, the report's telling us that skimming over an ad is good for getting that brand into a person's consciousness (and making them more likely to recall that brand in future purchase decisions). Conversely, targeting the "click-heavy" demographic and getting traffic through display campaigns might produce much narrower targets than we might hope.
In simple terms, here's some probable winners:
The Movado ad at top targets wealthy readers of the New York Times, many of whom probably have gifts for Valentine's Day on their minds this week. The ad doesn't demand a clickthrough - it barely requests one. Instead, it's designed to make the reader consider the upcoming event and brand them with Movado's name. Similarly, the Sheryl Crow ad (which I found in the movies section on Salon.com) is barely targeted at clickthrough. It's much more like a TV ad - attempting to use endorsements from trusted sources to remind the visitor that this is a high quality album.
Now some possible losers:
Both of these require reading, not just scanning, and serve to entice a click, rather than attempting to simply impart a quick brand message. The first, from LendingTree.com, I found on Yahoo! Finance, while the second comes from HotJobs - obviously, these are both well-targeted from a content perspective, but the study would recommend that the clicks they receive may not only be less valuable than they think (as they're continually over-reaching NBCs), but that using CTR as a formula for success is a flawed metric.
The thing is - I'm not totally convinced that display advertisers should give up on the CTR, as the research suggests:
...said Erin Hunter, executive vice president at comScore. “For many campaigns, the branding effect of the ads is what’s really important and generating clicks is more of an ancillary benefit. Ultimately, judging a campaign’s effectiveness by clicks can be detrimental because it overlooks the importance of branding while simultaneously drawing conclusions from a sub-set of people who may not be representative of the target audience.”
If you can earn positive ROI on display ads, for goodness sake, don't stop just because half your clickers all come from a similar demographic. If you can pay for only unique IP address clicks or reach into sites whose demographic profiles differ drastically from the NBCs, you might have even better success, but it doesn't detract from the already existing value of the investment. I'm forced to wonder whether this study might have a negative impact on ad placement, or whether the advertisers will read between the lines and see the extra value that they're currently not calculating.
BTW - Has there been any credible research on over or under-represented demographics/psychographics among those who click PPC ads at the engines?
A similar study was done a while ago by the folks at AOL.
https://blogs.mediapost.com/spin/?p=1085
They identify the Heavy Clickers as:
Predominantly female, indexing at a rate almost double the male population. They are older. They are predominantly Midwesterners, with some concentrations in Mid-Atlantic States and in New England. What kinds of content do they like to view when they are on the Web? Not surprisingly, they look at sweepstakes far more than any other kind of content.
(Both studies are focused on banner ad clicks. Not PPC search)
I have designed hundreds of banner ads over the years. I have found that the key to getting people to click on your banner ads is to make them not look so much like ads. However... Getting people to click on the banner ad isn't always the goal.
In many ways banner ads are more measurable than PPC search because you can cookie people who do nothing more than view the ad but not click on it and track it to a conversion. This is called a "View Conversion"
You can do banner ad tests where half the audience will only see a generic banner for something like a public service message and the other half the audience will only see your actual banner ads for a month or so.
Then... You can measure the difference between the group who never saw your banner ads and the group who did see your banner ads to calculate "Lift".
You then use the "% of Lift" to calculate the effectiveness and ROI of your banner advertising.
You can NOT do this with PPC search. You can NOT cookie people just for making a query. You can not measure the purchase intent of people who are not exposed to your PPC ad.
It would be great if you could do that with search. Can you imagine?... Let's say you rank #1 for something like your brand name: If you could cookie people just for making a query, you would know exactly what % of people would make a purchase with or without the extra PPC ad.
If you're tracking Return On Investment instead of just clicks, this finding is not a cause for concern. Oh no, of the folks clicking on my software ads, many of them might be Serial Clickers! (Not actually a shocker, especially as I'm selling solidly to the demographic that AOL identifies as the heavy clickers.) But if I'm making 100%+ ROI on PPC ads, remind me why I'm supposed to worry about this again?
