Once your Twitter Ads campaign is live, how do you optimize so it keeps performing at a high level? Let’s take a look at a few key points.
So, you have successfully set up your Twitter Ads Campaign (we’ve covered 3 steps related to planning and set-up of Twitter ads campaigns in Part 1 of this blog post). What do you think happens once your campaigns are up and running? That’s right, optimization, optimization, and optimization! No time to relax – yet. Below we’ll talk about how to maintain and optimize your campaigns to keep them performing at a high level. And again, these steps are applicable to campaigns for any company and industry.
Step 4. Optimization
One of the biggest lessons we’ve learned while running a campaign for our tourism industry client Visit Myrtle Beach (check out Twitter case study on this successful campaign) is that you can’t just “set it and forget it.” You have to do an analysis to see which different elements of campaigns and tweets work best. Here are some of the questions to ask as you are looking at your Twitter Ads:
- Which targeting is the most effective? You should be analyzing first and foremost by your Key Performance Indicator (KPI). So, if you are looking to gain followers, look for the campaign with the highest Follow Rate and the lowest Cost Per Follower. If you are running a general awareness campaign, look for the lowest cost per engagement, or cost per view, in case of a promoted video campaign. If you are going for website clicks, as we at Visibility and Conversions often are, make sure you use an outside tracking system, because Twitter doesn’t report website clicks separately. The column “Clicks” in the campaign analytics (see the screenshot below) includes, per Twitter, “clicks to links, the Website Card, favorites, hashtags, embedded media, your profile and other Tweet details in your ad. This includes clicks you are not billed for.” This means if you calculate your Cost Per Click (CPC) based on that number, it will be very, very underestimated. So what we usually do is use Google Analytics, Adobe Omniture or whatever tracking tool our client prefers to track the links used in tweets. Here’s an article from our blog on campaign tracking if you would like to find out more about it. Later, when optimizing and evaluating a campaign, we go back to this analytic tool to see how many clicks each campaign has earned. Then we divide the spend of a specific campaign by the number of clicks, and voila, we have the REAL CPC that helps us in social media decision making.
- After a bird’s eye view of your campaigns, prepare to dig deeper. Out of all the targeting elements used for your campaign, which ones are the most effective? As you might know, Twitter now allows you to target simultaneously by keywords, handles, interests, platforms, TV shows and more. Once your campaign is set up, you can go back and evaluate which of these work well for you. For example, you want to know which keywords bring you the most impressions at the best engagement rate within a particular campaign, or which device is your audience mainly using. Click on the right to choose a particular targeting element of your campaign, and then sort by the metric that’s most relevant to you.
- As your campaign takes off, check back to see if you need to increase or decrease your bids to keep your advertising cost-effective.
Here are a couple of actions you can take once you’ve conducted the above analysis:
- Build more campaigns similar to the best-performing ones.
For example, if you are running a promoted account campaign, and a particular segment responds well to it, why not also run promoted tweets campaign using the same segment for targeting? You can also run a different type of offer to the same segment.
- Shift budget from the worst-performing campaigns to the ones that perform well.
Step 5. Testing
You probably already know that testing is the only way to make sure you are using the most effective targeting, message and budget. If you didn’t have much diversity in your tweets and targeting from the beginning of your campaign, you will probably get some ideas on what might work while optimizing your campaigns. Then, you need to test what works.
Try combining all different targeting elements (keywords, TV targeting, usernames, etc.) to increase your audience size. Also, try the opposite: For example, if your message and product are device-sensitive, break down a campaign into two or more campaigns each targeted to a different device.
Test different elements of the tweets – photo, copy, call-to-action buttons (if using Twitter website cards or Twitter lead generation cards). Create multiple slightly different tweets (see the screenshots below for an example). In a few days, evaluate their performance. Turn off the tweets that didn’t get as much engagement and/or were not cost effective. By the way, never delete tweets, as you’ll lose all the data related to them.
Step 5. Refreshing the tweets
The big Twitter ads project we undertook for our client Visit Myrtle Beach in 2014 lasted two months. What I initially hoped for was that I can set up all the tweets, tailor them to the target audience segments, and happily rest on laurels. Not so easy. Soon enough, I discovered that once the tweets have been live for some time, engagement starts to decline even on “successful” tweets. It makes sense – why engage with a two-weeks-old tweet when there are plenty of new ones appearing every second? Yes, replacing your tweets with fresh ones is sort of a pain. But if you don’t want to see your metrics sag, it is worth the time. That being said, they don’t have to be completely new tweets – the copy and images may remain the same, especially if they worked well; you just need to re-inject them in your campaigns.
Step 6. Evaluating you campaign
If you have been optimizing throughout the campaign, you probably already have a pretty good idea on whether or not you are close to reaching your target KPI. At the end of the campaign, take the final measurements, much in the same way you measured your KPI(s) during the campaign. However, also tally up any additional effects of the campaign, besides the KPI. How many followers did you gain? How many impressions, retweets and engagements your campaign earned? How many mentions did you receive, and how does it all compare to the months when you didn’t run any campaigns? These additional analyses may change your approach to Twitter as it works differently for every brand. Maybe your goal was to generate leads, but instead your campaigns generated a lot of buzz? That topical awareness can be a strategy, too, in its own time. The additional data makes you aware of the platform’s capabilities and allows you to make informed decisions when approaching your next Twitter ads campaign.
So these are some steps we take to optimize and maintain Twitter ads campaigns for our clients. Now, over to you: What elements do you usually look at when optimizing your campaigns? Do you test different creative and copy? What valuable insights did you gain as a result of optimization and evaluation of your Twitter campaigns?
In Categories: Destination Marketing, Internet Marketing, Social Media