Monthly Archives: June 2010

Does a name change lose you followers?

UPDATE, June 30: This is really, really interesting: Our old account @Twitter_Tips has GAINED almost the same amount our new account @TweetSmarter has lost—1753 followers as of early June 30. And by “lost” I mean the difference between what our followers would usually have been and what they are (see chart below).

This is a unique situation, because Twitter did us a special favor: They let our old account name continue to exist. All that is there, however, is a tweet saying:

Our name has changed! @Twitter_Tips is now @TweetSmarter. Learn more: http://j.mp/TweetSmarter

So this means that people are following our old account by the hundreds a day without ever checking what it is. People are still continuing to unfollow us as they become confused by the name change, however, so I’ve tweeted a few more times about our name change.

UPDATE, July 11: Our account is gaining followers again (chart below) so it looks like the people following again comfortably outnumber the people unfollowing:

Here’s the original post:


I knew some people would be confused by our name change. I thought I would share some of our experience for anyone considering a similar change. (Why we changed.)

There are a couple of issues in figuring out how much it affects our follower numbers. We get unfollowed and followed a lot every day no matter what. The follow/unfollow numbers have to be taken together to make an educated guess at how much the name change is affecting things.

What’s actually happened to our follower numbers as of June 27:

So, how can we make a guess at what effect the name change has on our followers? With an account as big as ours (nearly 200,000 followers), calculating unfollowers is difficult:

  1. Some will unfollow out of confusion, wondering why this account with a name they never followed before is appearing in their stream.
  2. Others may think the name change means that the tweets will change—we’re not going to do anything differently!—and unfollow for that reason.

And a very important question is: When will the name change stop affecting our unfollowing number? I think it will take the better part of a month for most everyone to take whatever action they are going to take because of our name change.

Will a name change cause fewer people to follow?

Secondly, we have to figure out if we are getting fewer followers because of the name change. Some followers undoubtably follow a new account in part because of it’s name. If our new name was, for example, “BingoStatistics” I would expect fewer new followers :) Of course, a name change could cause you to get MORE new followers.

While there are many accounts that grow at a similar rate to ours that do NOT have “Twitter” in their username, I think in our case we can reasonably expect to get fewer new followers because of not having “Twitter” in our username.

What is the total effect?

I don’t think it’s possible to project the long-term effect in just a few days. It looks like we’ll lose somewhere in the range of <1% of our followers because of the name change though. I’ll update this post in a week or two when it looks like the changes are stabilizing

Some Twitter spammers got caught out by our name change

UPDATE: Twitter has finally informed me they are looking into this issue:


If you see anyone’s tweets beginning with “Tweetsmarter: ” please report them for spam.

Do NOT report someone simple for including @Tweetsmarter. It’s only tweets that do NOT use the “@” and start with exactly the username followed by “: ” that are automatically reposting tweets.

This will NOT get them automatically blocked from or search or suspended. It is merely a notification to Twitter that they are engaging in one of the practices that Twitter defines as spam: They are automatically reposting our tweets as their own. Including our conversations.

Chats being echoed by retweeting bots

You’ll be helping users that we chat with in particular, which gets really annoying to people we are trying to talk to when everything we say is repeated to them several times by a variety of accounts..

Meaning if we send you a message, suddenly our tweet to you will appear again and again from several accounts. Thanks to @gilman_gal for suffering through a conversation with us and seeing everything we said to her repeated by many other accounts over and over to bring this to our attention.

Many of them became visible suddenly because they had been smart enough to automatically remove “Twitter_Tips: ” from their tweets, but haven’t noticed the name change to TweetSmarter yet :) Some of them have already caught on and adjusted their rss-to-twitter feed string regex to remove our username, but for a little while, several are exposed.

What makes this spam?

Here’s where and how Twitter defines automatically retweeting others’ tweets as your own as spam, saying ”Some of the factors that we take into account when determining what conduct is considered to be spamming are:

Will this work?

