When it comes to social media ads, Facebook and Instagram are often the default social networks. For the most part, results - be it clicks, impressions or engagements - cost less on Instagram and Facebook and there are more people to target.
Twitter, however, has some unique targeting options that can give it an edge over the other social networks for particular requirements, such as targeting followers of accounts, event targeting, TV & movie targeting and the topic of this blog, keyword targeting.
What is keyword targeting?
In a nutshell, Twitter’s keyword targeting allows you to target ads to people who have tweeted with keywords you define. It also targets people who have engaged with (liked, replied, etc) tweets that include those keywords.
So, for example, you want to reach someone who has talked about planning a holiday, you might include phrases like “holiday planning”, “planning a holiday”, “where should I go on holiday”, etc. You might also want to include people who have tweeted with one or more keywords, like holiday planning, were holiday or holiday destinations.
Keyword targeting will only target users who have a “hit” on your keywords or phrases in the last 30 days. If they tweeted with your keywords over 30 days ago, they won’t see your ad.
If you’re familiar with search marketing, Twitter keyword targeting has a lot of similarities.
What options do I have for choosing keywords?
It’s not just a simple list of keywords you can use. You can include single words, multiple words or phrases into your targeting.
Single words: like a holiday. Any person who tweeted with the word “holiday” in the last 30 days could be targeted. (Note, don’t do this, it’s far too broad a term!) Note that hashtags can be included as single words too.
Multiple words: like holiday planning hotels. Anyone who has sent a tweet that includes all three of those words, e.g. someone who tweeted “Planning my holiday to Cyprus, can anyone recommend any hotels?” in the past 30 days would be targeted.
Phrase searches: like “planning a holiday to Cyprus”. Only people who have tweeted in the last 30 days with the exact phrase above would be targeted. Someone tweeting “planning a vacation to Cyprus” in the last 30 days would not be targeted.
Excluded keywords: If there are keywords or phrases you want to exclude from your targeting (consider this like a NOT in Boolean search terms), you can add these in here.
It’s good practice to include a range of keyword/phrase combinations to broaden your potential audience. Luckily, Twitter’s Ad tool has a bulk upload feature where you can type, or better still paste in a bunch of keyword/phrase combinations.
Pro tip: use a spreadsheet to list out all of your keyword combinations and paste them into Twitter’s bulk tool. You can then use spreadsheet features like concatenation to quickly create sheets of multiple keywords for targeting.
Keyword targeting can be layered up with Twitter’s other targeting options, e.g. location, ages, gender, etc.
How could keyword targeting help me?
Well, there are lots of ways keyword targeting can be used by different brands for different purposes. But here’s a couple of examples to trigger your thinking:
You could display ads to people using a particular hashtag, be that a common hashtag, one used by your brand or even one used by a competitor. It might make sense for some brands like gyms to target people using #MondayMotivation, in a particular location who also express an interest in fitness, for example.
Customer happiness targeting
If your business sells multiple products and customers are likely to buy more from you, you could target people talking about your brand, or simply target your followers, who mention positive keywords and phrases like “amazing”, “great”, “great service”, “fantastic”, etc. Perhaps you could offer extra products, request reviews or offer some sort of reward to people who are captured as part of your ads targeting.
Competitor unhappiness targeting
You could target people who have tweeted about your competitor unfavorably. This could be set up by using combinations of your competitors’ brand name and/or Twitter handle with negative terms like “bad”, “awful”, “terrible”, etc. Feel free to add in some more offensive keywords here too!
We’ve used this approach for one of our clients and whilst the number of conversations is low, we’re able to display ads to people expressing dissatisfaction, creating very targeted warm leads. Consider it an effective long-term tactic.
Let’s say you own a restaurant close to a gig venue in a major city. Let’s say, for example, Katy Perry is playing in Dublin. You probably want to attract visitors to the gig venue to come to your restaurant before the gig.
You could set up an ad that targets people who mention the artist and the venue or city. For example “Katy Perry Dublin”, “Katy Perry 3Arena”, “KatyPerry Dublin”, “KatyPerry 3Arena”. Perhaps there’s a hashtag for the gig too, like “#KatyPerryIreland”. Include that too.
Your ad should display to those Katy Perry fans excitedly talking about going to the gig in the days preceding the event, or on the day of the event itself.
Other things to note
You can add up to 1,000 keywords per campaign.
Punctuation is not supported.
#hashtags are automatically included when targeting a keyword, for example adding the keyword "holiday" will include "holiday" and #holiday in targeting.
Some categories are defined as “sensitive by Twitter. In their own words, the following categories are excluded from keyword targeting:
Alleged or actual commission of a crime
Genetic and/or biometric data
Negative financial status or condition
Political affiliation or beliefs
Racial or ethnic origin
Religious or philosophical affiliation or beliefs
Trade union membership
Twitter keyword targeting is a very powerful feature and makes sense for some scenarios. We see it as a bit of a “slow burner” for many of our clients. The rewards often come from quite niche targeting, and with niche targeting, you usually get a small audience to target. But, for example, displaying an ad to people who have expressed unhappiness with something, particularly in industries where customers are in long-term relationships like energy, broadband or mobile contracts, can be particularly timely and is likely to achieve a higher-than-average click-through rate.