Inbound marketing is all the rage these days, but that doesn’t mean that traditional outbound marketing (a.k.a. advertising) can’t still be effective.
In fact, it has a huge advantage over most content marketing efforts…
With content marketing, the average campaign takes at least a few months to see any real results.
You invest money upfront, but need to wait for the return.
With advertising, on the other hand, you spend money upfront, but you see the results almost immediately.
Additionally, there’s no reason why you can’t use both types of marketing.
That way, you get both immediate and long-term results.
I write a ton about content marketing. If you want to learn more about it, I’ve got you covered.
What I haven’t written about a lot is paid advertising.
In particular, there are a few advertising networks that provide you with a better opportunity than most.
One of those is Twitter.
And that’s what I want to teach you about today.
Twitter has a global advertising revenue share of 0.84%, which makes them one of the top 10 largest advertising networks on the planet.
You can reach just about any audience on the social network, and the ad platform is really simple to work with (and you can scale up easily when needed).
The cost of your Twitter ads depend mostly on one factor: their quality scores.
I’m going to show you how to optimize your Twitter ads’ quality scores, which will have a huge impact on whether or not your campaigns are profitable.
The key ingredient of a great quality score and how it saves you money
In case you haven’t used Twitter ads much in the past, let me give you a quick refresher on the quality score factor.
Twitter, like all businesses, has a great interest in making its users’ experience as good as possible.
On social media, this means engagement.
If users on Twitter find posts useful, they will favorite them, comment on them, re-share them, and click on links contained within them.
From Twitter’s perspective, the more users are engaging with posts, the happier they are.
Twitter allows businesses to show specific Tweets to a certain audience even if that audience doesn’t follow them. All they have to do is pay. This is also known as advertising.
But as you and I know, most ads suck.
And while they are necessary for Twitter’s business model, they can have a negative effect on the average user experience—that’s a problem.
In order to encourage businesses to create ads that users actually like, Twitter introduced a quality score.
It’s a simple concept:
The more engagement your ad gets, the less you pay. The less engagement your ad gets, the more you pay.
Larry Kim revealed some interesting numbers from experimental campaigns he ran:
You can see the engagement rate in the picture right next to the cost per engagement column:
- a 68% engagement rate resulted in a cost of $0.01 per engagement
- a 37% engagement rate resulted in a cost of $0.02 per engagement
- a 21% engagement rate resulted in a cost of $0.03 per engagement
- a 7% engagement rate resulted in a cost of $0.08 per engagement
Those are big differences.
Those numbers show that an effective ad gets 8 times the engagement of a poor one.
In practical terms: You aren’t going to have a profitable campaign if your engagement rate sucks.
That’s all you really need to know about the quality score.
While you can’t currently see what the quality score of any of your ads is, it is mainly affected by your engagement rate, so focus on optimizing that.
1. Always use Twitter cards when possible
If you’ve used Twitter much, you’ve probably noticed that some Tweets stand out more than others.
That’s because smart users have taken advantage of Twitter cards.
Here’s what a regular Tweet looks like:
And here’s what a Twitter card looks like:
Obviously, having images and videos embedded into Tweets is going to not only increase the attention they draw but the engagement as well.
There are a few different types of Twitter cards, all of which have their own purpose. For example:
- Summary card (optional w/image) – Typically used to share links. Shows a description, title, and thumbnail. You can also configure it to use a large image instead of the thumbnail.
- App card – You can promote an app on a Twitter card, and users can install the app directly from it.
- Player card – embed a video (like the first picture I showed you above)
There are few times where it makes sense not to use a Twitter card in your ad. But as a rule of thumb, always include one because it will improve your quality score.
How to make your own Twitter card: Regular users can create Twitter cards by including the right meta tags on the pages they link to.
For advertisers, it’s much easier (at least now it is).
When you first log in to Twitter ads and create a new campaign, you’ll need to select an objective.
Most often, you’ll want to use one of the first two: Tweet engagements or website clicks/conversions.
Once you select that, you’ll be taken to a page with four main sections where you fill out all the details of the ad.
If you scroll down to the Compose Tweets section, you’ll see a form where you can create a new Tweet to promote.
In this form, there is another form specifically designed for a “card”:
It’s pretty straightforward to use. You can customize the picture, headline, and call to action. The card description will be taken from your website if available.