If by work, you mean will these users be blocked from or search or suspended, the answer is…maybe. Twitter takes a number of factors into account whether to review and take action on an account. The number of people reporting an account for spam is one of those factors.

If they are blocked or suspended, they can contact Twitter and be unblocked from search or reinstated. I’m all for educating people about the right way to use Twitter! But I’ve tried communicating with these kinds of accounts in the past, and they often don’t read tweets to them. They are mainly or completely automated. So the only way to educate them is for Twitter to take action. Then they contact Twitter to ask what happened and can learn.

Can’t Twitter do this?

Twitter relies partly on us in some cases to help determine what is and isn’t spam. They are getting better and better all the time at identifying accounts that are completely spam. It’s the automated accounts that are only mostly or partly spam where our help makes a bigger difference.

Over 5 MILLION people respond to Twitter ad…in just 24 hours

UPDATE: June 1, 2011—

  • A Volvo ad had 50% engagement rate.
  • A Radio Shack promotion ran for one day on Twitter. Next three days, instore exchanges and purchases, up double-digits from day before ad ran. And ad didn’t run anywhere else.

The Next Web reports that the second ever Promoted Tweet had 6% engagement on 85 million impressions.

Whoa.

That’s over 5 million people engaging from just one tweet.

Did Twitter just became the gold standard for online advertising?

Even at a more typical .02% response to online advertising, that would represent 170,000 people engaging. So, while it’s too early to say for sure, Twitter appears headed into the stratosphere of the online advertising world. I can’t decide if the response rate or the total engagement is more impressive.

The advertiser, Coke, saw these results in 24 hours, according to Yahoo Finance and the Financial Times, from this tweet:

Captured by UK Tweeter Neville Hobson/@jangles for TheNextWeb

Of course, the US and UK have huge populations on Twitter, and this tweet ran during the US vs England World Cup match. While Coke is clearly not a newcomer to recognizing good advertising opportunities, Promoted tweets seem to have a potential never seen before in online advertising.

Best of all: You can ignore them

But you know what I like best? Promoted Tweets are minimally intrusive. Easy to ignore, but still apparently great for advertisers.

What do you think?

  1. Do you mind Promoted Tweets?
  2. Would you respond to a Promoted Tweet? Did you respond to the Coke ad?
  3. Is this a good way for Twitter to pay its bills while keeping the user experience positive for it’s users?

UPDATE: @DaveWiner points out in fact, that this is really, really hard to believe. Good point. Wouldn’t be surprised to hear some backtracking on this later. But for now, just…wow.

New Record: Twitter Reaches Peak Rate of Nearly 200,000 Tweets Per Minute

On June 24 during the Japan vs Denmark World Cup group match, Twitter reports tweets were posted at a peak rate of 3,283 tweets per second (TPS). As The Next Web points out, that translates to a rate of 196,980/minute.

Think about that: if sustained, that would be a rate of nearly a million tweets every five minutes.

Thanks to sporting events (mainly the Word Cup), Twitter has set five all-time records for TPS on different dates in June. Here are the records…so far:

Left column is date in June a TPS record was set: the 11th, 14th, 17th and 24th

An even more important record

But: who cares? Until recently, this would have represented nothing more than a drain on Twitter’s resources. The heavy traffic caused a lot of Fail Whales for users around the world. Twitter even went completely offline.

So the equally amazing and likely more important record was set during the US vs England World Cup match on June 12. A Promoted Tweet posted by Coke is claimed to have been responded to by over 5 million people engaging on 85 million impressions.

This was either the greatest advertising coup of all time…or Twitter is about to turn the world of online advertising upside down.

Considering that World Cup viewership is increasing towards the finals, I predict a few more records for Twitter before all is said and done.

