How to Manage Social media
Governance is all about how you manage and operate social media within your organization. It’s about connecting differing stakeholder groups and getting them to work together and pursue a common strategy and goal.
Everyone has a role to play in social media. Governance sets out what those roles are and establishes mechanisms that enable stakeholders to work together and achieve success while managing risk and monitoring progress.
Measurement of social media success is an important part of governance because it will inform decision making. In this blog, we will cover some of the key performance indicators that can be used to track and manage your social media programme. We will also discuss some of the more controversial ways that organizations have attempted to control social media risk.
Roles and responsibilities
Managing social media can be complex because of the wide range of stakeholders who have an interest in it. Defining clear roles and responsibilities makes it easier to visualize your social media landscape and to understand what drives different stakeholder groups.
If roles and responsibilities are not clearly defined it increases the risk that the business’s strategy will not be well executed and will increase the risk of something going wrong.
I like to define the stakeholders within three groups: those working for the organization; those external parties who are outside of the day-to-day operation of the company; and those that are transient and sit somewhere in the blurred line between inside and outside the company.
Within the company, there will be, undoubtedly, a number of departments or stakeholder groups who have an interest in social media use. The marketing department is probably the most obvious group who use social media but there are others who also have an important role to play, such as those in IT, Risk and Compliance, Recruitment etc.
Companies are structured and operate differently and sometimes you may find one person who plays a number of roles, while in others you may find whole teams of people responsible for one particular area with little crossover to other areas.
It’s important to remember that it’s not just those who have a formally defined role related to social media within a company who have a role to play; there are those outside the company too. Perhaps you have engaged an external PR company or social media agency to manage some or all of your accounts.
Your suppliers may be using social media and may wish to engage with you on it. Don’t forget that your competitors and their customers have a role to play too. Conversations and interactions on social media are often open to the public, which can be an opportunity for you to keep abreast of developments elsewhere in your industry.
Keeping an eye on what your competitors are doing in social media and how their own customers are interacting with them can give you real insights into the competitive landscape and alert you to issues sooner.
For example, if you know that both you and one of your competitors have a supplier in common and you begin to notify customers of your competitor complaining about an issue related to that supplier, it can act as a warning that can be flagged to your senior management, allowing you to plan for potential issues.
I like to describe employees and the connections and friends of employees like those in the ‘transient’ layer. They are obviously part of the organization but will probably not have a formal role to play in how it is managed.
The reason that I describe them as ‘transient’ is because, while they may follow a company’s social media accounts and occasionally share some of its content or post links to new products or services, they will have other personal interests that they share on social media.
Marketing is often the owner of social media because social media is traditionally seen as a medium that can be used to market products and services and connect a business with its consumers. Marketing will be interested in improving brand loyalty and driving sales and will focus on hitting its social media targets.
Communications professionals will want to use social media to post information to the public or to their customers but not necessarily with the goal of driving sales. Instead, they will be more focused on informing them of something.
Communications professionals will use traditional social media to broadcast information publicly or announce press releases. They will use an enterprise social network to broadcast messages or news internally.
Human resources (HR) professionals will be the guardians of the policies that employees must follow, including social media. HR will want to ensure that employees are not spending excessive amounts of time on social media when they should be working and that they are not using social media inappropriately, by engaging in cyberbullying or leaking company information.
The IT department is the technical owner of an enterprise social network and is responsible for its accessibility and how it is used. IT will control how employees can access an enterprise social network, as well as whether or not employees can access traditional social networks and for how long each day.
Nowadays, many people find it easier to simply send a tweet to a company instead of calling customer services to discuss some kind of issue or complaint. Many companies, particularly in the United States, have realized this and have trained up their customer services agents to use social media to engage with and respond to customers.
Many recruitment agencies already use LinkedIn aggressively to headhunt candidates for the jobs they are trying to fill. Employees of a company may also use social media to advertise positions that are open at the company.
Recruitment managers are using social networks to do extra research into potential candidates, and many now feel that a candidate’s LinkedIn profile can be a fairer reflection of a candidate because it has not been tweaked for a particular position, as CVs are.
