Social Media is Surprisingly Similar to Direct Marketing

Chris Brogan Blogchat with Mack CollierWhen Twitter rockstar Chris Brogan (@ChrisBrogan) agreed to host #blogchat, expectations were high. Mack Collier (@MackCollier), the creator of the successful Sunday night chat, used the proven direct marketing tactic of building anticipation to promote Chris’ appearance.

Mack started in July with a few “you aren’t going to believe what’s coming” tweets. He followed them with a blog post on July 26 announcing the August schedule, delivering on the promise. Marketing maven Beth Harte (@BethHarte) kicked the month off with her integrated marketing expertise, followed two weeks later by Brogan.

What happens when a Twitter celebrity with 150K followers hosts a chat?

Is it different from a direct marketing campaign? A good response for a promotion going to prospects from a rented list is 0.5% – 1.5%. When added to the customer results, successful campaigns pull 2.0% – 8.0% overall depending on products and promotions.

Brogan is the first blogchat host with a six-figure list of followers. If the response from his community followed typical direct marketing results there would be 3,000 – 12,000 participants!

A typical blogchat has 300-600 contributors and 2,000 – 3,000 tweets. The peak volume was 4,000+ tweets. If only 1% of the @chrisbrogan community participated, Mack was looking at tripling his peak participation.

The organized chaos of blogchat is part of its appeal. Quite often, there are multiple conversations going on simultaneously. In an effort to manage the expected overflow, Mack posted some participation guidelines for the hour that Chris was hosting.

The guidelines requested that everyone stay focused on Brogan’s topics during the first hour. Any off-topic discussion should be saved for after the main event.

The numbers are surprisingly similar to a direct marketing campaign.

Last week’s blogchat without a celebrity host had 498 participants and 2,700 tweets.

blogchat participation

(Please note that the source is It compiles data for seven days. The number of participants and tweets shown includes seven days.)

During the hour that Brogan hosted, the comparable numbers were 875 participants and 4,849 tweets.

Chris Brogan Blogchat Participation

Unless we compare transcripts, it is impossible to tell who is a regular blogchatter and who came to the party from Brogan’s community, so let’s look at different options.

If everyone who participated came from Brogan’s community, it is a response rate of 0.58%. This is a number that usually represents a rented list in a direct marketing campaign. It seems low for an engaged fan base.

If the 498 from the first week were there as regular blogchatters, only 0.2% of Brogan’s community showed up. Most direct marketers hesitate before renting those lists again.

Most likely, the real number is somewhere in the middle.

Several arguments can be made about timing, topic, and lurkers, but they don’t answer the question that keeps me awake at night.

How do you get people more actively engaged?

In a quest to find that answer, we routinely run activity analytics on our clients (and their competitors) followers. We check their community to see if the members are active, resting, or MIA. If they aren’t showing up to the party, how can you encourage them to play?

Here are the results from Brogan’s community:

Size: 148,896

Active (Tweeted in the previous seven days): 81,893 or 55%

Resting (Tweeted between 8 and 29 days): 19,356 or 13%

MIA (Hasn’t tweeted in 30 days or more): 47,647 or 32%

Drilling further down, 29,779 community members had tweeted in the last 24 hours. We consider those ultra active. If we presume (I know, I know!) that only the ultra active received the promotional messages tweeted by Brogan and Mack, the response rate range is 1.7% – 2.9%.

Does this mean that a company has to have thousands of followers to get a response?

No, it’s still like direct marketing. It’s all in the quality of the list.


  • Check your community member activity on a regular basis. If you’re seeing a decline in participation, your active members may be declining.
  • Don’t presume that your competitors are winning the social media game because they have more community members. Their participants may be inactive or uninterested.
  • Social media isn’t transparent. You have to dig through the data to find your answers.
  • Go for the quality. The best community begins with your customers. Invite them to participate.