Serial Clickers are just a cost of doing business, like the 60% rake from Google on Content Network ads is a cost of doing business. You don't say "Darn, if Google wasn't here I could negotiate directly and save 60%" or "Darn, if it weren't for those midwestern ladies I could save 60%", you say "Bring me the clicks until the marginal ROI ceases to be positive. Thank you sir, may I have another."
Don’t forget about Re-Targeting PPC visits with banner ads:
This is when banner ads are shown to people after they have visited the website. It’s a fairly simple process.
1) A small strip of HTML is added to all of your landing pages so the ad network can place a tracking pixel on the user’s browser.
2) When users visit sites in the ad network, they see your banner ads which are designed to lead them back to our website to complete the purchase or at least keep your website in their mind during the consideration phase.
The pricing for that sort of thing is also very economical. 100,000 (Yes… 100K) banner impressions only cost $200 with leadback.com (Advertising.com network). Tacoda (Encore) also offers a banner ad, re-targeting program but I’m not sure about the pricing.
Most people won’t click on banner ads but re-messaging visitors with several banner ads after a visit can increase the overall conversion rate for search.Let’s say that you spend $20,000 or more every month on PPC… Why not follow up spend a measly $200 to show those same visitors an extra 100,000? (Yes…$200 buys 100,000 banner ads)
In my opinion, you can’t target a more qualified audience than your own PPC visitors.
Dave, Awesome concept, thanks for the reference!
We are in a fairly unique niche that is growing and branding ourselves is very difficult to a larger audience. We have recently tried newspaper & radio ads and both of them have flopped. But, we have also implemented an email follow-up campaign with our recent visitors and it has increased our overall sales by 10-15%.
Meaning...the people who are aware of our industry and have visited our site, are extremely valuable to us. Anytime we can re-target these visitors is definetely worth a shot. Thank you very much!
BJ
This report makes quite a few bold claims, and yet is very light on details. For example, what brands were tested? What niches? Travel, music, tennis shoes, cars, financial? Young people would be more likely to click on ads for sneakers than for retirement accounts.
An interesting component of this article is the lack of SEO benefitit provides the site. This is a dead end article, with no links back to the original site - it is a link sink. Unless Starcom creates links back to their core website, this viral traffic spike will be lost.
It is an interesting article, but the end result is nothing. We already know that web-action is strongly banded from studies about online communities.
The article says that 25-44 year old $40,000 households click more and spend less, that is useful if you are targeting that demographic, but most advertising managers are more concerned with the bottom line. As you can see in the comments here. Plus, CPM campaigns are cost independent of clicks.
Probably the biggest effect that this article will have is now PPC managers will have to listen to more complaints from semi-informed clients. The Natural Born Complainers will probably be all over this.
Most banner ads are purchased as "CPA" or cost per thousand banner impressions. (Not cost per click)
Even if you design banners to appeal to heavy clickers, its unlikely the extra clicks will drive up your costs or negatively impact your ROI.
Maybe I missed someone mentioning it earlier, but there is another demographic angle to this: the "Make Money Surfing the web" crowd.
Every single newspaper I've opened up in the last couple of years has ads for people who are paid to just surf the web. Of course, reading the fine print, you are paid a couple of pennies to surf to a list of sites (owned by the person paying you) and are asked to click on at least 2 ads (Adwords, etc).
If this sounds like click fraud to you, then you probably know something about PPC.
Look at the list of sites these people are visiting. they all have one thing in common: free stuff. Free money for clicking on ads, free games, free stuff for filling out sweepstakes, free contests, free samples, etc.
Do a search for "make money on the web" and visit a couple of the sites that show up. They target the demographic you've named (ie low/no income people and/or stay at home types like housewives) and tell them to do exactly the things that they are doing.
These people may be a target demographic for some companies, but due to the free/paid surfing incentives, I don't think I'd consider them to be visitors in the normal sense, since the goal is to click ads, not to buy/learn about products.
Oh, and that idea that you are "branding"? Yeah, OK. Honestly Rand, I'm not buying that there are very many "winners" at all in this (other than the surfers themselves, of course).
Remember that old saying about 50% of your advertising budget is wasted, but you don't know which 50%?
Well, this 6% demographic would probably be a good start. Interesting that it accounts for...what was that number again? Let's see... Oh - 50%!
What an amazing coincidence.