The concept of social selling has emerged and we’re now seeing salespeople turning to social media to makes sales. Social media cuts down the amount of travel and face-time that has historically been required to sell a product. Sales professionals now use social networks to identify their targets, engage with them and build relationships online as a pre-sales activity.
The job of the knowledge worker is to ensure that information can be found by those who need it and is a role that is common in knowledge-centric businesses such as consulting, engineering or law.
Knowledge workers historically relied on archaic knowledge databases or repositories, however, the rise of enterprise social networks has meant that they have turned their attention to encouraging knowledge sharing through these networks because it can be far more effective.
Risk, compliance and legal
The risk and compliance professionals within an organization will, more likely than not, have an opinion on how social media can, or should, be used. These people have a thorough understanding of any regulatory requirements that affect your business, and it’s their job to ensure that these requirements are met.
Risk and compliance professionals wield a lot of power when it comes to what can and cannot be done in the course of business and it’s important, therefore, to engage them early in any plans to implement an enterprise social network to ensure that any specific requirements are included.
Organizations are waking up to the fact that social media poses a risk to the business. Social media is now being listed on the risk register and internal auditors have begun including it within the scope of internal audits.
They review the policies, procedures and other documentation related to social media in order to obtain reassurance that social media is being managed appropriately.
Public relations or social media agency
Many organizations outsource the management of their social media presence to external agencies instead of employing in-house teams. In many cases, the external agency has complete control over the organization’s social media accounts. This means that the organization will need to provide clear guidance about what the external agency can and cannot do in social media.
Your competitors may be active in social media, perhaps more active than you are. They will be monitoring your posts and may even be analyzing your own social media engagement to see if they can convert some of your followers to their own brand. You need to be aware of what your competitors are doing so that you can defend yourself if needed, or to exploit any potential opportunities.
Your investors and shareholders may have an interest in what your organization is doing on social media due to the impact that social media can have on your reputation and your share price.
There are often strict legal or regulatory rules that govern how certain information can be released to investors or the public, so it’s important that you are cautious about how you use social media to publish investor information.
By monitoring your suppliers’ social media feeds you may be able to spot any issues in advance and implement any necessary contingency plans. Social media is the most effective way to detect any widespread negative sentiment about one of your suppliers. It may also make you aware of any potential ethical issues that your suppliers are facing and that goes against the values of your own company.
Your customers play an important role in social media. Your customers may be reviewing your products or services online and will be able to influence other customers when it comes to the decision to buy your product.
Media or news organizations might be following your company to report on any newsworthy events. This also means that if you make any embarrassing mistakes in social media, they will hit the mainstream news faster than you expect.
Your regulator plays an important role in social media as they set out the rules with which you must comply, some of which will have an impact on how you use social media.
Ownership and sponsorship
Social media needs a clear owner who takes responsibility for and is accountable for, social media. This is important because of the wide range of stakeholders who have an interest in social media at any given organization.
It is also important because in times of crisis or when decisions need to be made quickly, having someone with authority will greatly improve the speed in which any issues can be resolved.
It will depend on your organization as to who exactly takes ownership of social media, but the key thing is for someone to take this responsibility and be held accountable for its use.
Social media sponsorship is when someone in a senior position at a company, preferably on the board of directors, pledges support, or ‘sponsorship’, for a social media programme. The sponsor will provide support, funding, and champion the programme at the senior levels of the company.
Having a social media sponsor will give more weight to how social media is perceived and how it is operated. Social media is a board-level issue, and the board should be aware of it and have comfort that it is being managed appropriately.
As we have seen, social media spans multiple departments, which can make it difficult to agree which department should take the hit to their budget for any investments that are needed. Having a board sponsor will also make it easier when it comes to budgets and investments, because the sponsor may be able to approve investments in social media, which will benefit multiple departments.
Working groups and oversight
When it comes to managing social media on an operational basis, it’s good to regularly arrange working groups where key representatives of each stakeholder group can meet to discuss what’s happening in social media and give some oversight to any potential issues.
The purpose of the meetings is to flag any new developments such as new accounts, new product launches, major events, press releases or risks that the organization may face and which may permeate social media.