Or promote an existing Twitter card: It’s also possible that you already have a Tweet (or Tweets) that you’d like to show to a larger audience.
To do that, just click the “select an existing Tweet” button to the right of the above form:
I’ll show you why you might want to do this later in the post.
2. How to use “tailored audiences” to only reach users you care about
It’s marketing 101: You need to show your message to the right audience in order for it to be effective.
It’s difficult to sell ice to someone in Alaska, but not too difficult in Florida. Same product, different audience.
More commonly, it’s called targeting in advertising.
Twitter has millions of users, and you don’t want to target all of them with your ad.
Instead, you want to show it—target it—to a small percentage of those users who are actually interested in your product.
If you’re promoting a local event, you want to show it only to the users interested in your industry who also live in your area.
This is something else that Larry Kim tested. Specifically, he tested an ad targeted to people who live in the South Florida area against an ad with no location targeting.
Not surprisingly, the one that was more targeted had an engagement rate three and a half times higher than the other.
That’s a big difference.
The more accurately you choose your ad settings, the higher the engagement rate (and quality score) you will get.
Targeting on Twitter: When you create a new campaign, you’ll be taken to a long page to set up your ads.
The targeting options are contained in step 2 of the process. At any point, you can see an “estimated audience size” on the right hand side:
As a rule of thumb, you want to target under 100,000 people with an ad. If your audience is much bigger, it probably isn’t targeted enough.
How do you decrease your audience size? You start picking different aspects of the targeting system.
For example, I specified that I only wanted to target users in the United States with my campaign. Instantly, the audience was reduced from 316 million to 66 million users:
In addition to the obvious targeting options such as location and gender, there are several other criteria you can pick from. Once you start adding them, the audience size will decrease further.
Knowing which options to pick is what separates beginner advertisers from experienced ones.
Here’s a hint: You don’t need to use them all.
Many beginner advertisers start by trying to use every targeting option available. Then, when something works well or doesn’t work well, they don’t know which targeting option actually helped or hurt.
Instead, keep it simple, and focus on the main things that separate your target audience.
Overall, however, this is one of the most difficult parts of advertising, so expect to get better at it over time.
One reason why this can be difficult is because ideal targeting options can (and should) be different for every business.
For example, if you sell a fitness course, you can target all of the main countries (US, UK, Canada, etc.). Location isn’t a big issue, but you still may want to filter out countries that you can’t accept payment from.
However, you’d probably want to start by adding an interest. You could browse the “health” section of the interests and then target users interested in weight loss:
In addition, you might want to target only followers of certain Twitter users who are in the fitness niche.
You can do some really in-depth targeting with the Twitter campaign creator. Invest the time upfront to explore all the different options so that you can find a way to describe your ideal audience.
Find more of your ideal users with a tailored audience: There’s one specific option that can take your quality score to the next level.
Under the “add interests” option, there’s one that says “add tailored audiences.”
When you click it, it will expand to give you two different options.
Tailored audiences can be really powerful. Essentially, they allow you to create a list of ideal users and let you target them on Twitter.
So back to your two options. For either one, you will need to have some sort of an audience already because you need those ideal users on your list. So, if you have a brand new business, you won’t be able to take advantage of this yet.
Otherwise, let’s walk through what you can do.
The list is pretty self-explanatory. If you upload a spreadsheet with the email addresses of people (probably your existing subscribers), Twitter will see if any of them are associated with existing Twitter accounts.
Then, you can add that tailored audience to your campaign settings and target them with specific Tweets.
Where might you want to use this? Well, say you’re in the process of selling a new product.
It would probably be useful to use Twitter (in addition to emails) to keep the product in your prospects’ minds.
Alternatively, and more useful in most cases, you can target your website visitors.
First, you click that second option, which will show you a screen like this:
Give it a name, select “site visit” from the dropdown menu, and make sure the tailored audience box is checked.
Then, save the tag and generate the code:
This tracking code can be pasted into the HTML of any page on your website. Twitter will check if a visitor is a user, and if so, they will be added to your tailored audience.