Why @Twitter_Tips has become @TweetSmarter

UPDATE 1: Wow!!! Your outpouring of love through tweets, DMs, emails and comments has been deeply touching *wipes tear from eye* Thanks to everyone for being so supportive! (The change has had a few stressful twists and turns, but everything important has gone smoothly.)

UPDATE 2: OMG, even more messages than we realized! We partially misconfigured after the name change and missed a TON of tweets. Apologies to not getting back to everyone personally—it’s just too much for us right now.

UPDATE 3: Our old account, even though the only thing there is a message telling people to follow our new account, has been getting thousands and thousands of followers. Here’s more of the story on that.

In a time long, long ago, Twitter changed their policy to not allow the word “Twitter” in usernames. And then…

A few months after the policy change, a newish account came to the attention of a Twitter employee. It was our just a few months-old account, @Twitter_Tips! It seems having “Twitter” in the username was causing minor technical (engineering) problems on the account. It had to do with how they engineered the interface to prevent accounts from having “Twitter” in their name.

We offered to change our username

At that time we offered to change our username (we figured Twitter would ask us eventually). There was no response to our offer. We created the @TweetSmarter account as a placeholder so no one else could take the name and contemplated what to do. Since Twitter had not taken us up on our offer to change our name, nor asked us to, we decided to keep it, willing to change at any time Twitter should ask.

Of course, the longer we waited, the more people became familiar with us as @Twitter_Tips. But there was always the hope we could keep the name. All we ever did from @Twitter_Tips was send out news, tips and status about Twitter, and answer questions for users. We didn’t even have a blog for a very long time. We still don’t have anything to sell :)

The dreaded password reset

Then one day Twitter reset our password, along with those of several other large users. (Because of some hackers attacking Twitter.) We were unable to reset the password because of the old problem with having “Twitter” in the username. In fact, Twitter told us, not even they could reset our password (it would have taken some custom engineering that had never been done for anyone before). We were locked out of our account from the web interface. Fortunately, we were able to remain logged in via the wonderful HootSuite and keep tweeting and answering questions for people.

Twitter decides it is time: Our username must change

All the changes happened automatically. (We kept the Avatar the same to minimize confusion.)

What happened was, having come back to the attention of Twitter, they decided it was time for us to change our username. They told us how much they loved what we did, and offered all kinds of help. They even offered to set up a temporary @Twitter_Tips web page to direct people to the new @TweetSmarter account. (We said yes.)

What happens when your username changes?

It used to cause a variety of problems, but no longer. The name simply changes. All your followers find themselves following someone by a new name (which can admittedly be confusing!) You remain on all the same lists, and your old tweet URLs stay the same. We have decided to take advantage of the name change by creating a Twitter educational site at TweetSmarter.com. We have a cool infographic there to give you a taste of things to come :)

Please help us by letting people know!

Because of the confusion, we would really appreciate if you get the word out by tweeting this blog post so people can know what happened. Here’s a picture of the two of us (although actually, Dave has a beard now, lol):

Volunteering
Love,
Dave and Sarah Larson

Should You Auto-Follow Everyone That Follows You?

Update: On September 2, it finally got to Chris, and he unfollowed everyone to try to shake off all the spammers he had autofollowed. He also said he would quit Twitter if they didn’t like that he had auto-unfollowed everyone.

Brogan’s: “Power Twitter Tip #2:

He recommends: “Follow anyone who follows you (and unfollow spammers/jerks).”

✔  Tip: This is usually done by turning on “auto-following” in a third party Twitter application such as SocialToo, because if you’re going to follow anyone and everyone, doing it manually is inefficient.

I asked Chris what his reasoning was. He said “Benefit to me is that I get several hundred followers a day. It’s a full time job if I want to do it by hand.” I asked if he unfollows both manually and automatically, he said “Correct. I unfollow mostly manually, and then let Socialtoo.com take care of the rest.“ Have to disagree with Chris on this one.