31 Ways to Supercharge Your Email Marketing

How to Get More Customers, Sales, and Profits without Spending a Fortune

The guide is filled with actionable tips that increase sales, improve retention, and reduce costs. We are so confident they work we offer a 365 day money back guarantee.

Learn More

Comments on this entry are closed.

  • Mack Collier

    Hi Debra, very interesting stats and analysis.

    A couple of thoughts:

    1 – My guess is that some (a lot, maybe?) people simply didn’t participate in #blogchat last nite because they thought it would be TOO busy. I actually saw a few people tweeting this. So while the ‘organized chaos’ approach likely attracted some people, it probably discouraged some people from participating. And I hate that.

    2 – There are advantages to having both a small AND large Twitter chat. If the chat is smaller, that means you can more easily connect with your community, and it probably means you can dive down and have ‘meatier’ conversations. If the chat is larger like #blogchat, it means the volume can simply be too high for some people, and scare them off. BUT, that high volume also attracts top co-hosts, that greatly improve the quality of the chats.

    As you know, I am a stats junkie, so I love this stuff, thanks SO much for taking the time to put this together, Debra!
    .-= Mack Collier´s last blog ..Here’s why you shouldn’t treat social media as being ‘free’ =-.

  • Debra Ellis

    Hi Mack,

    First of all, thank you for the work you put into making blogchat a good place for sharing information.

    In response to your thoughts:

    I saw the tweets of dissatisfaction too. It makes me wonder if making such a major change has a negative effect on your brand.

    You suggested that people have a “regular” blogchat after the Brogan appearance. Apparently people followed your suggestion. I pulled the numbers two hours after Brogan left to see what happened. 150 new people joined in after Brogan left and there were another 1,180 tweets.

    Maybe you could have the best of both by having a “Special Blogchat Event” at a different date and time. You change the schedule for holidays, so maybe once a month you could have a headliner event.

    You’re very welcome.

  • Amy Africa

    Debra, Debra, Debra. I adore you but what the hell kind of fuzzy math is this?

    I’ll agree with you that Mack used direct marketing techniques to promote #blogchat — although if he did the same thing via e-mail, we all know it’d be considered SPAM — but I don’t agree at all with the results as you’ve outlined them. And when I say — I don’t agree at all, I mean “What kind of crack are you smoking?”

    In direct response, we typically have ONE SHOT to get it right with the list. You rent the list, you mail your catalog/direct marketing piece and you wait for the results.

    Do you know how many times @mackcollier promoted #blogchat to his 20,000+ followers? It was insane.

    And that was just Mack. Chris did it. His followers did it. Mack’s followers did it. Other #blogchat people did it. Hell, you even did it. Look at the transcript — dozens of people plugging the Brogan #blogchat with hundreds of thousands of views.

    This, coupled with the fact, that a lot of the folks at #blogchat were regulars means that even LESS than .02% of Brogan’s audience showed up.

    In the direct marketing world, you’d be fired for that kind of performance.

    Even if you do the math using only the “active” tweeters, you still won’t get the numbers you mentioned above. (1.7%-2.9%) NOT. EVEN. CLOSE.

    I get the value of social media. I really do. I’ve been recommending things like User Reviews before Amazon had them — and I also think it’s important to get the facts straight.

    Yes, new folks were brought to #blogchat.

    Yes, #blogchat got increased exposure over the past week.

    Yes, having Brogan as a guest host was LIKELY a good thing for #blogchat overall.

    And yes, if this were direct marketing, and not social media, you’d likely be polishing off your resume for last night’s performance.

    P.S. Brogan, as co-host, dropped affiliate links throughout his presentation. How did that impact the #blogchat community? Regulars, who attend week after week, like @casudi, went into #UNTWEET mode. WTF? Untweet mode because Brogan was there? Why did the community allow that — she’s been a consistent contributor for over a year for goodness sake!

  • Debra Ellis

    Well Amy, what can I say? I’ve been thinking about entering the political arena and am working on my math skills. My tenure as an engineer and direct marketer has conditioned me to a high level of accuracy. If I’m going follow Meg and Carly over to the dark side, I need to polish my fuzzy math skills.