There are a few situations where this could be really useful:
- To spread your content – If you tweet a link to a blog post to a random Twitter user (interested in the subject), they have to first click the link (which only a small portion will do), and then decide to engage after. But if you target people who have already seen the content, they already know whether it’s good or not, and you’ll get a much higher percentage of shares.
- To improve conversions – You can also put the code on specific landing pages. Then, you can offer those users a different bonus (on Twitter) to get them to go back to your site and opt in to your list. This is very similar to remarketing and is a great way to get high quality subscribers for low cost.
Note that it can take a while to create a large enough list to target. If you have a low-medium traffic site, you might have to let it run for a few weeks.
Once you’ve created a tailored audience, it will be available under the “tailored audience” section of the main setup page.
When you have a tailored audience, you don’t need to use any of the other targeting options on the page because you already have a highly targeted audience. Aside from that, your tailored audience list size likely won’t come anywhere near a hundred thousand people.
3. Be selective and reap the rewards
It can be incredibly difficult to predict which of your Tweets will get a ton of engagement.
That’s why you might not want to just jump into creating new Tweets as ads and spending money on them. That’s because if they don’t end up getting much engagement, it’ll hurt your quality score and, therefore, your budget.
Instead, you can first test Tweets by posting them on your account and then use only the very best ones for advertisements (or create similar ones).
Once again, Larry Kim did a pretty cool experiment that shows how effective this can be.
He had a Tweet that was already getting good engagement with his followers. He then promoted that Tweet using Twitter ads. He ended up getting over 100,000 visits and 1,500 retweets from his advertising.
Guess how much it cost him?
In practical terms, he was able to get 400 visits per dollar he spent.
Just about anyone can find a way to get this many fairly high quality visitors for that cheap.
So, if you have a few Tweets that have already gained some good engagement, you can choose “select an existing Tweet” in the campaign option:
4. Advertising to get more followers? Testing comes first
I briefly mentioned it earlier, but there are many types of ads on Twitter.
Most businesses will focus on just one or two of them.
One common type is an ad to get more followers.
This is good if you plan to continue to use Twitter in the future and would like to get more followers so that you wouldn’t have to pay to show them your content every time you tweet.
If you go with one of these ads, engagement still matters, of course.
When you post one of these ads, this is what users will see:
In the sidebar, there’s a section for suggested people to follow, and there’s a little “promoted” icon on it if you paid for your profile to be there.
Can you optimize this? Absolutely.
There are two types of reactions that users have after they see this type of ad:
- I recognize that name or picture in a good way, so I’ll follow this person.
- I’m not sure who that is. I’ll either ignore it, or I’ll look at their profile for more information.
There’s nothing you can do about the first one other than keep building your business and brand. Brand recognition will come into effect eventually.
But the second scenario brings up a great opportunity.
First, your image needs to hook users’ attention. If it looks spammy or uninteresting, the users aren’t going to bother learning more about you.
You should either use a really good-looking logo, or better yet (if possible), use your picture—people are interesting to other people.
The big opportunity to raise your engagement, and quality score, here is to optimize your profile.
When someone clicks on your name, they see your profile, which consists of your picture, description, Tweets, and background profile picture.
Too many businesses ignore this area, but it can make a big difference in converting the users who visit your page into followers.
I’ll use Brian Dean’s Twitter page as an example of how to create an attractive one:
Of course, this will perform better than a run-of-the-mill profile page.
As I’ve pointed out in the screenshot, there are three main areas in your control:
- The big background area – a chance to display your brand/logo as well as social proof (i.e., the reason why someone should follow you).
- Your profile icon – which also shows up in sidebars.
- Your description – a little bit about you and why someone might want to follow you.
You should optimize all three areas.
How do you do that?
Essentially, you change one part at a time, run an ad to get followers, and measure the conversion rate.
Then, you try a different description or picture, and see which one’s better.
For example, you use your logo as the small profile picture for one test and see that you end up paying $0.05 per follow.
Then, you use a picture of yourself, keeping the background image and description the same, and see that you only pay $0.03 per follow.
You can test whether you have a large enough sample size by using this simple and free sample size validator.
If you did have a valid sample size in that experiment, you could conclude that the picture of you was more effective. Next, you could test another element on the profile page.
5. How to study the competition and learn from them
Coming up with the perfect Tweet to advertise isn’t easy.
