Digital display advertising: (Still) looking like the weakest online promo tactic.

untitledI’ve blogged before about the lack of engagement with online banner advertising, and as time goes on … the picture doesn’t change much at all.

When you break it down, online banner advertising is a bust on several levels:

 

  • As of the most recent stats, clickthrough rates on online banner advertising are running about 0.08%. That translates to fewer than one click for every 1,000 times the ad is served.

 

  • Based on current pricing for online banner ads, that one click might be costing anywhere from $5 to $10 (and it might have even been an accidental click).

 

Despite these “inconvenient truths,” nearly two-thirds of digital ad spending continues to go to online banner advertising based on a “cost per impression” pricing model. Why?

One answer is that it’s an easy way to advertise a product or service. Simply supply ad creative to the publisher and let it be served online.

Another may be that advertisers consider banner advertising to be a basic component of any promotional campaign: prepare a mix of direct marketing, some search engine marketing, some print advertising and some digital display advertising, and you’re off to the races.

A third reason — related to the one above and I suspect one big reason why so much digital display advertising persists in the B-to-B realm in particular — is that publishers who offer a suite of promo tactics as part of a specially priced integrated program always throw in digital display advertising as part of the mix. It becomes the default option for advertisers as they approve bundled programs and the discount rates that come along with them.

Here’s a suggestion for advertisers going forward: Push back a bit and ask publishers to come up with alternative program options that don’t include digital display advertising.  The revised program might not look as promising at first blush, but then remember the stats above and you may well see the attributes of the alternative program in a more positive light.

What types of word terms perform best in social media?

Words that sell in social mediaEver since the rise of social media platforms, marketers have wondered if the terms and phrases that generate the best response in direct marketing also perform as well in the social arena.

One reason why:  There have been plenty of experts emphasizing how consumers don’t wish to be “sold” in their social interactions, but instead prefer to develop a relationship of give-and-take with brands.

Dan Zarrella, Social Media Scientist at HubSpot
Dan Zarrella, Social Media Scientist at HubSpot

Now we have some empirical analysis to guide us, conducted by Dan Zarrella, a social media scientist at SaaS inbound marketing firm HubSpot based on reviewing ~200,000 links containing tweets.

Mr. Zarrella found that the tweets that contain more verbs and adverbs experience higher clickthrough rates than noun- and adjective-heavy tweets.

Zarrella’s research also found that when social media posts ask for an explicit action on the part of the recipient, that tends to increase clicks and engagement.

For instance, retweets are three times more likely to happen when people are specifically requested to do so.

Interestingly, the most “retweetable” words in the HubSpot analysis turn out to be the same terms that do well in e-mail marketing and other forms of direct marketing:

  • You
  • Please
  • Post
  • Blog / Blog Post
  • Free
  • Media
  • Help
  • Great
  • How To
  • Top
  • Check Out

In a parallel research endeavor, a recent evaluation of blog posts by writer and software analytics specialist Iris Shoor reveals how much a post’s title impacts on the volume of “opens.”

In her analysis, Ms. Shoor studied posts on 100 separate blogs, using an evaluation technique that rank-sorted blog posts from the most read to the least shared.

What were the words that resulted in the most opens?  Shoor calls them the “blood in the water” terms:

  • bleeds leadsKill
  • Fear
  • Dark
  • Bleeding
  • War
  • Dead
  • Fantasy

Translation?  Negative terms are more powerful for shares than more ordinary terms (e.g., positive ones).

It’s very much like the old adage in the newspaper world:  “If it bleeds, it leads.”

That’s another takeaway from the most recent research:  What’s worked in the offline world over the years appears to be working very much the same way in the online space today.

Plus ça change, plus c’est la même chose …

The ad-supported web: Will it fall under its own weight?

Banner advertisingFor the past (nearly) 20 years, the biggest thing that’s kept the Internet free for users is advertising – banner display advertising in particular.