Update: I asked Chris some follow-up but didn’t hear back. @SherryinAL points out that maybe Chris meant that manually unfollowing saved him time vs. manually following because he gets so many followers per day. I still think this isn’t a situation most Twitter users find themselves in. I’ve responded further in the comments.

Sometimes users with a lot of followers give advice that doesn’t work as well for users with not so many followers. This seems like one of those times. Autofollowing is kind of the land mine of tips for using Twitter. Potential problems include:

1. Auto-following everyone makes you a target for spammers

Spammers make lists of accounts that auto-follow, so auto-following everyone makes you a target for spammers …and your stream gets cluttered with spam. Even Chris has run into autofollowing problems. So, while it seems like checking out people before you follow them takes more time, becoming a target for spammers by auto-following them can end up costing you more time (and more spam) in the long run. However, like everything else on Twitter, your mileage may vary. Always remember Twitter Rule #1

.

"Twitter In Real Life" by HubSpot

2. You have to either manually follow, or manually unfollow

While both are work, autofollowing and manually unfollowing attracts spammers, and makes work for yourself. Chris himself wrote how he had to suspend autofollowing some time back because he was getting too many spammy DMs. I think any advice to autofollow needs at least an asterisk on the reason for doing so and the dangers of doing so.

3. You don’t need to auto-follow to get a lot of followers

Whether Chris autofollows for this reason or not, it’s important to consider whether this is something users should consider doing to get more followers. Users ask me frequently if they can grow their accounts without doing things like autofollowing, or following lots of other users. Of course, Chris has been a HUGE name in Social Media for years and recently had a bestselling book. So his results are difficult to use to gauge how well his methods help the average user.

We have a lot of followers because we focus relentlessly every single day on sharing and writing good tweets and helping people on Twitter. That creates a lot of friends, and a lot of retweets. Chris has a lot of followers because he provides similar value. I just want to point out that you don’t have to autofollow to get followers! We actually have 50,000 more followers than Chris does and we do NOT autofollow. (Stat comparison.)

Twitter in real life, by Hubspot

So to get followers, you don’t need to:

  1. Auto-follow;
  2. Spend money advertising your Twitter account;
  3. Trade tweets—you tweet me, I tweet you;
  4. Follow more people than follow you hoping many will follow you back;
  5. Tweet things just because they are popular or likely to get retweeted, i.e @GuyKawasaki‘s strategy.

None of things are necessarily bad in all cases, I just want to point out that they are not necessary. As of June, 2010, we do NONE of these things. When users with a lot of followers give advice, often what they say is take as gospel: you must do these things to do well on Twitter. I just want to say it ain’t necessarily so :)

✔  Tip: To see what’s currently popular on Twitter, check out the TwitterSphere. To search what’s popular, I recommend Topsy.

So what should YOU do?  All I can say is, remember Twitter Rule #1 :) Also, here’s a related view on auto-following from the very helpful Fernando Fonseca (@fjfonseca on Twitter)

Stat comparison

Stats current as of 6/18/2010 8:25:41 AM CST.

@ChrisBrogan’s account:

  1. FOLLOWERS: 140,841
  2. Following: 128,346
  3. Joined: 24 October 2006
  4. Listed: 12,576
  5. Retweets received: 104,754

Our account:

  1. FOLLOWERS: 192,636
  2. Following: 126,996
  3. Joined: 22 August 2008
  4. Listed 9,372
  5. Retweets received: 523,702

Compare last three months of Followers/Following/Tweets at this Twittercounter link.

✔  Tip: To see when someone joined Twitter, check out the When did you join Twitter? tool. To see who’s getting retweets, enter their username into the Chirrps cool profile search tool.

Twitter rule #1

Twitter the company does have rules and suggested best practices you should follow. But beyond that, what’s true is:

Twitter rule #1: Make your own rules.

(c) Jennifer Berman, Humerus Cartoons

If someone tells you you’re doing Twitter wrong, feel free to ignore them.