    You’re right about so many points. I’ll begin with direct marketers having ONE SHOT to get it right. This means that we have to do our work up front or lose the one opportunity we have to convert a prospect into a customer.

    What if we treated social media the same way? There’s a lot of talk about “like minds”, but it isn’t followed with appropriate action. (Suddenly I feel like channeling Toby Keith and singing “A Little Less Talk and A Lot More Action.”)

    What if, instead of amassing thousands of followers with little or no interest in our brands, we focus on the people who really matter? You know the ones I’m talking about – we used to call them “customers”. Sometimes we called them “friends”. We never called them “advocates”, “influencers”, or “community members.”

    What if we combined the best of direct marketing with the best of social media? We could finally fulfill the promise of one-to-one marketing.

    Thank you for commenting. Your insight is always on target.

    I missed the affiliate links and had to go look. Since this is a post on statistics…

    Brogan had 79 blogchat tweets. Two or three were affiliate links. (I’m not sure about the Flowtown tweet.) The bottom line is that 2.5% – 3.8% of the “sharing” was self-promotional. Isn’t that against the social media rules?

    It saddens me that regular contributors like @casudi opted out. They are what make blogchat a special place to be on Sunday night.

  • Z. Kelly Queijo

    Debra, I did not “do the math” to see if I agreed with your comparison of last night’s #blogchat to a successful direct mail campaign and here’s why — I know that my own level of participation was likely not on the radar to be counted and there were probably others like me who followed the chat.

    I submitted one tweet prior to the start of the chat and without using hashtag, invited a friend to join in. Then I read Mack’s guidelines and I took them to heart, therefore I did not greet @CASUDI when I saw her tweet or engage with any of the people I look forward to chatting with during #blogchat. As host of my own chat and an avid fan of #blogchat, I know how challenging it is to manage “organized chaos” so I “listened” to @chrisbrogan rather than join the conversation. My level of engagement was perhaps one-sided as I chose to “favorite” various tweets to review later rather than retweet as I normally would. Now I’ll follow up and scan the transcript and the twitter streams of those who I missed communicating with last night which is bound to “count” but maybe not in a trackable way.
    .-= Z. Kelly Queijo´s last blog ..Virginia Public Colleges &amp Universities =-.

  • Debra Ellis

    Hi Kelly,

    Thank you for commenting. If you didn’t include the hashtag, you weren’t included in the count. I’m sure there are others who watched without participating. Listeners are very important. Hopefully they’ll become participants.

    This is also similar to direct marketing. Sometimes people have to receive multiple promotions before they are moved to action.

    Both you and Mack do a good job with your chats.

  • Mack Collier

    One thing I’ve noticed over the past couple of months with #blogchat is that more and more people are saying ‘Yeah, I am going to just wait and check out the transcript tomorrow’. I’ve also noticed that people that live in Europe and later time zones have started to figure out that we also publish the transcript so they are coming and getting it on Monday morning when they wake up.

    So tracking the # of contributors unfortunately does not take into account the lurkers, or the people that got the transcript later on, and opted to not try to participate in the live event.

    As for @casudi, she apparently chose to not participate because she felt she could get more out of the chat by listening and observing. Ultimately, that’s what I care about, that the people that DO interact with #blogchat do so in a way that’s the most valuable to THEM. I almost never participate in #IMCchat, but I often lurk, because that’s how *I* get the most value out of that chat.
    .-= Mack Collier´s last blog ..Here’s why you shouldn’t treat social media as being ‘free’ =-.

  • Mack Collier

    Kelly you just hit on one of the issues I am really running into with #blogchat. It’s growing so fast, how do I continue to make it welcoming for the ‘regulars’ and for people that are completely new to it, at the same time.