Bloggers and other online publishers large and small rely on revenue from web banner ads to fund their activities. That’s because the vast majority of them don’t have pay walls … nor do they sell much in the way of products and services.

Because of this, the temptation is for publishers to serve up as many display ads as possible on each page.

It’s not unusual to see web pages that tile ten or more ads in the right-hand column. Usually the content of these ads has no relevance to readers, and the overall appearance isn’t conducive at all to reader engagement, either.

And that’s the problem.

Because of conditioning, people don’t even “see” these ads anymore. The advertising space has become one big blur – as easy to gloss over as if the ads weren’t even there to begin with.  (When’s the last time you clicked on a banner ad?)

Attempts to come up with other display advertising types – pop-ups and pop-unders, animations and other rich media, skycrapers and so forth – haven’t done much to change the picture. Indeed, they’re so ubiquitous – and so predictable – we don’t even consider the ads to be annoying anymore; they’re just part of the “décor.”

I’ve blogged before about how clickthrough rates on banner advertising are bouncing along in the basement, making them less and less valuable for advertisers to consider placing. And ads that are priced on a pay-per-click basis can’t be giving advertisers much in the way of revenue either, since relevance and engagement rates are so abysmally low.

The bottom line is that we now have a “lose-lose-lose” situation in online advertising:

  • Advertisers lose because of near-zero user engagement, thereby limiting their potential to drive business.
  • Publishers lose because ad revenues aren’t sufficient to bankroll their activities.
  • Readers lose because of lack of relevance and an incredible degree of page clutter.

So it seems that the ad-supported online publishing model is in a bit of a fix – and the question is how things can evolve to create a more satisfying result for all parties.

I’d be interested in hearing your thoughts on this issue:  What will online publishing look like in another five or ten years?  Anyone willing to hazard a guess?

Online display advertising’s clickthrough performance: Now it’s just embarrassing.

poor online display ad clickthrough ratesRecently, several executives at the Advertising Research Foundation, the online analytics firm Moat and media buying company Accordant Media created six completely blank online display ads – no copy and no images – in three standard sizes and two colors.

The idea was to test the veracity of information being collected on clickthroutgh rates in online advertising and the algorithms that drive automated ad buying and selling — ads being automatically served where they will generate the most clicks.

Once the ads had been created, the crack team had them served via a demand-side ad platform, using both run-of-exchanges as well as being trafficked to publishers that would accept unaudited copy.

The idea was to see what sort of clickthrough rate would be garnered by these “faux” ads.

It’s common knowledge that clickthrough rates for online display advertising are abysmally low. But what transpired in the ARF-led team’s “nothing-doing” ad campaign was startling, to say the least.

The average clickthrough rate on these ads, across 1 million+ impressions served, was 0.08%

That clickthrough rate rivaled the better results achieved by online branding campaigns … and isn’t very far off the average performance for a direct response effort!

The team noted that the modest cost of this research effort (less than $500) was a “pretty good deal for a diagnostic check-up on a $100 billion machine.”

And what are those results telling us? A couple things, I think:

  • If you believe these clickthroughs are legitimate, there’s not much difference between someone clicking on an ad by mistake and the degree of intentional interaction with actual online branding campaigns.
  • If you believe that these clickthroughs might be happening to generate click rates that someone can roll up into an ROI calculation … you might be correct, too.

But ARF’s Ted McConnell reported that a follow-up screen appearing once an ad was clicked asked why the person who clicked did so — out of curiousity or by mistake. 

Here’s what McConnell reported:  “We detected no click fraud in the data we counted.  Half the clickers told us they were curious; the other half admitted to a mistaken click.”

McConnell noted that the team went beyond the follow-up query in its investigation.  “To obtain further insights, we tracked hovers, interactions, mouse-downs, heat maps — everything.  (Heat maps detect click fraud because bots tend to click on the same spot every time.)”

So, what’s the major takeaway finding from this experiment? 

If it’s that online brand advertising campaigns are going to deliver abysmal engagement, you already knew that. 