Ian
Hi Ian,
Banner advertising is tricky but you can measure the ROI pretty well if you have the resources to conduct the test described in my first comment where you measure the diffrence between people who see your ads and people who don't. It's an expensive test though because you need to purchase a LOT of public service messages in order to conduct the test.
If you have the ROI, why would the heavy clickers matter? Seriously businesses need to start paying more attention to the advertising budgets and not just worry about industry news releases.
That's like saying that as long as your store is profitable, you should not worry about shoplifting.
To a point, that's true. But only to a point.
I suspect that many of these heavy clickers tend to work at certain times during the day. I'll bet you if you tracked your ROI by dayparts instead of overall monthly/weekly totals, you'd start seeing a correlation between time of day and ROI - ie you'd get a much higher bounce rate and lower site penetration with ad clickers. And almost all would be on the content network.
The answer, of course, is to watch your content network ads very carefully, instead of just lumping them into the "I don't care what they do as long as I make a little money off them" category. Because if you don't, your competitors will.
Having a good ROI doesn't help much if your competitors all have much better ones than you. Business does not occur in a vacuum (unless you are a monopoly).
To paraphrase, businesses need to spend more time measuring "why" and "how" they are making money than on the bottom line. There are a lot of companies that are seriously in trouble but don't know it because they have one or two profitable areas that are holding up the rest of the business - on paper.
The bottom line (and ROI) usually hides far more critical information than it reveals. As an analogy, Microsoft is a very profitable company - does that mean that they don't have to fix anything?
You are better off measuring ROAS, and many other things, in addition to ROI. Not knowing who your visitors are or why they are on your site (and not caring) is a great way to go out of business.
Ian
Yes, you would like to cut back on your shoplifters, as this should be a part of managing your ad budget. But my point was for companies to stop only paying attention to what the media is telling them, rather start watching their own findings. It's amazing how many multi-million dollar companies, don't know how well their advertising budgets are being spent. I've personally dealt with a few of them.
To carry on with the point that you also raised, there was a case study I read recently that talked about a lingerie website. They would cut off all their adverstising mediums in the afternoon to avoid the late night crowd. Then again when morning came they would turn it back on, this made a huge difference in their advertising budgets.
Sorry for the late response, I didn't recieve any emails within SEOmoz notifying me of any new posts.
But my point was for companies to stop only paying attention to what the media is telling them, rather start watching their own findings.
I'll certainly agree with that! It's important to keep track of trends and news, but only in context of improving your business, not just doing the old "monkey see, monkey do".
Cheers,
Ian
The "ad-blindness" is something the ppc crowd is striving for. Google Adwords are a lot less recognizable as such than used to be the case.
My father's ad-blind too - he can't seem to tell the difference between AdWords that match a site design and the content of the site itself.
It's downright scary.
Ian
I'd say my wife very much fits that profile. She's in that age-range and eBay addicted. She also clicks ads like she gets paid for it (she doesn't). The funny thing is, she has no idea those are ads.
She's "ad-blind" in a way more like color blind. Where we just don't see the ads she can't tell them apart from the content.
I've thought about educating her and then realized it was much more valuable to let her be an average user so I can learn from it.
Still I cringe as she jumps from ad to ad to ad...
This is a great topic and happy to see some thoughts on display media here!
The Lending Tree example is likely a different spend model. It may be CPA or PPC so it's a little unfair to compare it to the CPM models used in branded campaigns.
What's exciting is that with new types of display units such as Gadgets and Widgets advertisers are getting metrics on actions in the unit itself that can be used to help optimize and create new performance models. Advertisers are even getting Page Views brought back to them for interactions with these units on other properties.
All this is great news for search marketers since the action based ROI and segmentation strategies needed to be successful in the new world of display are very similar to search. I guess it's no wonder then that Google bought DoubleClick!
This post instantly reminded me of my best friend who clicks his mouse so loudly and frantically that you can audibly hear his clicking on a phone call. He's broken 2 mice last year and one this year already! He believes that every web page should be attacked like a beta tester.
More on topic: I wonder how this demographic behaves with text ads. Are they equally clicky on PPC ads and do they make up a large percentage of clicking and a low percentage of buying? In that case they are all working for Google and they don't even know it.