Having a community that meets regularly like this will make it far easier to manage social media effectively, and the representation from other stakeholder groups will allow issues to be raised and debated collectively to ensure that appropriate actions, in line with the business strategy and risk appetite, are taken.
Creating an email distribution list for all attendees or stakeholders is worthwhile in case there are absences from a meeting or in case certain people are out of the office at the time of the meeting.
Any communications can also act as an audit trail to show that issues were raised and addressed, should there be a need to do so. A virtual private group within an enterprise social network can also be an effective way of both cutting down on email traffic and allowing better collaboration and communication around specific issues.
Regular meetings like this are also an opportunity for metrics to be tracked and assessed so that performance can be measured and any performance issues or risks can be addressed early to allow remediation plans to be implemented.
The use of a RAG (Red, Amber, Green) status is a common way of assessing progress against objectives or the potential impact of any given risk. There will be a lot happening at any given time and the use of a RAG status is something that, when used correctly, can help focus attention on the things that matter most.
Be warned, however, that RAG statuses should be used honestly. I’ve seen progress dashboards in the past that have shown ‘Green’ for weeks when, in fact, there have been problems and an ‘Amber’ rating would have been more appropriate, but some people have a tendency to manipulate the status to hide those issues.
I recommend that if a status is challenged during one of the working groups and the consensus is that it should be changed, that the dashboard is updated and retained as an audit trail.
A good way of keeping the board sponsor engaged in what’s happening in social media is to present a monthly dashboard to them. This will ensure that social media continues to get the oversight it needs and will give the board sponsor the opportunity to comment on any issues or developments that need their attention.
Approvals and processes
Any organization that takes social media seriously should have well thought-out and documented processes and should understand what sorts of activities need approval. What will need approval will depend on the nature of your business and its culture.
For example, if you operate in a highly regulated industry, such as banking and finance, you will probably have more stringent approval processes around what content you post on social media and how you engage with people on social media than a business operating in a less regulated industry.
Seeking approval can take time, and in the fast-moving world of social media, being forced to obtain approval for social media posts can hamper how effectively an organization engages in social media. This may cause a risk that a response to any particular incident could be slowed considerably, causing the incident to gain prominence.
I think the best balance is to allow a specific team or group of individuals the ability to post to social media without having to go through an approval process. These individuals, of course, should be well trained and experienced in understanding the impact that posts can have.
They should also attend the working groups and have regular contact with other key teams within the organization to ensure that they are aware of any major developments or potential risks.
Many social media risk and compliance platforms can be configured to support your approval processes. You can grant permissions to a subset of users so that their posts do not need to be approved by another user. Many platforms also allow you to configure rules so that, for example, any use of a specific term forces the post to go into an approval queue.
Having resilient processes with agreed timescales is a good idea. It’s worth always thinking of the ‘what if’ so that you can understand where there might be a breakdown in the process if, for example, one of the approvers is unavailable, sick or in a meeting when they’re needed.
One of the tasks of the working group should be to design and document the process for the approval of content. Because the working group has representation from the key stakeholders, it should own the process and, once finalized, the documentation can be distributed to the wider social media stakeholders for reference.
I recommend that a document repository is created where all of the documentation related to social media processes and policies can be stored. The repository could be as simple as a shared network drive or as complex as a document sharing platform such as Microsoft Sharepoint.
It could also be stored within a private virtual group on an enterprise social network, which would allow comments and suggestions to be posted. What’s more, most enterprise social networks keep a version history, making it easy to look back at previous changes.
What is moderation?
In social media, moderation is the act of removing or censoring posts that appear on your various pages. This could be moderation of the comments that appear on your videos on YouTube for example or moderation of the content within your own enterprise social network.
Moderation is an interesting topic because there are a number of different viewpoints and it can be difficult to agree on what should be moderated and what shouldn’t.
On one hand, the comments that some social media users post can be abusive, threatening or contain many unsavory uses of language so you could argue that this content should be removed as it may offend other users. On the other hand, the removal of posts or comments can be fiercely opposed by some people who see it as censorship and an infringement of free speech.
It’s understandable why it can be tempting to moderate posts if they contain false accusations or information, or if they reflect badly on your organization or its brand. Likewise, if inappropriate information or comments are posted on your enterprise social network then you will have a responsibility to respond to it and may be required to remove it.
