Everything's cyclical - it's just the length of the cycle that changes.
Big chain stores and mass-produced products are the norm right now. I don't know if this will ever change.
But I suspect that there is a growing trend of small producers and sellers chipping away at the long-tail of the behemoths. Led largely by the efficiencies that can be experienced at even a small volume online, but also appearing in physical locations, there are a lot of "cottage" companies out there. Why buy the same, generic, everybody-has-one stuff when you could get something of (probably) higher quality that's substantially more unique, interesting, and personal?
It will be interesting to see if this develops into a trend, and if the big guys respond with niche sub-brands of their own.
Wednesday, October 21, 2009
Wednesday, October 14, 2009
Whither The Clickers?
Last time, I explained why I don't think that clickthrough rate is a dead metric for online advertising.
But this still leaves the big questions:
Why do most campaigns have such low CTRs?
And why do so few people click at all?
I think there are two simple reasons (requiring not-so-simple fixes):
1. It takes a certain type of person to click much.
2. Most campaigns suck. Even the good ones.
A Certain Type of Person
People are people, regardless of the medium they use.
A small portion of the audience are hard-core coupon clippers, sweepstakes entrants, infomercial-callers, letter-to-the-editor-writers, and so on.
It's hardly surprising that a relatively small group of users account for the vast majority of clicks. Even if average clickthrough rates were to dramatically improve, I suspect that the heavy-clickers would keep up the pace and continue to dominate.
I don't have any stats to back me up (anybody know of a good study on the topic?) but I'd bet that this group are a strange mix of extremely high-value consumers (the type who will buy anything, who are easy to sway, and like to tell all their friends about their purchase) and virtually-no-value consumers (the type who don't actually buy anything, but browse anything and everything purely for the sake of their own curiousity). I don't think there are many of the sought after in-betweens in this group: those who browse anything and everything and become experts for their friends. Those guys need to actually be targeted properly.
Most Campaigns Suck
It's true.
Typical online ad creative is awful. Unclear message, weak call-to-action, low-quality imagery or animation, too much copy, not enough copy...
And the typical campaign is very untargeted. Who are you trying to reach? It still amazes me that the bulk of online ads aren't even geo-targeted.
Is it any wonder that very few people respond immediately to an ad that doesn't speak to them, and is just plain lousy?
If advertisers took the time and made the effort to segment (and sub-segment) their target audience, identify targeting filters and techniques, carefully select media providers (not just pick the top 10 in a category on Comscore, say), created multiple creative variations, matched creative execution to placement/audience, and other basics (that we should all understand by now)... Well, I suspect we'd see clickthrough rates improve (along with all the other applicable metrics).
Of course, I completely recognize that sometimes this just isn't realistic. Deadlines, budgets, resourcing requirements, and a thousand other obstacles get in the way.
But I'm sure there are plenty of cases where these obstacles can be avoided and the people involved just didn't bother to do things right. You could write a book on Internet advertising best practices, and hundreds of people have, yet many media planners, brand managers, creative directors, and so on simply don't apply these.
When this changes, brand marketing through online display advertising will have a chance to bounce back. Who knows? Maybe one day it will be the sexy-tactic-of-the-minute...
But this still leaves the big questions:
Why do most campaigns have such low CTRs?
And why do so few people click at all?
I think there are two simple reasons (requiring not-so-simple fixes):
1. It takes a certain type of person to click much.
2. Most campaigns suck. Even the good ones.
A Certain Type of Person
People are people, regardless of the medium they use.
A small portion of the audience are hard-core coupon clippers, sweepstakes entrants, infomercial-callers, letter-to-the-editor-writers, and so on.
It's hardly surprising that a relatively small group of users account for the vast majority of clicks. Even if average clickthrough rates were to dramatically improve, I suspect that the heavy-clickers would keep up the pace and continue to dominate.
