Twitter is probably the coolest marketing tool of our generation. A few things to ponder:
- Twitter lets you see what people are interested and talking about right now.
- Twitter lets you search user profiles to find what tribes people align themselves to.
- Twitter lets you see lists of people who follow your competitors
- Twitter lets you interact with all of those people.
When I actually put those facts together, it was one of those “holy crap we’re sitting on a gold mine” moments that marketers live for. It took a couple years to figure out the right strategy, but now if my startup (Grasswire) doesn’t get 50 sign-ups per day (not to mention 150 new, active Twitter followers), I know I’ve done something wrong.
Auto-favoriting and retweeting (the crappy way)
Have you ever noticed those accounts that seem kind of spammy always favoriting and retweeting your tweets? That’s because it works.
For example, there was a time when tweeting anything that contained “Hacker News” was favorited by @HackerNewsOnion.
My first tweet that contained “hacker news” was something a little bit condescending about how a Hacker News thread had turned into a huge circlejerk. When @HackerNewsOnion favorited my tweet, I thought, “Hey, these guys are laughing along. That’s cool.” I checked out their profile, and their tweets were pretty funny, so I followed them.
Then a couple minutes later I tweeted something else that contained “Hacker News” to a friend. Again, it was favorited. But this time it didn’t make sense; there was no reason @HackerNewsOnion should have favorited that one.
I opened up their profile and looked at the things they had favorited – it was just a list of every tweet that said “Hacker News” in any context. As I watched the account over the next few weeks I realized I wasn’t the only one being duped by a stupid little “favorite” bot. Dozens were following @HackerNewsOnion each day.
Method 1: Auto-favorite tweets containing pertinent keywords
If you’re ever up for a laugh, try tweeting “#SEO #social #socialmediamarketing #SEM #socialmedia SEO social media marketing.” Your tweet will be inundated with Twitter favorites and retweets from bots.
It was a little cheap and a little bit spammy, but for them it worked. It might for you, too, just make sure to not piss people off along the way. (I unfollowed @HackerNewsOnion).
Yet, despite the fact that it works, I don’t recommend it. Because there are better ways.
Selective following (the mediocre way)
If favoriting a tweet shows solidarity, following an account shows respect.
In a similar way to how @HackerNewsOnion used favorites, we can follow people to gain those couple seconds of attention you need to attract someone’s interest.
Twitter lets accounts with less than 2,000 followers follow up to 2,000 people. Those with over 2,000 can follow 1,000 people per day (so long as the number of people you follow isn’t more than 110% of the number of your followers). In other words, you can, by brute force, get a split second of attention from up to 1,000 Twitter accounts daily. Set up your profile right, and that traffic drives through to your product or landing page, and interested parties can sign up from there.
Let’s say you have a 10% follow-back rate (a pretty crappy one in my experience). That’s 100 new followers per day; 3,000 followers per month; 36,500 followers per year. It’s not pretty, but it works. What could your startup do with 3,000 interested Twitter followers? How quickly could you turn those 3,000 followers into users?
Manual following isn’t ideal; it takes a long time, and your fingers will wear out from hitting buttons. But as you begin the process of using Twitter as a marketing tool, it’s best to really get your hands dirty.
Method 2: Manually following people with similar interests
So let’s say, just as an example, you make iPhone screen protectors. How would you find potential customers?
You could look at the users who follow Zagg. Zagg (the largest iPhone screen protector company) has the Twitter handle @zaggdaily, with almost 39,000 Twitter followers — all of whom are presumably at least somewhat interested in iPhone screen protectors. You could open up a list of their followers, and follow them all in succession.
This would work, but you’d end up with a ton of spammy and other business accounts. Just look at this screenshot.
What we really need to do is find a search syntax that would find the right people.
Let’s go to Twitter and search “screen protector.” (make sure to hit “all” to get the most recent results, instead of the “top” results of the people who have the most followers).
That’s much better – everyone is talking about screen protectors. There aren’t any business accounts, not many spammy accounts, and each person there would be a potential customer.
But what if there were something better? If I were a person who was not only talking about screen protectors, but who was looking to buy, what would I be tweeting about? Let’s search “need screen protector” instead.
Bingo. There they are; our target market, ready and looking to buy, and perhaps most importantly tweeting about it right now. We could follow them, and they would see our profile, and hopefully head to our landing page or store. We might include a discount code and a link to our store. Sales would come in.