    I would hate for the regulars that helped make #blogchat what it is to feel like you couldn’t connect with each other. What I have been trying to do is encourage the regulars to come an hour early, if they can. Then we can all interact and meet and say hi. This is also the perfect time for everyone to share links to their site/blog. I had a lot of complaints about people doing this DURING #blogchat, so I wanted to encourage everyone to do this as well before the official chat gets started.

    I want to make sure that #blogchat is always seen as a welcoming place for regulars. I am really going to try to make a better effort to make sure that y’all have more time to interact before the chat starts, plus I want to catch up with y’all as well.
    .-= Mack Collier´s last blog ..Here’s why you shouldn’t treat social media as being ‘free’ =-.

  • Debra Ellis


    Thank you for providing additional insight.

  • Steven Woods

    thanks for putting together this data, it has obviously started an interesting discussion. I would, however, challenge your analysis. It’s not that it’s suffering from an apples-to-oranges problem, it’s suffering from an apples-to-apple-tree problem.

    You’ve created an equivalent between the “response” to a direct campaign (usually a click of some sort, very passive) with an interaction with a blogchat campaign by crafting and sending a tweet. Not at all a comparison that’s accurate.

    To compare the two (and this might be hard), one would have to compare either the views of all participants’ tweets, or at the very least, the clicks on any and all links shared.

    If the goal of a standard direct campaign is to drive a change in mindset or awareness, that is the only fair comparison. I’m sure that that analysis would show a very different comparison.

    Thanks for putting together the numbers to start the discussion.
    .-= Steven Woods´s last blog ..The Sales Team as a Content Testing Crucible =-.

  • Mack Collier

    BTW Debra and Amy, Chris actually disclosed the affiliate links he included, here is an example:

    I don’t know how many he did, but I heard someone else say it was 3. Personally, I have zero problem with a fraction of his tweets being disclosed affiliate links.

    Now if he was dropping undisclosed afflinks in #blogchat last nite, then yeah, that would be an obvious problem.

    BTW in complete disclosure, I found this out the hard way. I DMed Chris pointing out that he didn’t disclose the links and that he should in the future. He informed me that he DID, and he was exactly right. So I was the jerk for not checking for myself before I DMed him about it.
    .-= Mack Collier´s last blog ..Here’s why you shouldn’t treat social media as being ‘free’ =-.

  • Mack Collier

    Well I should probably clarify that previous comment as the context of the disclosed aff links is important too. Of the afflinks I found that Chris shared, they were disclosed, and in reply to a question that someone else had asked him. To me, that seems pretty harmless. But it’s an interesting topic to cover.
    .-= Mack Collier´s last blog ..Here’s why you shouldn’t treat social media as being ‘free’ =-.

  • Debra Ellis

    Hi Steven,

    Thank you for adding to the conversation. I like your apple to apple tree comparison.

    Yes, there are different actions involved, but I’m comparing the ability to move people to action, not passive clicks versus composing tweets.

    It is actually much harder to get people to respond to a direct marketing campaign (which BTW include clicks, mail, and calls) than to create an interaction in social media. The participants are more vested on the direct marketing side because they have to part with their money to participate.

    If we look at it that way, then the response for blogchat should be much higher than direct marketing.

    There is a long tail in both direct marketing and social media. Our conversation is increasing the blogchat tail as we type. Hopefully it will bring more people to blogchat.

    Thank you for your kind words and participating.

  • Amy Africa

    Actually I think one of the things that impressed me most about Brogan was that he used affiliate links. It was refreshing to see a “social media rockstar” or whatever it is that #yousocialmediapeoplewhokillme actually sell something and not apologize for it. My P.S. was not that he should not have used affiliate links, it was that he did use them and that was the FIRST time I’ve ever seen them used in #blogchat.

    My issue with this post is that Debra’s numbers are completely inflated and that if it was a direct mail campaign, the person who designed it would have been hung, quartered, lynched, shot, tarred, feathered and whatever else people who actually use metrics that matter do to torture those who don’t.
    .-= Amy Africa´s last blog ..What Can You Learn From a Guy Who Cleans Roofs =-.