You basically have two options…
You can create Tweets from scratch and run them as ads, and if they aren’t working well, you stop them.
Or you can look at what is already working and create your ads based on that information.
The second option is far superior, and you’ll save a lot of money by finding out which ads work and don’t work before you create your own.
The next question is where do you find these Tweets to base yours on?
Well, you want to make sure that your advertisement Tweets will be well received by your target audience. Therefore, you want to look at the Tweets of your competitors.
They have largely the same audience, which is perfect.
If one of their Tweets is really popular, it would likely perform well as an ad. Conversely, if they get no engagement at all on certain Tweets, try to figure out why your audience doesn’t like them.
If you were running a blog on social media, one of your biggest competitors would probably be Social Media Examiner.
So, your first step would be to find their Twitter page, which you can do with a simple search on Google or Twitter.
Then, look through their past Tweets, and note down which ones have more shares, favorites, and comments than the rest. These are the high performers.
You’ll want to gather at least 5-10 popular Tweets in the past few weeks/months.
These will tell you three things:
- What types of Twitter cards perform best – i.e., videos, pictures, galleries, apps, etc.
- What topics perform best for your audience – The above picture clearly shows that the audience likes social media tools (which I would agree with from personal experience)
- Any particular format that works best – Some audiences like Tweets with questions, others like descriptions, yet others just like a good headline. Mimic the writing style of the most popular posts.
When you take those three factors into account, you’re much more likely to produce a Twitter ad with a high engagement rate (and quality score) than if you randomly created one from scratch.
6. On Twitter, everything moves fast (this affects ads as well)
There are many reasons to like Twitter as a user.
But one of the most common is that there’s always new content.
Even if you return to Twitter an hour after you visited it, chances are you’ll see a completely different feed.
Users on Twitter like this rapid state of change and expect to see new content constantly. If they see old Tweets or similar to old Tweets, they just skip them.
This has implications for your Twitter ads as well.
The first few days you run your ad, it will likely perform as well as it’s going to.
Then, it quickly goes stale, and your engagement rate continues to drop.
This is common on just about every advertising platform, but it happens even faster on Twitter.
If your ad is shown to the same person, it will often be ignored, which can have a serious impact on your engagement rate.
On top of that, the ads will be shown to the most active users in your target audience first. As you approach the last parts of your audience, the less enthusiastic members of Twitter, your engagement rate will, of course, be lower.
How often should you change Tweets? Twitter itself recommends using at least three different ads for a campaign to slow down the speed at which they go stale.
But even if you create several Tweets at the start, they’ll still go stale—it might just take a few weeks.
The only way you’ll know how often to create new Tweets is by monitoring the performance of your ads.
Once the cost per engagement of a specific ad reaches a level that you don’t think is profitable, stop running that ad.
Then, create a new ad.
Here, you have two main options:
- Immediately create a new, similar ad – You can create a new description and use a new image in your Tweet. Even though you’ll link to the same content, it’ll seem like a new Tweet to users.
- Switch to a different topic – You don’t want to show the same users the same content right away even if the Tweet is different. Instead, you can switch to a different topic and come back to promoting the first piece of content in the future.
The second option is better in most cases. If you target a similar audience you targeted with the first ad, some people who engaged with the first ad will also engage with the second ad.
If you show them the exact same content, you can’t blame them for not being very interested.
However, if you show them new content, they’ll be as interested as they were the first time.
Then, when you start promoting the first piece of content a few weeks later, it won’t be so fresh in their minds. They won’t mind getting a quick refresher and still might retweet or engage with your Tweet in some way.
Twitter is one of the biggest social media sites on the Internet.
It has hundreds of millions of users, which means it can be used to engage with the target audiences of almost any business.
And while growing an organic presence on Twitter is a good idea, it can take many months to see real results (traffic, shares, conversions, etc.).
Instead, you can get those results a lot faster by using Twitter advertising, which is a fairly well developed advertising platform at this time.
But if you don’t understand the effect the quality score of each of your ads has on the cost of your advertising, you’ll have a difficult time creating a profitable campaign.
If you understand everything I covered in this post, you should know enough to get started with Twitter advertising and be successful.
If there are any parts of the quality score that you don’t fully understand, just let me know in a comment below, and I’ll try to clear things up.