I don't have any stats to back me up (anybody know of a good study on the topic?) but I'd bet that this group are a strange mix of extremely high-value consumers (the type who will buy anything, who are easy to sway, and like to tell all their friends about their purchase) and virtually-no-value consumers (the type who don't actually buy anything, but browse anything and everything purely for the sake of their own curiousity). I don't think there are many of the sought after in-betweens in this group: those who browse anything and everything and become experts for their friends. Those guys need to actually be targeted properly.
Most Campaigns Suck
It's true.
Typical online ad creative is awful. Unclear message, weak call-to-action, low-quality imagery or animation, too much copy, not enough copy...
And the typical campaign is very untargeted. Who are you trying to reach? It still amazes me that the bulk of online ads aren't even geo-targeted.
Is it any wonder that very few people respond immediately to an ad that doesn't speak to them, and is just plain lousy?
If advertisers took the time and made the effort to segment (and sub-segment) their target audience, identify targeting filters and techniques, carefully select media providers (not just pick the top 10 in a category on Comscore, say), created multiple creative variations, matched creative execution to placement/audience, and other basics (that we should all understand by now)... Well, I suspect we'd see clickthrough rates improve (along with all the other applicable metrics).
Of course, I completely recognize that sometimes this just isn't realistic. Deadlines, budgets, resourcing requirements, and a thousand other obstacles get in the way.
But I'm sure there are plenty of cases where these obstacles can be avoided and the people involved just didn't bother to do things right. You could write a book on Internet advertising best practices, and hundreds of people have, yet many media planners, brand managers, creative directors, and so on simply don't apply these.
When this changes, brand marketing through online display advertising will have a chance to bounce back. Who knows? Maybe one day it will be the sexy-tactic-of-the-minute...
Tuesday, October 13, 2009
The Reports of The Clickthrough Rate's Demise Have Been Greatly Exagerrated
It's an ongoing debate in digital marketing circles, but really brought to the forefront lately: How valuable is clickthrough rate (CTR) as a success metric?
A recent report, for example, indicates that the majority of clicks come from an extremely small group of people.
At an IAB event last week, a speaker basically said "Clicks are dead. You've got to get past looking at clicks."
Within a typical ad campaign, it's normal to expect a clickthrough rate of around 0.10 to 0.20%. And even that's a stretch for some campaigns. In other words, only one or two people click on the ad for every thousand that see it.
And, of course, clicking on an ad is just one of many steps in any consumer interaction -- alone, it does not give us a very complete picture of what actual happened in that interaction. Many interactions don't even involve a click (such as in cases where a user sees an ad, doesn't respond immediately, but at a later time performs a search for the advertised product or brand).
In addition to all this, CTR isn't even an exact measurement. In most campaigns, the number of clicks registered by the ad server is different than the number of visits registered by a site's log files (or other metrics methods).
So, with all these negatives, why do I think that the CTR isn't dead yet?
Lots of reasons. Among them:
1. Clickthrough rate is still better than most traditional metrics.
How many clicks did your latest out of home ad garner? How many people responded to your multi-million dollar TV campaign? How does this compare to the number that responded to your radio campaign that ran at the same time?...
Most traditional marketers simply cannot answer these questions. Those who can do so by spending a significant amount of time, effort, and money on research and analysis. Or they use direct response tactics (such as a unique phone number, coupon code, or other identifiers) to differentiate between ads, channels, offers, etc. (Which is great, but has all sorts of not-so-good impacts on branding and consistency).
2. Clickthrough rate is good for quick analysis of trends and general status.
At a glance, CTR can tell you if Creative A or Creative B is making a bigger impact, or if Placement A or Placement B is more relevant to your target audience.
Watching clickthrough rates change over time can help warn that your campaign is getting stale or, alternatively, tell you that you're reaching an optimal frequency level.
It's not perfect, and it's not a complete picture, but it can be invaluable when you need a quick-and-dirty understanding of what's going on.