The most transparent way to moderate social media is to publish guidelines on what is acceptable and what is not, and moderate strictly according to these guidelines. Another effective way of performing moderation is through peer moderation, which we will cover in the next section.
Peer moderation is performed by the users of a social network. You rely on the users to moderate posted content by giving them an option to mark certain posts as inappropriate or abusive.
It’s an effective method of moderation as it means that you, as moderator, only need to review those comments that are flagged by other users as abusive. You can review and then make a decision on what action to take, which could be to allow the comments, censor them in some way such as by replacing swear words with asterisks or to remove the comment altogether.
When rolling out an enterprise social network it can be advantageous to implement peer moderation as it means that content does not need to be approved before it is posted. Configuring your system to instantly remove content that is flagged as inappropriate until someone has reviewed it is a sensible way of moderating content.
You will need to consider who will receive alerts and should consider global time zones and public holidays if your organization is global because the content may have been marked in error and your users will expect a quick turnaround.
If you have engaged an external PR or social media agency you should consider what timescales you want the agency to respond to content which has been marked as inappropriate. Ensuring that these terms are written into your contract should give you some comfort that the accounts are being managed appropriately.
Automatic moderation with dictionaries
A common way of automatically moderating content is to use a dictionary of terms that the application will look up to ensure that none of the users’ input contains any disallowed terms.
This is a nice way of catching swear words or another offensive language, which you may want to keep off your enterprise social network and indeed most networks allow a dictionary to be configured.
Some dictionaries of offensive terms are available online and some come pre-bundled with the application or network you purchase. Some social risk and governance tools can also be configured to look up posts against a dictionary of terms. It’s often possible to configure the tool so that certain words will be automatically blocked, or for certain words to force the post through a review process.
While dictionaries are useful to identify and remove offensive language, they also have other uses. For example, if you are going through a sensitive deal such as a merger or acquisition, and you do not want your employees to discuss or speculate on the deal, you could add the name of the other party or terms related to the deal to your dictionary as a way of stopping posts related to the deal going onto your network.
You may have made a policy decision not to allow clients to be discussed on your enterprise social network, so adding your clients into the dictionary of terms would be another neat way of controlling their use.
Some industries are obviously more regulated than others and the regulator may have specific requirements around how certain work can be carried out. Adding keywords into a dictionary to stop certain topics from being discussed, such as employee bonuses or other HR issues, can be effective at managing these risks.
Be careful not to add common words, however, so that you don’t end up inadvertently discouraging people from using the network. And don’t forget that if you have implemented peer moderation (by the use of an ‘inappropriate’ button, for example) you are already on your way to managing the risks of the content posts by your users.
A word on censorship
Being in control of the content posted about your company on social media or within your enterprise social network is clearly important as it can expose you to a wide variety of risks if not monitored and managed effectively. That said, it’s wise to think very carefully before deleting any content.
The saying goes, ‘think before you post’, but ‘think before you delete’ is equally as relevant. When it comes to censorship, many organizations have learned about the wrath of social media users the hard way.
Many people fiercely oppose any type of censorship and indeed the question of online freedom continues to be hotly debated all over the world. There are many, many examples of companies trying to control what is said about them online, only to find the strategy backfires spectacularly.
From the small bed and breakfast that issued fines to its guests who wrote negative feedback on TripAdvisor to huge multinational companies who try to hide information from the public, censorship in any form can be dangerous and can cause a somewhat small anecdote of bad service into a full-blown national news story.
Questions around copyright are often raised when considering social media or marketing in general. While many marketing professionals will be aware of the constraints to the use of copyrighted material, others in your organization may not be so well versed in the risks of copyright infringement, particularly when it comes to the use of imagery.
If you’re looking for an image to add to a report or a presentation, it’s a bad idea to simply use Google Images. It can be hard to determine who owns the copyright for any particular image and whether or not you are allowed to use or reproduce it.
If you take a photograph of a beautiful sunset while you are on holiday, for example, you own the copyright for that image. Nobody can use that image for any purpose unless they have obtained your permission to do so.
