SOCMINT (Social Media Intelligence)

What Is SOCMINT?

Social media intelligence is a broad term that refers to one or more organizations’ collection and analysis of data from social media. Often, the process is facilitated by the use of specialist software, but it can also be carried out through more manual approaches such as data interpretation.

There are several uses for SOCMINT, and some are more controversial than others. For example, it can assist fraud prevention, law enforcement, and due diligence on new contacts and associates. It can also keep brands notified when they are discussed online. However, SOCMINT can also be used for unethical and illegal purposes.

According to Finances Online, 93.4% of internet users and 58.4% of the global population use social media. Publicly shared information is a treasure trove of data to help investigations of various types.

Is SOCMINT the Same as OSINT?

SOCMINT (social media intelligence) and OSINT (open source intelligence) are related, but they’re not quite the same, as the former sources intelligence just from social media and the latter gathers intelligence from various sources that are open to the public, including social media.

OSINT is an umbrella term that includes open SOCMINT. Anything that people make public on social media sites counts as OSINT data, as it can be freely accessed and analyzed manually or with the use of automated tools.

However, there are elements of SOCMINT that can also involve data that’s not open source – for example, closed data accessed by surveillance programs, or data collected and analyzed in ways that users haven’t consented to or are otherwise illegal.

Some critics, including civil liberties groups, argue that certain uses of social media intelligence are intrusive and cross the line between public and private, though there is no clear consensus over this at present.

How Does Social Media Intelligence Work?

SOCMINT always begins with information collected from social media sites, manually or automatically. Data enrichment can both utilize SOCMINT data and add to SOCMINT investigations by enriching any primary data associated with social profiles.

On a very basic level, this could mean manually visiting someone’s profile and gathering publicly available information such as a user’s location, who they are connected with, their job title, or anything else that’s visible online.

Automation can take this further in a variety of different ways. A wide range of software solutions, both commercial and open source, can collect and analyze social media data for a range of purposes. 

For example:

• A fraud prevention solution that cross-references a customer’s email address with a wide range of social networks. This helps to build the individual’s social footprint and assess whether any red flags exist that could indicate fraudulent or otherwise dishonest intentions.
• An HR professional who reads a candidate’s LinkedIn and Twitter profile and posts to gauge their level of knowledge of and public involvement in their field.
• Market researchers utilizing social listening solutions to monitor social networks for product reviews, brand mentions, and related chatter.
• A law enforcement agent who uses social media to determine who a suspect’s closest contacts are, or where this person has checked in from to take into account when locating them in relation to a crime.

Social media data can reveal far more than many realize. For example, posted photos normally include metadata that shows when and where they were taken. 

Moreover, when data is collated and analyzed in bulk, it can point to overarching trends and deeper insights into a demographic, thus contributing to customer segmentation.

What Are the Elements of Social Media Intelligence?

Broadly speaking, the elements of SOCMINT involve the different steps one would take to make use of this investigation method:

1. Data collection: Any harvesting of data is included in data collection – for example brand mentions, or location information on one or more subjects.
2. Automation: Outside of small investigative scenarios, SOCMINT usually involves some kind of automation to facilitate bulk and relatively low-effort data collection.
3. Analysis: Also usually automated, the analysis of data allows organizations and individuals to gain actionable information that includes trends and patterns.
4. Distribution: This provides usable data and learnings to companies and organizations in the form of reports, dashboards and/or API integration.

Uses for SOCMINT

The uses of social media intelligence are not only vast but fast-growing. Practical uses for SOCMINT include: 

Social Listening for Brands

A company can use a social listening solution to monitor Facebook or Twitter for any mentions of its brand(s), which might then classified as positive, negative, or neutral. They can then act on this social media intelligence in various ways, from responding to feedback to assessing the success of a marketing campaign.

Forensics for Law Enforcement

Law enforcement agencies can use SOCMINT investigations to identify accounts that are spreading misinformation or extremist content. They can identify links between suspicious individuals and determine the location of a person.

SOCMINT Challenges

Although social media intelligence provides a rich, real-time, and ever-increasing source of data, its use can be controversial. This is especially the case when it goes beyond data that was publicly, openly shared – in other words, when SOCMINT goes beyond OSINT.

Privacy International argues that SOCMINT “requires more specific regulation, policies and safeguards that take into account the very unique and specific nature of social media”.

For example, only data that is fully accessible to the public can be considered open source, and therefore OSINT. An individual with a Facebook account typically makes considerably more information available to those on their friends list. If a law enforcement agent were to act on this knowledge by impersonating a close contact of a suspect, the question arises about what specific warrant, if any, they should be able to use.

Finally, there are also questions surrounding the accuracy of social data. Would relying on it mean accepting unverified information, just because someone stated this as a fact? In the age of deepfakes, can we even trust images, videos and audio posted on social networks?

These and similar questions are not as easy to answer as one might assume. There remain considerable gray areas around SOCMINT, as well as concerns to do with privacy.

Why Is Social Media Intelligence Important?

Social media intelligence is an extensive source of granular data, which is very often accurate, readily available and can be sourced in real-time. A non-exhaustive list of what SOCMINT research can do includes:

• assist law enforcement officers with evidence-gathering and other investigations
• help verify someone’s identity or act as a pre-KYC step to weed out fraudsters
• provide confidence that a job applicant is qualified and a culture match
• conduct alternative credit scoring or additional info for underwriters
• catch fraudsters masquerading as good customers
• source contact information for sales leads
• help with segmentation for marketing, such as by finding high-value customers
• provide demographics data to companies

The list of uses for SOCMINT is almost endless. Some people even abuse SOCMINT for malicious purposes, such as to learn when a house is empty or to find out more about an individual in order to impersonate them.

With so much of the world using social media and new platforms, social media intelligence will only grow in value.

How Can SOCMINT Prevent Fraud?

Companies can make use of social media intelligence to assist with automated fraud prevention strategies, as well as with manual reviews conducted by fraud analysts.

Such processes will typically utilize open SOCMINT, as this will help answer questions such as whether a suspect’s email address is registered on any social media, whether the individual has shared any pictures, their location if shared publicly, and so on.

Here’s an example of how it can work in practice:

1. A customer looking to buy something online provides an email address.
2. The address is automatically cross-referenced against a host of information that’s freely available on social networks, to answer qustions such as:
   • Does the user have any online accounts?
   • Are they based in the same location they’ve provided?
   • Are they an active user? etc.
3. This information facilitates intelligent conclusions and the raising of red flags. For example, an email address that isn’t linked to any social network whatsoever would usually be suspicious. So too would a customer claiming to be in the US and using a US-issued credit card when every linked social account shows a location in Asia.

The resulting data points will inform the person’s fraud score, from where the transaction can be automatically accepted, declined, or sent for manual review by a fraud analyst.