3. Clickthrough rate is an important part of the big picture.
Although its importance as an individual metric is questionable, CTR can give us lots of great information when combined with other data.
Going back to Marketing 101 and the AIDA model (Awareness, Interest, Desire, Action), clickthrough rate can help us understand how well we're achieving some of the earlier metrics (Awareness to some degree, but primarily Interest). This info has a low value by itself (it doesn't do much good to gain someone's interest if they don't eventually Desire your product), but can help you understand where you're going wrong (or right).
A campaign with, say, a high CTR but low conversion rate might indicate that the creative is excellent, or very well targeted, or that the offer is strong at first glance, but that the landing page is weak, or that the details of the offer are disappointing when the user looks more closely.
On the other hand, a campaign with a low CTR but high conversion rate might indicate that the media plan needs to shift focus, or that the creative isn't getting the message across clearly, because the small portion of people who do click are actually quite impressed when they get to the landing page.
An ad with high engagement (like watching a video or playing a game) but low clickthrough rate might indicate that it's doing a great job of branding but needs a stronger sense of urgency.
A campaign with high CTRs -- but only for ads with high impression levels -- might tell us that the ads are too complex, or that a certain level of frequency is required, or that sequential messaging would be beneficial.
And so on and so on.
Without understanding the clickthrough portion of the bigger picture, our understanding of the end result is incomplete.
4. Clickthrough rate has become a standard.
Right or wrong, everybody understands (or can understand) how to calculate a clickthrough rate and what it represents. When we start talking about “engagement:” or “time spent” or “viewthrough” or “complete views” or “post click conversions” other metrics that are typically more valuable to an analyst, the waters get a bit muddy. What does an “engagement” really mean? Does 30 seconds spent playing an online game have the same value as 30 seconds spent viewing an online video? How long after an initial impression is made should its contribution still be attributed to a sale?
CTR might be simple, but that’s part of its value. GRPs don’t tell us much about the value of a TV audience, but they’ve become a standard metric for much the same reason, as have distribution and circulation for print.
So, if clicks are still important, this brings up the equally important question:
Why are clickthrough rates so low?
Some thoughts on this in another entry.
A recent report, for example, indicates that the majority of clicks come from an extremely small group of people.
At an IAB event last week, a speaker basically said "Clicks are dead. You've got to get past looking at clicks."
Within a typical ad campaign, it's normal to expect a clickthrough rate of around 0.10 to 0.20%. And even that's a stretch for some campaigns. In other words, only one or two people click on the ad for every thousand that see it.
And, of course, clicking on an ad is just one of many steps in any consumer interaction -- alone, it does not give us a very complete picture of what actual happened in that interaction. Many interactions don't even involve a click (such as in cases where a user sees an ad, doesn't respond immediately, but at a later time performs a search for the advertised product or brand).
In addition to all this, CTR isn't even an exact measurement. In most campaigns, the number of clicks registered by the ad server is different than the number of visits registered by a site's log files (or other metrics methods).
So, with all these negatives, why do I think that the CTR isn't dead yet?
Lots of reasons. Among them:
1. Clickthrough rate is still better than most traditional metrics.
How many clicks did your latest out of home ad garner? How many people responded to your multi-million dollar TV campaign? How does this compare to the number that responded to your radio campaign that ran at the same time?...
Most traditional marketers simply cannot answer these questions. Those who can do so by spending a significant amount of time, effort, and money on research and analysis. Or they use direct response tactics (such as a unique phone number, coupon code, or other identifiers) to differentiate between ads, channels, offers, etc. (Which is great, but has all sorts of not-so-good impacts on branding and consistency).
2. Clickthrough rate is good for quick analysis of trends and general status.
At a glance, CTR can tell you if Creative A or Creative B is making a bigger impact, or if Placement A or Placement B is more relevant to your target audience.
Watching clickthrough rates change over time can help warn that your campaign is getting stale or, alternatively, tell you that you're reaching an optimal frequency level.