This may seem pedantic, but when it comes to business, infringing copyright can be a very costly mistake to make. There are a number of ‘stock imagery’ websites where you can buy images for commercial use.
There are often a number of different licenses available depending on how you want to use the image, and the costs of the image may depend on the physical size (in pixels) of the image itself.
Some licenses will allow you to use the image only online, another license may extend the terms to include print, some licenses might be for specific countries, and others might be worldwide.
There are tools available that can search the web to find instances where your own images have been used – if the person does not have your permission to use the image, you can request that they remove it, or in some circumstances even request compensation.
Social media is a great place to share your marketing material but, the point here is to ensure that you own the rights to use whatever imagery you choose in whatever fashion you choose to use it.
The topic of copyright extends further than just images and is just as relevant when considering the content your users share within an enterprise social network. Your users may come across reports or research related to your industry and may decide to upload the report to your enterprise social network to share more widely with their colleagues.
However, the user should check for any copyright marks on the document, and if it is copyright protected they should seek permission from the copyright holder before uploading it. If they don’t you may find that your enterprise social network turns into a repository of illegally hosted copyright material.
In my experience, most users don’t realize that by uploading a document to an enterprise social network they are actually creating a copy of the document which may infringe copyright.
The simplest solution to this is to educate the users of the risks of copyright infringement and encourage them to submit a link to the report or document, rather than uploading the actual document itself.
If it comes to your attention that users have uploaded copyright information, I recommend that you contact them, firstly to make them aware of how copyright can be infringed by uploading documents to an enterprise social network, and secondly to ask them if they have obtained the permission of the copyright holder to store the material on your network.
I would include a note to say that you will automatically remove the content if you do not receive adequate assurances that the material does not infringe copyright within 30 days.
If the content is within a group, it is worth copying the message to the group owners as well. More often than not, the user will simply acknowledge the mistake and replace the document with a link.
Believe it or not, data can be of a good quality or a poor quality. At first, this may seem an odd concept as surely data is just data…? But no, data is not just data. In today’s society, data is being created constantly.
That includes such things as when we pay for something using our credit card, when we send messages to each other, or when we ‘like’ a social media post. So, what is data, and is it the same as information? The following is an example of some data:
31, 30, 31, 30, 31, 31
On its own, data is meaningless and is, therefore, not information – it’s just some data. The data above becomes information when I say that it is the number of days in March, April, May, June, July, and August.
Incidentally, you’ll notice that I deliberately started from March as February would have just over complicated matters…!
It’s true that the word ‘data’ is often used interchangeably with the word ‘information’, but for the purposes of this explanation let’s think about them as being separate. Information is data with a context – without a context or any way of knowing what the data is, it’s just considered data.
The problem of data quality comes when we start thinking of larger amounts of data. If you asked 100 people to fill in a form with details of where they live, you might find that a large number of respondents reported living in London.
So, it’s logical to think that if you wanted to analyze this data in some way, you could say ‘Show me all of the people who live in London’. However, you might find that some people typed ‘london’, others ‘LONDON’ and some might have mistakenly entered ‘Llondon’.
In this example, we have quite a lot of dirty data and need to perform a data cleansing exercise. This is the process of cleaning the data and making it accurate and in the same format.
While a human might be able to compare ‘London’ and ‘Llondon’ and deduce that it is the same place, a computer cannot. Thus, if you look at how many people live in ‘London’ in this example, a computer may give a lower number as it is not intelligent enough to work out that the two are, in fact, the same.
So, data quality is all about how uniform and accurate any given data set is. Other issues can exist too;
For example, just because five people stated that they live in ‘London’ and that ‘London’ is in the same format within the data set, it does not mean that the data can be fully relied upon – someone may well have answered London because they misread the question in some way or because of the difficulty of defining geographical boundaries.
For reference, in data analysis, a margin of error would normally be applied to counter this and make the results more accurate, but this is not relevant to this particular topic.
The reason that data quality is important is that, in today’s world, we are all far more reliant on data when making decisions than we were in the past and, indeed, companies make many business decisions based on data.