But if you think achieving a clickthrough rate of 0.04% or 0.05% for an online brand advertising campaign is an indication of anything in particular, you’re probably off by miles, too.

Was this “nothing-doing” research worthwhile, seeing as how it didn’t test “real” ads? 

I think it was … because the findings are telling us that below some threshold level, what we’re really experiencing is just noise. That’s not exactly what the online display advertising advocates want to hear …

Online Display Ad Effectiveness: Skepticism Persists

Online Display AdvertisingAs the variety of options for online advertising have steadily increased over the years, the reputation of display advertising effectiveness has suffered. Part of this is in the statistics: abysmal clickthrough rates on many online display ads with percentages that trend toward the microscopic.

But another part is just plain intuition. People understand that when folks go online, they’re usually on a mission – whether it’s information-seeking, looking for products to purchase, or avocational pursuits.

Simply put, the “dynamic” is different than magazines, television or radio — although any advertiser will tell you that those media options also have their share of challenges in getting people to take notice and then to take action.

The perception that online display advertising is a “bad” investment when compared to search engine marketing is what’s given Google its stratospheric revenue growth and profits in recent years. And that makes sense; what better time to pop up on the screen than when someone has punched in a search term that relates to your product or service?

In the B-to-B field, the knock against display advertising is even stronger than in the consumer realm. In the business world, people have even less time or inclination to be distracted by advertising that could take them away from their mission at hand.

It doesn’t take a swath of eye-tracking studies to prove that most B-to-B practitioners have their blinders on to filter out extraneous “noise” when they’re in information-seeking mode.

This isn’t to say that B-to-B online display advertising isn’t occurring. In fact, in a new study titled Making Online Display Marketing Work for B2B, marketing research and consulting firm Forrester Research, Inc. reports that about seven in ten B-to-B interactive marketers employ online display advertising to some degree in their promotional programs.

And they do so for the same reasons that compelled these comparnies to advertise in print trade magazines in the past. According to the Forrester report, the primary objectives for online display advertising include:

 Increase brand awareness: ~49% of respondents
 Lead generation: ~46%
 Reaching key target audiences: ~46%
 Driving direct sales: ~41%

But here’s a major rub: Attitudes toward B-to-B online display advertising are pretty negative — and that definitely extends to the ad exchanges and ad networks serving the ads. Moreover, most don’t foresee any increased effectiveness in the coming years.

That may explain why Forrester found that fewer than 15% of the participants in its study reported that they have increased their online display advertising budgets in 2011 compared to 2010 – even as advertising budgets have trended upward overall.

When you look closer at display, there’s actually some interesting movement. Google has committed to a ~$390 million acquisition of display ad company Admeld. And regardless of the negative perceptions that may be out there, Google’s Ad Exchange and Yahoo’s Right Media platforms have created the ability for advertisers to bid on ad inventories based on their value to them.

Moreover, new capabilities make it easier to measure and attribute the impact of various media touchpoints — online display as well as others — that ultimately lead to conversion or sales.

But the negative perceptions about online display advertising continue, proving again that attitudes are hard to change — even in the quickly evolving world of digital advertising.

The Ripple Effects of High Gasoline Prices

Shopping at home is rising along with gasoline prices.We’ve all heard the news reports about the effects that high gasoline prices are having on families who rely on automotive transportation for their livelihoods. It’s all well and good to promote the use of public transportation, but when your job is 25 miles away along suburban or rural roads, it’s often impractical to adjust commuting behaviors.

We’re also reading how high gas prices are affecting other aspects of the economy, such as the rising price of food items in the grocery stores due to higher transportation costs.

To this, we can now add another consequence of the high cost of petrol. Paralleling the gas price spike has been an increase in Internet activity.

Marin Software, a leading paid search manager platform for advertisers and agencies, has performed an analysis across more than $2 billion worth of paid search marketing activity. The firm established a benchmark based on the share of activity across the Google and Bing search engines, and then studied cost-per-click activity, clickthrough rates and conversion rates.

