Matthew J. Kushin, Shepherd University
Meltwater Media Intelligence Software
Cost: Free with Meltwater University Program
Contact: Carol Ann Vance, Director of Talent Acquisition, Americas, firstname.lastname@example.org
Meltwater media intelligence software (https://www.meltwater.com) is an online software suite that enables subscribers to conduct news and social media monitoring, as well as build media lists and perform media outreach (Business.com, 2018). Like many of its competitors, such as Radian6 and Cision, Meltwater follows the software as a service (SaaS) model. Users subscribe and log in via their web browser to access the software package. Users do not need to download any files or software to use the service. There is also an optional mobile app available in the App Store and Google Play that enables users to access many of the service features on their mobile device (“Meltwater Mobile,” n.d., 2018).
The software works by allowing users to search public social media data from platforms such as Twitter, Instagram, YouTube, and Facebook, as well as forums, blogs, comments, and product reviews (Business.com, 2018). Similarly, the software enables users to search articles from news databases. A search can be used to look back in time to past social media posts or news articles. Once set up, a search will collect social media posts or news articles going forward. According to the Meltwater help center website, news searches go back to the start of 2009, and social media searches go back a rolling 15 months (Apple, n.d.).
Since January 2017, Meltwater has offered a university program, which enables faculty and students to get free access to the software to use in the classroom for educational purposes. Carol Ann Vance oversees the university program at Meltwater. Software training was initially provided through an optional introductory video call with the professor and class; however, due to the scale of the university program, the company has moved to a model of making training videos accessible. In addition, online training is embedded into the help section of the software.
OVERVIEW OF MELTWATER
The Meltwater software is divided into functions via a navigation menu on the left. The first part of this review focuses on functions related to the news and social media search features. Following that, additional supporting features are noted.
Users set up searches for keywords or phrases and provide time and date parameters. Searches can be for either news or for social media content. As their titles suggest, news searches enable users to search a real-time database of news articles across a growing list of more than 300,000 news sources (C. A. Vance, personal communication, May 23, 2018). Multiple searches can be programmed to run concurrently. Social media searches enable users to search a variety of social media platforms, as well as forums, blogs, and product reviews.
A search can be set up by navigating to the “Search” menu item. A user enters a keyword to begin a search. Once a search is initiated, additional keywords can be entered. Further, after a few keywords are entered, a list of related keywords is generated and suggested by the software. The user has the option of including those keywords in the search. Similar to searching on other platforms, keywords can take a variety of forms, including but not limited to a word or phrase of interest; the name of a person, organization, or entity; a hashtag; or a username. Users have three options for the types of search they want: (1) all of these keywords, (2) at least one of these keywords, or (3) none of these keywords (see Image 1). These options can be combined in a single search. Users can also enter search terms using Boolean logic in the advanced settings. Date parameters and source parameters can be set. News source parameters include location, source type, reach percentile, among others. Social media parameters include source type and language.
Once a search is set up or after an existing search is edited, the user begins to see the results in real time. Searches default to a reverse-chronological view, with the most recent post at the top. The results can be sorted by a number of other dimensions, including reach, sentiment, and geolocation. Each item in the search results is organized into a “content card,” which contains metadata appropriate to the type of search. For example, for a news search, metadata information includes publication, headline, byline, date and time of publication, reach, and sentiment score. In each content card, the first few lines of text to a news post are shown, as well as a photograph. Clicking on the article headline takes the user to the original article in a new browser page. Similarly, users can sort metadata in social media searches and access the original post by clicking on the social media profile data associated with search results.
Dashboards are created from existing searches. They can be accessed by navigating to the “Dashboards” menu item. Therefore, a user must first set up a search for a topic of interest before creating a dashboard for that topic. Dashboards are visual representations of data from a search, and they are organized into interactive widgets (see Image 2). Widgets are displayed as windows within a dashboard and, as the name suggests, function to analyze and display information visually. There are a variety of widgets, including but not limited to widgets that display share of voice, potential reach, media exposure, sentiment, trending themes, top locations, top sources, a heat map, and a Google Analytics widget. Each widget can work with one or more of the following sources: news, social or RSS feed. Each widget can be interacted with and customized individually, enabling the user to modify the date range and the search source for that widget. Further, the location and size of widgets can be changed within a dashboard. As such, Dashboards serve as the key way to interact and analyze data in Meltwater.