It's not perfect, and it's not a complete picture, but it can be invaluable when you need a quick-and-dirty understanding of what's going on.
3. Clickthrough rate is an important part of the big picture.
Although its importance as an individual metric is questionable, CTR can give us lots of great information when combined with other data.
Going back to Marketing 101 and the AIDA model (Awareness, Interest, Desire, Action), clickthrough rate can help us understand how well we're achieving some of the earlier metrics (Awareness to some degree, but primarily Interest). This info has a low value by itself (it doesn't do much good to gain someone's interest if they don't eventually Desire your product), but can help you understand where you're going wrong (or right).
A campaign with, say, a high CTR but low conversion rate might indicate that the creative is excellent, or very well targeted, or that the offer is strong at first glance, but that the landing page is weak, or that the details of the offer are disappointing when the user looks more closely.
On the other hand, a campaign with a low CTR but high conversion rate might indicate that the media plan needs to shift focus, or that the creative isn't getting the message across clearly, because the small portion of people who do click are actually quite impressed when they get to the landing page.
An ad with high engagement (like watching a video or playing a game) but low clickthrough rate might indicate that it's doing a great job of branding but needs a stronger sense of urgency.
A campaign with high CTRs -- but only for ads with high impression levels -- might tell us that the ads are too complex, or that a certain level of frequency is required, or that sequential messaging would be beneficial.
And so on and so on.
Without understanding the clickthrough portion of the bigger picture, our understanding of the end result is incomplete.
4. Clickthrough rate has become a standard.
Right or wrong, everybody understands (or can understand) how to calculate a clickthrough rate and what it represents. When we start talking about “engagement:” or “time spent” or “viewthrough” or “complete views” or “post click conversions” other metrics that are typically more valuable to an analyst, the waters get a bit muddy. What does an “engagement” really mean? Does 30 seconds spent playing an online game have the same value as 30 seconds spent viewing an online video? How long after an initial impression is made should its contribution still be attributed to a sale?
CTR might be simple, but that’s part of its value. GRPs don’t tell us much about the value of a TV audience, but they’ve become a standard metric for much the same reason, as have distribution and circulation for print.
So, if clicks are still important, this brings up the equally important question:
Why are clickthrough rates so low?
Some thoughts on this in another entry.
Monday, October 5, 2009
Don't Get Defensive
Good advice for personal relationships, and equally so for business -- when confronted with a potential argument (or even just a disagreement), don't get defensive.
Two stories for this one:
A few months ago I was at a local home improvement warehouse store, waiting in line with some lumber. This store has a "contractors" line-up for people with large, bulky items. Due to a bit of ad planning in the store layout, the cashier area only has room for one or two big shopping carts and the line-up is then intersected by a major aisle through the store. So, the line breaks up for twenty feet or so and then reforms. Anyway, I was waiting in the third position (the first person after the break) with two customers in front of me. The second customer had a couple of carts, so they were standing with one by the cashier while their spouse waited with the other a few steps to one side. The person in front of them finished, and the person in line moved forward. Their spouse brought the other cart over to join them.
At this time, a staff member walked past and confronted the spouse. "The line up is back there", she said, pointing in my direction.
Okay, common situation for almost every store. The spouse had a legitimate reason for being there, though, and simply said "Oh, I'm with her".
Instead of smiling and moving on, the staff member decided to get upset. "There's no way I could know that!" she snapped. I still remember this because of the way she said it. You know that episode of Family Guy where Brian has a new girlfriend who's a complete moron? Near the end of the episode, Stewie and Brian are talking about how she talks, where every sentence.. ends with an upwards inflection... as if it's a question. That was exactly how Miss "Angry That A Customer Pointed Out Her Mistake" sounded.