Marin evaluated the rise and fall in the volume of clicks along with the rise of gas prices over the time period January – March 2011. Voila! It found a positive correlation between rising gas prices and increased click activity.

In a similar vein, digital market intelligence firm comScore is reporting that U.S. e-commerce sales were ~$38 billion during the first quarter of the year. That’s up ~12% compared to the first quarter of 2010. And while e-commerce volume has been up over the past six quarters, this is only the second time the growth as been in double digits.

So the premise that the higher gas prices climb, the more the propensity is to shop from home and avoid the cost of driving appears to be on target. And it’s probably being helped along by the plethora of “free shipping” offers that are also out there — along with avoiding paying sales taxes.

Looking forward to the day when gasoline prices may plateau or fall back, it’ll be interesting to see if Internet activity drops back as well. Or will more people have become used to the comfort of shopping from home in their boxer shorts – so that online activity remains at an elevated level?

I have a suspicion it’ll be the latter.

Click Wars Opening Round: Plaintiffs 1; Facebook 0

I’ve blogged before about the issue of click fraud, which has many companies wondering what portion of their pay-per-click campaigns are simply wasted effort.

Until now, Google has been the biggest target of blame … but now we’re seeing Facebook in the thick of it also.

It’s only been in the past year that Facebook has made a real run for the money when it comes to paid search advertising. There are some very positive aspects to Facebook’s advertising program, which can target where ads are served based on behavioral and psychographic factors from the Facebook profiles of members and their friend networks. This is something Google has had a difficult time emulating. (Not that they haven’t been trying … which is what the new Google +1 beta offering is all about.)

But now, Facebook is the target of a lawsuit from a number of advertisers who contend that there are major discrepancies between Facebook’s click volume and the companies’ own analytics programs which suggest that the purported clickthrough activity is significantly inflated.

As an example of one company that is a party to the lawsuit, sports fan site RootZoo alleges that on a single day in June 2010, its software programs reported ~300 clicks generated by Facebook … but Facebook charged RootZoo for ~800 clicks instead.

While contesting the allegations vigorously, Facebook’s attorneys have also argued against the company having to disclose the source code or other details of how it calculates clickthrough activity, citing fears that the proprietary information could be leaked to outside parties (competitors) as well.

But that argument fell on deaf ears this past week. Instead, Facebook has been ordered by the U.S. District Court in San Jose, CA to disclose a wide range of data, including its source code for systems to identify and filter out invalid clicks.

In making this decision, Magistrate Judge Howard Lloyd stated, “The source code in this case implemented Facebook’s desired filtering, and whether that filtering [has] lived up to Facebook’s claims and contractual obligations is the issue here.”

This ruling appears to call into question the sweeping terms and conditions that Facebook advertisers are required to sign before beginning a media program. The relevant language states: “I understand that third parties may generate impressions, clicks or other actions affecting the cost of the advertising for fraudulent or improper purposes, and I accept the risk of any such impressions, clicks or other actions.”

[This isn’t the only incidence of Facebook’s broad and restrictive stipulations; another particularly obnoxious one deals with “ownership” of content posted on Facebook pages – basically, the content creator gives up all rights of control — even if the content came to Facebook through a third-party source.]

But in this particular case, evidently the terms and conditions language isn’t sweeping enough, as Judge Lloyd ruled that the plaintiffs can sue on the basis of “invalid” clicks, if not “fraudulent” ones.

Touché! Score one for the judges against the lawyers!

Of course, it’s way too soon to know how this particular case is going to play out – or whether it’ll even get to court. It’s far more likely that Facebook will settle with the plaintiffs so as not to have to disclose its source code and other “trade secrets” — the very things that cause so many marketers to see paid search advertising as a gigantic black hole of mystery that is rigged against the advertisers no matter what.