If a certain dataset (for example, sales data) was being relied upon for decision making (say, to compare with marketing spend), but the sales data was of poor quality (perhaps it included duplicates), then the decisions made based on this data are likely to be poor and have negative consequences for the business.
There is a vast amount of data within social networks and there are many tools and packages available to help you analyze that data and gain insight to help you make decisions.
It’s important, therefore, that you understand the data you are analyzing as well as the quality of that data. Ensuring that you maintain good quality data offers many benefits. When it comes to negotiating budgets, having confidence in your numbers and being able to cut the data in different ways to show different insights can be extremely valuable.
Later in this blog, we will look at how you can measure the success of a social media programme and what metrics you can use. A good consideration and awareness of the quality of your data will help cut down on the number of unexpected surprises that you may face.
It’s likely that you’ll have a number of key performance indicators (KPIs), which you will use to assess how effectively your social media programme is running. Understanding the data that drives these KPIs will give you greater confidence and more reliable insights into the data.
Types of data: structured and unstructured
There are two types of data: structured data and unstructured data. Structured data has defined attributes and is presented in a consistent manner. Data within a database or a spreadsheet is considered to be structured data as it is held within various fields within a table or column in a spreadsheet.
Therefore, a postcode would be stored in a postcode field, a name would be stored in the name field, and so on. Sometimes structured data will include references to other fields or pieces of data;
For example, an order reference number may refer to order data in another database table or spreadsheet. Structured data is, therefore, usually stored in some kind of logical and uniform fashion.
Unstructured data, on the other hand, usually refers to data where there is no structure or logic. For example, a field containing ‘I’m reading a great book about social media risk and governance’ would be classed as unstructured.
It’s not possible to easily analyze the contents of the field because it could contain absolutely anything – there is no structure to it. While the format of a postcode is standard, the contents of a text field, or a social media post, is not standard.
Sentiment analysis is the term given to the analysis of a social media post in order to determine the ‘sentiment’ of the post. The sentiment is usually expressed as positive, negative or neutral, or is displayed on a scale, to describe the mood.
The purpose of sentiment analysis is to determine whether social media posts about any given term or criteria are generally positive, negative or neutral. Therefore, if you want to determine the overall sentiment towards your brand during a particular timeframe you could use sentiment analysis.
Many people try to use sentiment analysis in order to measure how people feel about a brand or company. Sentiment analysis is inherently difficult to do accurately due to the complexities of the languages we speak.
This is mainly due to the fact that the data you can analyze from social media posts are usually in an unstructured format. For example, a post that says ‘I love the products from company X’ would show positive sentiment, ‘I hate the products from company X’ would show negative sentiment, and ‘I just bought a product from company X’ would show neutral sentiment.
The complexities in the languages that we speak can make it very difficult for a computer to understand the real meaning of the words. Sarcasm is notoriously difficult for computers to understand.
Most of us would probably understand a post saying ‘massive queues at all of the tills – brilliant!’ to be negative, but computers have a tough job identifying such linguistic nuances and may class this as positive due to the use of the word ‘brilliant’.
There are many applications and platforms available that claim to be able to analyze sentiment towards a brand, however, you should be careful how much trust you put into the algorithms that drive the sentiment scores due to the inherent difficulty of language analysis. If you use sentiment scores as part of your regular metrics you should be wary of how accurate (or inaccurate) the score can be.
CASE STUDY Risk in action: UKIP – sentiment analysis difficulties
The United Kingdom Independence Party, commonly known as UKIP, is a euro-skeptic, right-wing political party in the United Kingdom. In May 2014 the party decided to use a hashtag, #WhyImVotingUKIP, to give their supporters a way of voicing their support for the party and give them the ability to articulate the reasons for their vote.
It all started innocently, with UKIP members and supporters tweeting about why they were voting for the party. Unfortunately for UKIP, however, the idea backfired and the hashtag was quickly hijacked by other Twitter users who used it to poke fun at the party. Many tweets marked with the hashtag #WhyImVotingUKIP contained obviously sarcastic, and sometimes offensive comments.
I don’t wish to comment on UK politics, however, the hashtag hijacking was fairly easy to predict by anyone who regularly uses social media.