Second story:
Stopped for dinner at a restaurant a couple of weeks ago. One of us wanted to order the soup and salad combination. Unfortunately, they didn't want the soup of the day and tried to substitute French Onion in its place. The waitress informed us that there would be a $2 charge for the substitution. I pointed out to her that, on the menu, the price difference between the two items was $0.50. Unless the soup in the combo is four times bigger than the one on the menu (doubtful, since the combo cost in total was only double the price of the soup alone), the extra charge seemed a bit high.
We didn't care about the $2. It's not a big deal. But the principle of charging a significant premium for a slight alteration seemed like a really poor business practice.
Instead of simply apologizing and saying something like "Unfortunately, we have to charge a standard $2 for all substitutions" or some other excuse, she became very defensive. She proceeded to tell us how she isn't responsible for the pricing. And how she has to enter the substitution into the computer when she orders and can't instruct the cook to make the change. And how it's a much bigger soup bowl than the one on the menu... And on and on. Meanwhile, we're just telling her "Okay fine. It doesn't make sense to charge that much, but we'll just order something else."
In the first case above, the staff member risked making something of nothing, and obviously created a bad impression for any customers within earshot. In the second case, the staff member risked turning a mildly irritating bit of bad pricing decisions into an entirely bad customer experience.
In both cases, the defensiveness was completely unnecessary and was worse than the initial problem itself.
Two stories for this one:
A few months ago I was at a local home improvement warehouse store, waiting in line with some lumber. This store has a "contractors" line-up for people with large, bulky items. Due to a bit of ad planning in the store layout, the cashier area only has room for one or two big shopping carts and the line-up is then intersected by a major aisle through the store. So, the line breaks up for twenty feet or so and then reforms. Anyway, I was waiting in the third position (the first person after the break) with two customers in front of me. The second customer had a couple of carts, so they were standing with one by the cashier while their spouse waited with the other a few steps to one side. The person in front of them finished, and the person in line moved forward. Their spouse brought the other cart over to join them.
At this time, a staff member walked past and confronted the spouse. "The line up is back there", she said, pointing in my direction.
Okay, common situation for almost every store. The spouse had a legitimate reason for being there, though, and simply said "Oh, I'm with her".
Instead of smiling and moving on, the staff member decided to get upset. "There's no way I could know that!" she snapped. I still remember this because of the way she said it. You know that episode of Family Guy where Brian has a new girlfriend who's a complete moron? Near the end of the episode, Stewie and Brian are talking about how she talks, where every sentence.. ends with an upwards inflection... as if it's a question. That was exactly how Miss "Angry That A Customer Pointed Out Her Mistake" sounded.
Second story:
Stopped for dinner at a restaurant a couple of weeks ago. One of us wanted to order the soup and salad combination. Unfortunately, they didn't want the soup of the day and tried to substitute French Onion in its place. The waitress informed us that there would be a $2 charge for the substitution. I pointed out to her that, on the menu, the price difference between the two items was $0.50. Unless the soup in the combo is four times bigger than the one on the menu (doubtful, since the combo cost in total was only double the price of the soup alone), the extra charge seemed a bit high.
We didn't care about the $2. It's not a big deal. But the principle of charging a significant premium for a slight alteration seemed like a really poor business practice.
Instead of simply apologizing and saying something like "Unfortunately, we have to charge a standard $2 for all substitutions" or some other excuse, she became very defensive. She proceeded to tell us how she isn't responsible for the pricing. And how she has to enter the substitution into the computer when she orders and can't instruct the cook to make the change. And how it's a much bigger soup bowl than the one on the menu... And on and on. Meanwhile, we're just telling her "Okay fine. It doesn't make sense to charge that much, but we'll just order something else."
In the first case above, the staff member risked making something of nothing, and obviously created a bad impression for any customers within earshot. In the second case, the staff member risked turning a mildly irritating bit of bad pricing decisions into an entirely bad customer experience.
In both cases, the defensiveness was completely unnecessary and was worse than the initial problem itself.
Subscribe to:
Posts (Atom)