There are three default dashboards: monitor, benchmark, and analyze. Additionally, users can set up custom dashboards by picking and choosing the widgets of their choice. All dashboards can be customized by adding and removing widgets from the settings in a dashboard. Multiple dashboards can be programmed to run concurrently, and users can set up multiple dashboards for the same search. It is worth noting that data from dashboards can be downloaded in CSV format for analysis in external software, such as in Microsoft Excel.
The influencer tool enables students and professors to build searches for influencers working at news outlets and to build media lists. Users can search for influencers and their associated news source through Meltwater, import a list of contacts from a CSV file, or create individual influencers manually by entering their contact data.
To build a media list using the Meltwater search software, users start a new search and select whether they are searching for “Contacts” or “Sources.” Several filters enable users to focus their search on key parameters, such as beats, source reach, geographic focus, language, media channel, and others. Once a search is run, the user can select from the results to see detailed information about the reporter or media outlet (see Image 3).
In addition to the features above, Meltwater contains several additional features.
The “Home” menu item is a default view that a user sees when logging into Meltwater.
The “Inbox” menu item is a place to organize RSS feeds and searches that a user has programmed.
Tags are created by users in the “Tags” menu item. They can be used to classify and thus organize content for easy access later. Articles or social media posts can be tagged via the search results.
The “Outputs” menu item enables users to output, or share, search content from pre-programmed Meltwater searches to a third party outside of the Meltwater platform. This may include internal parties such as direct reports or external parties such as clients or website visitors. In the classroom context, students could create outputs to share with the professor or with class clients.
Outputs take one of two forms: 1) newsletters and 2) newsfeeds. A newsletter allows a user to pull posts from a preprogrammed search to send via email. The newsletter is organized by sections, which contain user-selected posts from a search and which can contain explanatory text that the user can add. Each section can contain data from a separate preprogrammed search. Once set up, the newsletter can then be emailed to an email list, which can be imported in CSV format. The newsfeeds feature enables a user to create a feed of posts from a single preprogrammed search. A website administrator could then use the generated RSS or XML to produce a web feed to be hosted on an organization’s website. These features are not likely to be used in the learning environment.
This feature is not available through the university program. It enables users to create custom reports of Meltwater data.
The “Settings” menu item is where users manage their account information and customize preferences. Importantly, the settings feature is also where users can connect Instagram and Google Analytics to pull data into Meltwater from those services.
LEVEL OF EXPERTISE REQUIRED FOR USE
The user should have a working knowledge of social media, social media analytics, and media relations. The interactive, visual nature of the software is approachable but can be cumbersome. The available help tools within the software provide resources, including step-by-step videos, for professors to learn the software and teach it to students. But, learning to use the software effectively requires a substantial time investment. Given that the software can be accessed via a web browser or via the native apps (Business.com, 2018), it can be readily accessed by professors and students both on and off campus, irrespective of operating system.
MY EXPERIENCE WITH THE SOFTWARE
I integrated the Meltwater software in my undergraduate 300-level social media course in fall 2017. To me, the power of the software is in its ability to quickly pull historical data from a variety of sources. Unlike other software that I have used, Meltwater does not require a user to set up a search ahead of time to begin collecting data for future analysis. This enabled my class a lot of flexibility in what we were able to search. With other software, I have had to pre-program searches at the beginning of the semester and wait several weeks to collect enough data to use for class projects.
However, there were some hiccups and challenges with using the software. I will discuss several below.
We began with a brief activity where students set up a few searches to monitor social media content related to a well-known health and beauty brand and its competitors. The purpose of the activity was to familiarize students with what the software was capable of monitoring on the social media side. Here, we ran into a few issues where software usability challenges and user inexperience conspired to create problems. First, when setting up their searches, students struggled to get creative in generating the search keywords. Beyond the brand names themselves, I tried to encourage creativity. For example, I suggested using the brand’s social media account handles and using various spellings of the brand names (e.g., with and without an apostrophe in one case). Students also did not tend to look at the search results to look for false positives. In some cases, students needed to go back and add keywords to the “None of these keywords” textbox to refine results. Altogether, students tried to rush this important process.
