TechCrunch basically made the point yesterday: TwittAds is a bad idea. I wanted to add my two cents from a the perspective of someone who spends a lot of time asking himself “would my company want to buy ads on this site?”
The point of TwittAds is thus: put advertising on your personal Twitter page and hope that your followers click on your Twitter ID, view the ads on your Twitter page, and hopefully execute some monetizable action (like clicking on the ad.) This advertising model indicates a fundamental misunderstanding of advertising from a buyer’s point-of-view.
Seth Godin says thus about advertising in his influential book, Permission Marketing:
From an advertiser’s point of view, the single most important tactic is frequency. Frequency is a simple concept: How many times is your ad presented to a single individual? In practice, though, frequency can create a number of pitfalls.
When advertising agencies measure their campaigns, they look at reach and frequency. Reach is a fairly simple metric. How many different people were exposed to the ad? Frequency, as we’ll see in a moment, involves some more artful measurements.
Seth’s wisdom is still relevant ten years later. Twitter is weak on reach to begin with; the Twitterers with the most followers, like Mr. Scoble, have around 34,000 followers. Assuming that the TwittAd advertisers are trying to target the followers of influential Twitterers running an ad on a single Twitter page, even for someone like Scoble, has insufficient reach to even merit the attention of modest advertisers. The obvious solution to that problem is to place the same ad across a number of Twitter home pages, which I will dissect in a moment.
On the surface Twitter appears to be a perfect mechanism for frequency, at least until you look under the hood. Let’s look at a sales funnel I cooked up:
So where did I get these figures? I made a reasonable estimate based on contextual advertising performance for the final Twitter home page to Advertising Landing Page CTR rate, and the first two rates are determined by my own personal experience with Twitter (I actually read about 1 in every 100 Tweets that I am subscribed to.) I’m not trying to perform exact science here, I’m trying to make a point about how pitifully small a reach of 34,000 is.
So here’s a formula for how we determine the CTR for a particular Twitterer:
Advertising Clicks per Day = Audience * (Number of Tweets Per Day * Average Number of Tweets Actually Noticed by Followers) * CTR from Noticed Tweets to Twitterer’s Home Page * CTR from Twitterer’s Home Page to Advertisment
Let’s plug in some values for Robert Scoble, who is the best-case advertiser that for TwittAd:
Audience 33,482 Avg # of Tweets Per Day 100 Tweets Noticed by Followers 0.01 CTR Tweets to HP ~0.01 CTR HP to Ad 0.01 If you do the math using the numbers that I’ve supplied you learn the Robert Scoble would generate between 3-4 clicks on an advertisement per day. Pitiful.
As Erick Shoenfeld pointed out in the original TechCrunch, the home pages of specific Twitterers are simply not a popular destinations for Twitter users. What Erick didn’t mention is the huge noise problem with Twitter - even if Scoble sent out 100 tweets a day no one reads all of them; in fact, most of them are not read at a level beyond mere skimming.
The other problem with advertising on Twitter is the lack of context of the advertisements - how in the hell do you make sure that the right Twitterers with the right messages become advertisers for the right products? It seems like an impossible challenge to me. Anyone who wants to advertise on Twitter pages is going to have to advertise on more than one, period, and how you do that with any level of appropriate audience discrimination seems infeasible. I guess the philosophy of TwittAd is to take a handful of ads and broadcast it to one amorphous, untargeted audience and hope that the combination of “frequency and reach” can yield enough sales for advertisers to overcome the huge waste of marketing dollars spent on the uninterested portions of the audience.
This is just Web 1.0 advertising with Web 2.0 spoilers - indiscriminate broadcast advertising is ineffective and smart advertisers avoid it. This is the principal reason why the original Web 1.0 bubble crashed - the failure of the indiscriminate broadcast advertising model. Once again Web 2.0 entrepreneurs disappoint me by failing to learn the lessons taught in the first crash.
If you enjoyed this post, make sure you subscribe to my RSS feed!
Hello-
You provide a very well thought out advertising scenario. As the CEO of Twittad I would just like to give my take on things if I can.
Twittad knows that the background images on a Twitter users profiles are not clickable. Intelligent Twitter users know the same. Hence why you see the easy entry point prices being posted on Twittad. But I see Twitter evolving. My speculation is that if Twitter wants to gain more than just a simple 140 character business they will begin to develop their users pages more.
Allowing the users to upload applications, provide more than one simple link to a blog, the ability to upload photos, etc. This could be where Twitter is headed. Increasing the time a user spends on Twitter and having users viewing a profile page will increase page views, the time spent on the web version of Twitter, and bring in advertising dollars.
In addition to an Ad being seen on a profile background, when a user clicks on another user to send a Direct Message (DM) that users profile background image (with the Ad) is displayed. So one more spot for advertisers to gain a “eye”. If and advertiser wants to track effectiveness they could do so by displaying coupon codes and a different trackable URL, etc in their Ad.
I think you bring up some very good points, and they are all ones that were in our business model at the very beginning. It is our view that Twitter is in version 1.0. There are going to be many more versions of Twitter, and based on the amount of Twitter users signing up on Twittad..over 430 total as of today (9/3) people are open to this concept.
We are working on a lot of development and in the next few weeks we will roll out the second version of Twittad that will more effectively tie advertisers to Twitter users worldwide.
Regards,
James Eliason
President/CEO Twittad.com
James,
Thanks for responding to my comment. I totally overlooked the fact that those background ads are not clickable. Rather than provide you a full reply now I’m going to let the article air out for a bit before I get back to you.
Ok James, as promised here’s my reply.
Although I don’t think you intended to call my an unintelligent Twitter user, let me assure you that I use the service regularly. I totally forgot that you can’t even click on the background of the ads, and in my opinion that’s just another area of conversion that’s going to be added to the sales funnel. Based on the sample ads that I saw via TechCrunch I suppose the idea is to get Twitters to follow a “sponsored” Twitter account, i.e. one that belongs to a corporate blog and so forth. That idea has some merit to it - it uses basic permission marketing to get people to opt-in for updates from a service and I’m all about that.
However, how do you gauge performance for TwittAd then without CTR metrics? If your advertisers use TwittAd as the only method of promotion for a said Twitter account then measuring the total number of subscribers acquired makes sense as a metric, but what if they promote their account through other means, such as on the corporate blog itself per my earlier example?
I still think the idea is poor given the sales funnel - there’s just too much loss at any given step and the reach isn’t broad enough. However, I like your ideas about designing TwittAd to move alongside the evolution of Twitter and given that I left my crystal ball at home this morning I can’t speak to what Twitter may do and how it will be received by its user base.
Even if I don’t think the idea is good I still wish you the best of luck. Go get em!
Thanks for the follow up and I definitely was not calling you a unintelligent Twitter user. I have read many of your posts and have read Marketing Ninja before, all of which I have enjoyed reading.
There are a lot of things we are working on that I cant mention at this time with regards to ad performance on Twitter that comes from Twittad. But I will tell you that the #1 problem in our business plan is what everyone is talking about, the issue of not being able to click on a ad and gain the measure for the advertiser.
I will say that the Twitter community is a very intelligent and active audience. If a advertiser buys their profile I would bet many would promote that ad or service through their tweets or on their blog. Providing success for the advertiser not only comes from the views on the ad, but also the attention that comes from the purchase. Twitter users can mention the advertiser through Tweets and blog posts on the Twitter users website. To that degree, I will say we are working on something to solve this.
Again, thanks for the post and stay tuned.
James Eliason
President/CEO
Twittad.com