Did you know that many of Twitter’s features were invented by users? @msgs, #hashtags and retweets were all invented by users.

Part of Twitter’s magic is that you create your own experience! (You might want to read “Quit worrying about people that don’t want to follow you.”) There are an infinite number of ways to use Twitter. But because this is true, also remember…

Twitter rule #2: Be open to learning.

There are a lot of great ways to get more out of Twitter that you can learn from others.

If someone gives you Twitter advice, feel free to think about it. Ask questions. Investigate. Don’t take what you know for granted. Part of Twitter’s magic is that there are an infinite number of new things to learn about how it can be used.

How to tell if an article about Twitter gives bad advice: It relies on one point of view. When someone says “Twitter is about this one thing” they are automatically wrong.

Twitter is about a lot of things, not just any one thing. But because this is true, remember…

Twitter rule #3: Engage with and help others.

Once you learn how powerful connecting with and helping others is on Twitter, you can forget all the other rules :)

Tweets from the #WSEC2012 Conference


Here are Tweets with the #WSEC2012 Conference hashtag only.
#WSEC2012 Hashtag Tweets:
Note: Many tweets may be missing from this list (use the above list instead) because many of the accounts are new).

Twitter’s new link shortener give you LESS characters—but more security

Why you actually now have fewer characters:

The real big news is that since Twitter will be running existing links through their new t.co shortener, links that are under 20 characters will actually be lengthened! So using the URL shortening service that returns the shortest URL won’t help you gain any characters anymore.

Twitter is standardizing. Unfortunately, this means that by standardizing on 20 characters for the URL, all those of who were gaining a few extra spaces by using even shorter URLs will find our tweets actually getting longer.

How this makes links safer:

Again, from the Twitter engineering post:

“…We’re trying to protect users against phishing and other malicious attacks. the way that we’re doing this is that any URL that comes through in a DM gets currently wrapped with a twt.tl URL — if the URL turns out to be malicious, Twitter can simply shut it down, and whoever follows that link will be presented with a page that warns them of potentially malicious content. in a few weeks, we’re going to start slowly enabling this throughout the API for all statuses as well, but instead of twt.tl, we will be using t.co.”

Of course, all browsers already offer this same service. The reason you can’t have enough of this is that malicious web pages often catch a few people before they are discovered to be harmful, and the sooner they can be blocked the better.

An overview:

  1. This service won’t be available to most of us until later this summer
  2. You CAN enter links that make your tweet run over 140 characters. They will be auto-shortened by Twitter’s new t.co service.
  3. Your wording NOT including the link itself must be 120 characters long or less
  4. Twitter will even “shorten” (encode) links that are already shortened—making the URL always, automatically 20 characters long
  5. If you use a URL that is less than 20 characters, Twitter will lengthen it to 20 characters, leaving you less room for the wording of your tweet
  6. The GOOD news: All URLs pass through Twitter’s new t.co service—and thus all are checked to see that they are safe.
  7. Links may be displayed, depending on your interface, as non-shortened links. You may be shown the whole URL, or the first part of it. This is great—you’ll be able to easily see where you’re going. Twitter says they are “removing the obscurity.” Of course, via SMS and on some other interfaces, only the short URL will display.

Where the confusion came from:

Mashable tweeted this information as “Twitter to Change Links: They Won’t Count Against the 140 Character Limit” but later updated it to reflect the information here. Mashable at first had said:

“Summary: T.co links won’t be counted in Twitter’s famous 140 character count.”

Nope: they will be counted. They will just always be 20 characters long. Mashable quotes this passage posted on the Twitter Development Talk Google Group:

“the way the Twitter API counts characters is going to change ever so slightly. our 140 characters is now going to be defined as 140 characters after link wrapping. t.co links are of a predictable length — they will always be 20 characters. after we make this live, it will be feasible to send in the text for a status that is greater than 140 characters. the rule is after the link wrapping, the text transforms to 140 characters or fewer. we’ll be using the same logic that is in twitter-text-rb to figure out what is a URL. ”

I don’t blame Mashable for getting confused. Twitter is famous for engineering-speak. There are inevitably, after every announcement, Twitter clarifications. As usual, Twitter’s official blog post itself provides few details. Had Twitter made this slight clarification (added in brackets), it would have been clearer:

“The rule is after the [length of the] link wrapping [is included], the text transforms to 140 characters or fewer.”