Second, some students struggled to follow the instructions to make the search a social search, accidentally setting up the search as a news search instead. Students didn’t seem to understand how the “news” and “social” searches differed, despite having taken the video training provided in the software as well as my brief lecture in class. Both the variations in keyword choice in search set up and the mistake of creating a “news” search instead of a “social” search led to inconsistent results between students when they analyzed their search result data in a dashboard. Third, several students stumbled through understanding that a dashboard was needed in addition to a search to see analytics. They expressed that they felt that the search should auto-generate a dashboard. Because there are several different types of dashboards, the differences for which are not immediately clear, they grew frustrated and confused. These students did not understand that the search serves to pull in the data, and can be modified, and that the dashboard provides the output of the data via the widgets. Fourth, when setting up additional widgets within an existing dashboard, a user must select which search to pull the data from. Several students selected the wrong search unknowingly, and were thus believing they were seeing data for one brand when in fact they were seeing data for a different brand.
In ways such as those described above, students seemed to struggle with grasping how to navigate and use the software. Those struggles led students to believe they were reporting the correct data when in fact they were not.
Turning now to the data the software provides, there are a few limitations to be aware of.
First, be aware that the software is a monitoring tool and does not provide access to analytics of social media accounts. My students also completed a social media audit assignment and we were only able to use some features of the Meltwater software. For example, we were unable to find a way to use the software to explore analytics surrounding a specific social media account (e.g., a Facebook page and its followers, or audience demographics). This is likely because this information is not public. Therefore, for certain assignments, additional tools or client access to built-in analytics will be needed in the classroom beyond Meltwater.
Second, while the news media influencer tool is great for building media lists, it was a bit challenging to help students identify social media influencers via Meltwater. There are a few proxies which can be used to try and triangulate a search for social media influencers. I had my students sort searches by reach and by engagement. There is also a dashboard widget for finding the top posters by volume, as well as a widget for share of voice. However, a separate tool for identifying influencers in a social conversation would be a wonderful addition.
Third, in my own experience, I found the widgets a bit cumbersome to navigate. From a dashboard, one has to click into the widget by clicking the blue arrow before a widget can be interacted with. Interaction with the widgets feels a bit one-dimensional. When a data point – such as a date – is clicked on in a widget, the software displays all results pertaining to that data point. However, many widgets do not allow further drilling down. For example, if the widget showed search volume for a four-month period and I clicked on one day, the widget will not zoom in on that day and show me a timeline of posts across the hours of that day. In this regard, I was not able to see what time of day search volume was at its highest.
Upon completing our use of the software throughout the semester, I asked students for verbal feedback about their experience. Some expressed excitement and a sense of empowerment, others expressed a sense that they learned a lot, and a few expressed a sense of being overwhelmed or intimidated. A majority of students said that they wished I had spent more time in class showing them how to use the software. I recommend that professors planning to use the software in their classroom offer ample in-class opportunities for students to learn to set up searches properly and refine searches through keyword targeting. I also recommend walking students through the dashboard setup a few times and checking for understanding and task completion. Look for discrepancies between students in their results, and help students to reverse engineer how decisions they made in setting up their search led to those discrepancies.
The Meltwater media intelligence software provides a powerful software suite for teaching students about news media monitoring, social media monitoring, and analytics. Meltwater is also a powerful tool for teaching students key media relations skills, namely identifying relevant news outlets and reporters and building media lists.
By integrating the Meltwater software into classes in public relations and social media through activities and assignments (see https://www.slideshare.net/CarolAnnFunkhouser for examples), professors can expose students to media intelligence software used by more than 25,000 organizations around the world (O’Malley Greenburg, 2017).
Apple, M. (n.d.). Meltwater content: How much data do we store? Meltwater: Welcome to the help center. Retrieved from https://support.meltwater.com/hc/en-us/articles/115015823988-Meltwater-Content-How-much-data-do-we-store-
Business.com. (2018, July 12). Meltwater review. Retrieved from https://www.business.com/reviews/meltwater/
Meltwater mobile. (n.d.). App store. Retrieved from https://itunes.apple.com/us/app/meltwater-mobile/id959583041?mt=8
Meltwater mobile. (2018, July 9). Google Play. Retrieved from https://play.google.com/store/apps/details?id=com.meltwater.mobile
O’Malley Greenburg, Z. (2017, May 23). Google slayer: Meet Meltwater, the company that conquered Google Alerts. Forbes. Retrieved from https://www.forbes.com/sites/zackomalleygreenburg/2017/05/23/google-slayer-meet-meltwater-the-company-that-conquered-google-alerts/#221215216422