Also Psychologically: Why do you think they are so clicky? Is it the way that they behave with all applications (like my friend), or is there some burning need they are trying to fill that they believe is just beyond the next display ad?
And Demographics: Are they young poor people like Rand's article points to? Or are they middle aged female midwesterners like Hawaii SEO cites?
Also, the goal of some PPC campaigns is sometimes to enhance brand exposure rather than focusing on CTR per se. It is hard to measure the effectiveness of branding campaigns, but the general rule is - the more the merrier.
Perhaps the heavy clickers are a significant contributor to the fact that many listings convert better in positions further down the page... high conversion rates are more significant than high CTR for most campaigns!
A lot of PPC prospects ask about this - they say "but I don't click on those links - who does?". It's a hard question to answer and I'd love to see some research on that side of things.
As you say, as long as the campaign is profitable, this isn't necessarily a problem, but if you can profile and avoid clicks that aren't valuable to you, then you can turn unprofitable into profitable or profitable into more profitable...
I also work in general direct response and DRTV. When planning media campaigns with late overnights and rather obscure or new channels the "new" clients invariably say "I don't watch that, who watches that" or "I'm not up at that hour, who is up at that hour to see it"...
...the correct response is the same...politely remind them that they are not the one and only monolithic buying public, and it's not just the total number of viewers but the cost per relevant viewer.
That's not to say the work involved in placement and cost of every venue or campaign is worth while. If your client cares about anything more than simple brand impressions (and most if not all of them should care!) then, as you allow, the metric that matters in the end is the cost to "conversion" whether that's an email sign-up, a blog entry, or a purchase.
To which you can politely reply: "From my experience, the 2-8% of people who see the ads", then show them some reports from your previous campaigns. That usually answers the question.
I have a client who had use PPC before engaging our organic seo service. There were really quite a constant click thru and also they bounces.
Interesting when switched over to organic SEO, the incoming traffic become lesser at a relatively constant number on weekdays. But the traffic conversion rate is now higher.
Do the "Heavy Clickers" Negatively Impact Display Ad Campaigns?
Maybe.
I think to be successful at a PPC campaign you need to fully understand your ROI. Some clients want clicks to generate forms being filled out and leads. Other clients want branding.
What I find difficult is getting my sales team to understand the difference and then determine from the client what they hope to achieve. I think its also different if you are looking at ads just on adwords where click through rate helps improve your Quality Score and thus help with cost.
In an Adwords environment it’s my opinion that competitors are sharing clients and heavy clickers. It seems if you are going to click on one ad you are probably going to click on most ads.
I will preface this by stating I am a novice... but would this study possibly lend more weight to impressions in determining the effectiveness of a PPC campaign? If your ad is simply for branding purposes, as with Movado, it seems to me the views would be just as important as the clicks.... just a thought and I would be interested in any and all feedback so I may better understand :)
"If your ad is simply for branding purposes, as with Movado, it seems to me the views would be just as important as the clicks"
Not necessarily… In most cases the advertiser wants you to click-through to a relevant landing page with even more information. A quick glance at a banner ad is better than nothing but a visit is much, much better, branding experience, even if there is no conversion.
For example: The pharmaceutical industry can NOT sell prescription drugs on the internet. Drug manufactures place banner ads for prescription drugs online to create awareness in the hope that the ailment will become more closely associates to the brand of drug. Hopefully, people suffering from the ailment will click the banner to learn if the drug is right for them or not. (Conversions can NOT be tracked in this situation)
For me the lesson here is that PPC managers should always been on the look out for any information that's going to give them an edge in reducing CPA.
With the article's definition of NBC the demographic profile was too broad for me to take action. IF the article had said 90% of NBC are from the midwest state of Ohio and do their clicking between 5 and 7pm (yeah, i wish), I would be running to campaign manager to make adjustments.
But that's not the case here, it's a broadly defined group which can't really be adjusted for in bidding and as Patio11 put it, these NBC are just a cost of doing business.
hmm, this shores up my belief that the message is so important, and the need for actionable text (aka "click here, now!") is just not that important compared to brand, where display ads are concerned.