  • Amy Africa

    Oh and that apple-tree analogy was brilliant. Kudos to Steven Woods. I will credit you when I steal it, ok?
    .-= Amy Africa´s last blog ..What Can You Learn From a Guy Who Cleans Roofs =-.

  • Debra Ellis


    The reason I questioned the affiliate links weren’t because of the disclosure. I thought chats were supposed to be a promotion free zone.

    Everyone who follows me knows that I promote my products and services. I also promote others, freely share my knowledge, and provide my email address for additional contact.

    But, not in chats.

    Where do you draw the line for promotional tweets? Can the participants post affiliate links? They contribute valuable information every week.

    I’ve noticed that people often apologize in blogchat when they tweet links to posts that expand on the topic.

    Allowing affiliate links is a slippery slope…

  • Debra Ellis


    If it were a direct marketing campaign we could include profits #theonlynumberthatreallymatters in our metrics ; )

    It’s interesting that you have me over inflating the numbers and Steven has me under reporting them. It’s nice having the different perspectives.

    For the record, as a metrics that matter person, I prefer to use boiling oil when someone messes with my direct marketing numbers.

  • Lois Geller

    Somehow reading this discussion string..I’m lost in the apple to apple tree analogy. Why Amy wants to use it again is beyond me.

    Frankly I was surprised last night when I clicked on the flowtown tweet, which I assume is an affiliate of Chris’s.

    If he was going to put that in there, and we’re talking direct marketing techniques….where was the offer of the first month Free?

    It would have been an interesting test (for me, at least) to see how many people he would have actually signed up right on the spot.

    The funny thing about all those people on Twitter is that I rarely know what any of them are selling, except for the Trump. Trump folks come right out and ask you to do something.

    We direct marketers might want to try that also. Forget asking people to RT your tweets. Ask them to call you for whatever it is your selling. If they respond and buy ..then we can categorize it as a direct marketing vehicle.

    Oh, but we don’t dare sell on it. We need to build relationships over time. We do that in direct mail, and we always ask for the order in every effort.

    Not sure why I’m off on this tangent, except that I got a call today from someone who has a zillion Twitter followers. He said that he’s dying because he has no business.

    I told him I’d help him craft a direct mail letter to his target audience and it should work. Less Twitter time and more “back to business” efforts!

  • Debra Ellis


    Thank you for stopping by. I love your tangent. We could also start answering the phones with “Good morning, I had the best breakfast on the way to work. It was a rough night because of the neighbors, but I’m at work now. Are you still listening?”

    Seriously, it’s always a pleasure to have your comments on my blog. Come by any time.

  • Barbara Bix

    I reached a similar conclusion after tallying the responses to TopRank’s 2010 Digital Marketing Poll of their 6000+ followers. View that analysis here
    .-= Barbara Bix´s last blog ..Target Marketing- Still essential- still evolving =-.

  • Debra Ellis

    Hi Barbara,

    Thank you for visiting and sharing your post. Your information was interesting.

  • Kristof

    Hi Debra –

    WTHashtag is a nice app but, for more exact numbers, I reco TweetReports. Here’s a blogchat transcript with stats (there were 5160 tweets) and a complete list of all contributors with their tweet counts

    Also, here’s a transcript of just Chris Brogan tweets from that chat

  • Debra Ellis

    Hi Kristof,

    Thank you for sharing the information about TweetReports. It’ll provide additional analytics for future tests. I like how it breaks out retweets and @’s.

    The challenge I have with both tools is the inability to select by date AND time. I pulled the screen shot with the 4,849 tweets right after Chris signed off. It included the tweets building up to the chat. It would be nice if there was the ability to pull tweets by the time the chat was actively going and filter out the retweets. I think we would have a very different view of the results.

    Thank you for commenting and sharing an alternative source for analytics.

Real Time Web Analytics