But one thing is easy to predict: This won’t be the last time the issue of pay-per-click advertising is brought before the courts. Whether the target is Facebook, Google or Bing, these skirmishes are bound to be part of the business landscape for months and years to come.

E-mail early birds? The worm may be turning differently.

Best time to deploy marketing e-mail messages.One of the great benefits of the “online everything” world in which we now live is the ability to evaluate nearly anything about marketing not with hunches or speculation, but with hard data.

A perennial question is what time of day is best to deploy marketing e-mails to customers and prospects. The higher the propensity to open and read these messages, you’re closer to the goal of converting eyeballs to clickthroughs … and to sales.

ReachMail, a Chicago-based e-mail service provider, recently studied a large sampling (~650,000) of the millions of consumer and business marketing e-mail messages it sends out for clients daily in order to determine open rate differences based on the time of day. It normalized the data to account for different time zones.

What ReachMail found was that there are differing peak open rate times on weekends versus on weekdays:

 Weekdays: Peak e-mail open rates are between ~11:30 am and ~2:00 pm.

 Weekends: E-mail open rates begin trending upward at ~11:30 am, but don’t peak until ~4:00 pm.

John Murphy, ReachMail’s president, had this to say about people’s weekday e-mail open rate behaviors: “You would think it would spike in the morning, but they’re looking at work e-mails in the morning. Once they’ve cleared out their inbox, they’re looking at marketing e-mails in the afternoon.”

ReachMail’s conclusion: It’s best to deploy weekday e-mails between 10:00 am and Noon. For weekend e-mails, deploy them between Noon and 3:00 pm.

And this additional tidbit also: Don’t assume e-mails sent during the week will perform better than those deployed over the weekend. “People’s engagement rates are up there on the weekend,” Murphy maintains. “It’s our habit of checking e-mail all the time.”

He’s sure right about that.

Third-Party e-Mail Lists: Clicks to Nowhere?

Clickthrough fraudOf the various issues that are on every marketing manager’s plate, concern about the quality of third-party e-mail lists is surely one of them. It’s a common view that the effectiveness of a purchased e-mail data file is worse than a carefully crafted in-house list based on input from the sales team plus opt-in requests from customers.

Part of the reason is that there’s less likelihood for recipients to be interested in the products and services of the company, which only makes intuitive sense. But there may be other, more nefarious reasons at work as well.

Ever heard of a click-o-meter? It’s the way some e-mail lists are made to look more effective than they actually are. In its basic form, this is nothing more than people paid to open e-mails with no other interest or intention of further engagement. The more technical way is to have an automated click setting, usually done through a rotation of IP addresses.

To the casual observer, this gives the impression of recipients who are interested in a company’s offer, but the final analysis will show something quite different: near-zero purchases or other relevant actions. The problem is that for many campaigns, ROI will be slow at first, so the grim reality that the company has been punked comes later.

The growth of the autobot click-o-meter phenomenon tracks with the growing interest in purchasing third-party lists based on cost-per-click (CPC) performance rather than on the traditional cost-per-thousand (CPM) basis. Not surprisingly, when list vendors started being asked to sell lists based on a CPC versus CPM basis, for some of them the temptation to “juice the numbers” was too great. And since many of the databases come from other sources and are private-labeled, the problem is perpetrated throughout the system.

Many purchasers have wised up to this issue by settling on one or two list brokers that they know and trust, by asking about the data source, and by asking for client references for the lists in question. If an e-mail database has suddenly changed in pricing from a CPM to a CPC basis, that may be another cause for concern.

Another option is to hire a third-party traffic monitoring service to assist with back-end analyses of e-mail campaigns to see what’s working or not working in specific campaigns and nip any problems in the bud before they do too much damage to a marketing effort.

But like anything else, self-education is critical. Most companies who are victims of fraudulent e-mail practices become so because their staff members are unaware of the potential problems. But the information is out there for the asking, and that knowledge will soon become “intuition” – usually the best predictor of ROI!