Where Twitter says “140 characters is now going to be defined as 140 characters after link wrapping” they mean 140 characters after adding in 20 characters for the link.

Twitter’s Deadly Problem: The “Tweet At The Top” is Becoming Invisible

What if the police put up lots of fake speed limit signs—that you didn’t have to follow—and then set up speed traps wherever the speed actually changed after a real sign? They would catch a lot of motorists going the wrong speed. Many people would fail to notice which the real signs were at first (did you take a close enough look at the photo with this article?).

But then what if someone tied huge orange banners to the fake signs?

Very quickly everyone would learn to tune out any sign with an orange banner. This is why banner ads on websites work so poorly. Because of their recognizable wide size and position at top or bottom, and because they are rarely relevant to our purpose in coming to a site, we quickly learn to tune them out.

Humans are great at tuning out non-relevant, easily identifiable information.

Twitter is putting ads that look like tweets at the top of streams. Their deadly problem is that they are literally teaching us to tune them out because they aren’t yet relevant to most people. They will have to become extraordinarily relevant or interesting to break through the fact that we are being taught not to read them. They are failing to use promoted tweets effectively to teach us not to tune out the “tweet at the top.”

How expensive will this mistake be?

At one of this country’s largest Renaissance Festival some years ag0, I had 50 people working for me at a series of souvenir shops. During the first year, I determined that awareness of our products was low. It was because salespeople were trained to behave aggressively toward visitors. Yes, they were entertaining, but people for the most part avoided the areas around the shops with the most “entertaining” salespeople, realizing they were basically being accosted.

So I set up a product display—basic merchandising—at location that led to a shop, but was not near the shop. People were free to visit or ignore the product while being guaranteed to avoid the salespeople. Sales had been stagnant at this location for several years, always hovering just under $100,000 for the 15 days of the festival. I didn’t get the display up for the first two days of the festival, so the test would only run for 13 days.

The result? Even though we’d missed nearly 1/7th of the festival, sales were OVER $200,000. That’s right—sales more than DOUBLED when we overcame the problem of people tuning out. (Traffic—the number of visitors— was steady and weather was similar for all the years of this test.)

No matter how great Twitter’s sponsored tweets are, if they TEACH people to tune them out, they’re going to lose the majority of their financial potential as a business.

What should they do instead then? They’re doing some of it by showing promoted tweets at the top too. These are the most popular, relevant tweets culled from a particular set of search results. But they need to to mix the sponsored (ad) tweets in very gently. I’ve experienced so many search results with sponsored tweets at the top I’ve long since stopped looking at the “tweet at the top.”

Twitter needs to make adjustments immediately if they are to overcome the natural ability of humans to tune out, or it will kill their business model.

Will Twitter listen?

In a word: no. The management of successful businesses does not listen to outside sources, as a rule. Twitter’s great success in becoming popular and well-known is a powerful force pushing them into the kind of cloistered tunnel vision that is common in those kinds of situations.

So, what happened at the renaissance festival?

I told the upper management at the festival why what I had done worked. Their response? Build a giant castle in that location, forcing people to go into the Lion’s den—an enclosed space filled with salespeople—just to see the product. The result? Sales dropped well below $100,000. Did management understand why? No. They made their plans without me, and in spite of my explanation. But I had just done the same thing for a chain of about 50 Midwestern health food stores a few years previously. I knew the power of humans to tune out, and how powerful it could be if you could overcome it.

At the health food stores, I had more than quadrupled key sales.