How to Automatically Monitor Twitter for Mentions & Keywords [Tutorial]

Mar 4, 2021 | 6 minutes
brand-topic-mentions-illustration-alt

Mentions, hashtags, keywords: It’s undeniable that Twitter is a massive source of data that marketers, journalists, and traders (among others) can turn into useful information.

The problem, however, is that not everyone knows how to unlock Twitter data to its full potential. 

Manually searching for Twitter mentions isn’t really a solution, not after learning that on average, 6,000 tweets are sent every second. Before you do the math, that’s nearly 520 million tweets per day and over 15 billion tweets per month. Gasp!    

So, what to do? Three words: Automated Twitter monitoring. 

In this article, Andrew Davison, founder of Luhhu and certified Make partner, shows us how to automatically monitor Twitter in a step-by-step guide. 

Before we dive into it, let’s take a brief moment to learn how Andy has used this Twitter automation to move his business forward. 

Twitter monitoring for agencies and experts

Twitter is a fantastic tool that agencies and experts can utilize for self-promotion, and for building a reputation in a particular industry. 

As the founder of a no-code agency that specializes in process automation, Andy uses Twitter to answer questions, give advice, and find interesting people to network with.

“When I first started out, I quickly realized what a sheer firehose of unfiltered information Twitter is, and finding relevant tweets was difficult. I found a few tools to help manage the flow, but they were all quite expensive and overkill for what I needed”, he said. 

To deal with this, Andy built a workflow in Make to automate Twitter monitoring that does the following:

  • Compiles all the tweets mentioning Airtable in an Airtable spreadsheet

  • If any of these includes a question, he gets immediately notified 

Here’s how he did it.

Note: Twitter is a premium Make app, and you will need a paid account (starting at $9/mo) to create the integration. If you don’t have one yet, please register to Make.

Step 1: Setting up the Make scenario

After you have your Make account, you will begin by creating a new scenario from scratch:

After you do this, you will have to select the app module that will trigger your scenario: the Twitter app. Then, click the “Continue” button on the upper right side of the screen.

Note: app modules define the way an app responds to a request. For more information about basic Make terms, please refer to this page. 

After your press “Continue”, you will be directed to the scenario builder (aka visual canvas) which is what you will use to create the automated monitoring system. 

In the visual canvas, you will see that the Twitter app has been pre-loaded as your first option. Click on it, and select the “Search Tweets” module from your options. 

Step 2: Configuring the “Search Tweets” module

Once you’ve selected this module, please connect your Twitter account in the “Connection” field.

After connecting your Twitter account, the crucial part is constructing your search phrase.

Twitter supports a variety of search operators, but at a minimum, you’ll want to include the brand’s name and Twitter handle, separated by the word OR.

In our example, we use Airtable OR @Airtable (see image below). 

By doing this, our scenario will return any tweet containing the word “Airtable” or any tweet from the @Airtable handle.

One thing to keep in mind here is that this is a “dumb” search that considers the word and not the context. 

In the case of Airtable, it doesn’t really matter; however, if we were to monitor Apple brand mentions we would find ourselves inundated with tweets from fruit lovers as well as the tech fans. 

There are ways to solve this, but we’ll leave that for another occasion. Now, let’s keep pushing. 

Step 3: Filtering the noise

Now, we need to do some tidying up. For this, we’ll use a Make filter.

The reason for including a filter here is simple and useful: we don’t want any RTs, as they’ll clutter our list of mentions with useless content.

Behind the scenes, retweets come through from the Twitter API with an “RT” at the start, which makes them easy to filter for. 

We also applied a language filter to avoid getting tweets in a multiplicity of languages. You can of course adjust this second filter to match your preferences or remove it entirely.

Step 4: Configuring the Airtable Module

Another problem you’ll need to solve is duplicate tweets. Typically, this happens when people share an article from a news site with pre-filled text. 

Depending on the popularity of the brand you’re tracking, this could be the difference between a couple of extra rows and hundreds of extra rows on your spreadsheet every day.

So, to avoid duplicates, we’re adding a “Search records” Airtable module to our scenario.

This module will help us by searching the Airtable spreadsheet containing our tweets for any existing records that use the same text.

Note: we are using Airtable, but you could just as easily swap this for Google Sheets, Excel, or even an SQL or similar database tool. Anything that supports searching existing records.

To complete our anti-duplicate check, we add a filter between the Airtable module and the following module (“Create a record” Airtable module). This filter will set the scenario to keep running only if no record is found during the search (see image below). 

Step 5: Recording the tweet

Next, we will add the “Create a record” Airtable module to our scenario. This module will add new, non-duplicate tweets that mention Airtable as new rows.

At a bare minimum, you’re going to want to store the tweet text and a link to it, which you can construct using the [User: Name] and [Tweet ID] variables (see image below).

You can also store other fields, such as the author’s name and bio if you want as well. All it takes is to use the corresponding data operators. 

Step 6: Getting notified

As it was stated at the beginning of this post, this automated scenario will notify whenever a tweet with a question comes up.

For a no-code agency like Luhhu, providing timely answers is a great way to offer help, and an opportunity to pitch services and onboard new customers. 

So, each time someone tweets a question mentioning Airtable, this scenario will send those tweets straight into Slack. 

To do this, you will need to add a “Create a message” Slack module, and a filter between the last two modules (“Create record” Airtable module, and “Create a message” Slack module”). 

After you add the Slack module, you will have to configure it so it grabs all the relevant details from the tweet. See the image below:

To conclude, the filter. It’s a simple, but an important one, as it will only pass those tweets containing a question mark in the text (because questions are what we are after).

And that’s it!

Now all you have to do is a trial run of the scenario, activate it, and save it (all of this is done from the visual canvas - don’t leave before you do this, or your configurations will be lost). 

Conclusion

This is just one possible configuration of this particular workflow. As well as swapping out Airtable and Slack for your preferred tools, you could also switch the trigger from Twitter, to an RSS feed, or to multiple sources via an app like Feedly.

The end result will remain the same: no time spent on monitoring tweets, and you still get to have all the brand mentions, hashtags, or keywords that help you improve and advance your work and business. 

More automation ideas

Improve your work and business with Make. For more automation ideas, please read our selection of use cases.

You can also create custom Twitter integrations as well, as there are many available modules. 

Happy automating!

Andrew Davison

Andrew Davison

Andrew Davison is the founder and chief automator for <a href="https://www.luhhu.com/integromat-expert" >Luhhu</a>, a Registered Make Partner. They help businesses save time, money and scale their operations effectively with process automation. Andrew is active on Twitter as <a href="https://twitter.com/AndrewJDavison">@AndrewJDavison</a> where he regularly answers questions about Integromat and shares useful resources.

Like this use case? Spread the word.

Get monthly automation inspiration

Join 75,000+ Makers and get the freshest content delivered straight to your inbox.