“But that’s easy,” you might say, “Screen protectors have a huge market of people ready to buy. No one has even heard of my project yet.”
If that’s the case we might go a little bit more abstract. A good rule of thumb is to look at the last thing you tweeted. What was it about? Find people tweeting about the same stuff.
What if you’re Heroku? Your target market is hackers. So enter the mind of a hacker; what would they be tweeting?
You could plug in the name of a competitor (digital ocean or AWS). You could plug in “hosting” or “scale/scaling.” Maybe you would plug in the top link from HackerNews or r/programming. You could even try your own name if you’re that big, or perhaps the URL of an article you’ve been mentioned in.
This has worked for Grasswire many times in different forms. Grasswire lets everyday people create collaborative news reports from real-time social media aggregation, so we might find the people tweeting about breaking news. Perhaps something to do with Egypt, Syria, or #NavyYardShooting (depending on the day). When people can see our real-time news feed of retweeted first-hand content at the same time they’re tweeting about it, our follow-back rates approach 40%. That’s 400 followers per day.
Perhaps more interesting than that, though, is watching the traffic flow to our landing page. We can, in the course of 5 minutes: Follow people, watch them flow through to our landing page (the link on our Twitter profile), and track signups. If we don’t get 50 signups a day from Twitter alone, something is wrong.
But the application doesn’t stop there. After I spent the summer living in my car to get Grasswire off the ground, we wanted to get some press as we went into money-raising mode. That’s when TweetAdder comes into play
Method 3: Semi-manually following
TweetAdder is basically a piece of software that makes it easy to scrape Twitter and follow people. There are other programs out there that let you automatically follow, but they’re against the Twitter TOS, and drop like flies. TweetAdder makes you click “follow” manually, but it’s worth it. It’s also free for the first 5 users, and runs on a Mac.
With TweetAdder I can not only search and follow in real-time (sorting tweets by most recent first to ensure real-time relevance), but I can search profile data.
So if I’m looking for press, I can search ‘+technology +journalist,’ or ‘+tech +journalist’ and follow from my personal Twitter profile (that has a bio to emphasize the fact that I lived in a car to get a startup off the ground). Spoiler alert: this, as well as some other press-hacking, got us in a half dozen major publications, and, in my opinion, deservedly so.
Warning: Using TweetAdder on a personal profile will pretty much destroy the enjoyment of using Twitter.
Method 4: Interaction (Do this!)
Yet, when all is said and done, all of this pales in comparison to true, authentic communication with people who are interested in what you have to offer.
Following is great if you’re looking for a quick bump, but if you actually “reply” to people using the same search syntaxes, people actually care about what you have to say.
Search for something the same way as we did before, export the tweets to a CSV file, sort however you like (most recent makes the most sense), and individually open profiles by clicking on the link and begin engaging with people. Twitter is, after all, a tool for two-way communication, and should be used as such.
We’re going to download TweetDeck, and create a new stream for the terms we want to interact with. For example, here is a list of everyone tweeting about my name or my blog.
As you can see, it pulls in every tweet that mentions my name, or a link to austenallred.com (including shortlinks).
Using TweetAdder I can click on any tweet and respond instantly, without opening a new interface.
Don’t let this become a rote copy-paste script – not only because Twitter will ban you, but because this is an unparalleled opportunity to engage in an authentic, one-on-one conversation with a would-be user. And there are lists and lists of them. Think about it: lists and lists of people who are probably interested in what you have to offer.
Doing this we have found not only people who are interested, but people who are advocates and a little bit obsessed about the little company we’re building (have you ever had anyone tell you they encouraged their mosque to pray for your Twitter stream)?
Conversion: Twitter Cards.
So we now have lists of people interacting with us that are interested in what we’re doing. We can take that data and turn them into users/subscribers/etc. by using Twitter Cards.
Twitter Cards are an easy way to allow a Twitter user to perform an action without actually leaving the Twitter interface. With one click they can even create accounts on your system with one click. Or they can join an email list.
Earlier I recommended tweeting “#SEO #social #socialmediamarketing #SEM #socialmedia SEO social media marketing.” I could hypothetically respond to everyone that tweets that particular tweet and give them a Twitter card to sign up for future chapters of The Hacker’s Guide to User Aquisition. Or I could push them to Amazon to buy the full book (once it’s released). Or I could just follow them, favorite the tweet, and have a cool new way to know who is actually reading what I write. I’m not doing anything with it yet, but you should.