This example of a hashtag hijacking effectively illustrates how difficult it is to accurately analyze sentiment. When I read down the tweets containing the hashtag it was generally clear when I was reading a sarcastic tweet. Other tweets, however, were more difficult to determine and required me to look at the user’s profile to try to determine whether they were being sarcastic or not.
The point is that if a human is finding it difficult to determine whether something is sarcastic, and determine whether it includes positive or negative sentiment, then it must be considerably more difficult for a computer to accurately assess the sentiment.
Examples of Twitter hashtag hijackings are common and are often predictable, yet companies still fall foul of them. Social monitoring tools can give you a good awareness of current perceptions of your brand or company and should inform your decisions on which hashtag you create.
Taxonomy is the way in which social media content or posts are classified. In traditional social media, the most common tool for classifying content is through the use of hashtags, which are preceded with the hash (or number, or pound) symbol, ‘#’.
The idea of the hashtag is to make it easier to find content or posts related to a specific topic because you can search for the hashtag and pull up all posts that contain it. There are no rules as to what a hashtag can be – social media users have total free reign to make up whatever hashtag they wish.
Many organizations create specific hashtags to accompany marketing or awareness campaigns about their products or services. You may have also noticed that it’s becoming popular for a hashtag to be shown on screen at the start of some TV programmes to allow viewers to discuss the programme on social media.
If you are considering what hashtag to use, it’s worth doing some research and getting some opinions. You should search the major social networks to see if the hashtag has been used before and, if it has, in what context, so that you can avoid accidentally associating your organization with something unsavory or irrelevant.
It’s also worth looking to see if there are any other meanings of the hashtag in question. One entertaining example of a hashtag gone wrong was when it was used to promote the launch of the album of UK singer and Britain’s Got Talent contestant, Susan Boyle.
The hashtag that Susan’s PR company chose to use to promote her album party was #susanalbumparty. Unfortunately, instead of reading #SusanAlbumParty, it could also read #SusAnalBumParty. Hashtags don’t include apostrophes and so it quickly got noticed by many social media users and was reported in the media, causing it to be changed shortly afterward.
If you’re planning to use an acronym as your hashtag it’s worth doing some research to ensure that there aren’t any rude connotations that might cause harm to your brand.
Content within an enterprise social network also needs to be classified in some way to help the users of the platform find what they are looking for. Most networks allow content to be shared within specific groups and categorized in some way.
Often group administrators have the ability to create their own categories within their group so that new content can be marked as part of a specific category.
Furthermore, most networks allow content to be tagged with keywords or hashtags. These tags are important because when a user types in a search term, the tags that have been assigned to content will drive the search results.
Some users will find it frustrating to have to type in tags or keywords when they post content to an enterprise social network. However, including instructions and guidance for content categorization and tagging within your training will encourage proper use of tags, and will help your users find the content they are looking for more easily.
Like traditional social media, tags within an enterprise social network are not usually controlled, so any user can choose any tag they wish for a piece of content. This can cause problems, however, because what might seem like an obvious tag to one person may not be so obvious to another.
Some users may use abbreviations for their tags, others may use the full text. In this scenario, there may be two tags which mean the same thing, but which are spelled differently.
This links back to the issue of data quality because having content which is not categorized in a standard way will mean that the data is difficult to search or analyze. Therefore, you should give consideration to how your taxonomy will be managed. Things that should be specified include:
how many tags should be created; whether or not abbreviations should be used; and how many words a tag can consist of.
This guidance should be available to the users of the network so that they can refer to it and create correct tags from the outset. Some users, of course, will ignore the guidance and create their own tags that do not conform to the standards outlined in your taxonomy.
When this happens, you may find it worthwhile to clean up the tags. Your chosen vendor should be able to provide you with extracts and mappings so that you can more easily view all of the tags that are being used within your network and the number of items they are tagged to.
This will help you identify which tags need to be updated. By completing this exercise regularly, it will help your users to navigate the content within your network more easily and, in turn, find the content that they are looking for.
Social media is a place where conversations are happening every minute of every day, with people all over the world discussing any topic that they are interested in. There’s a huge amount of data being generated all of the time which can make it difficult to understand how to monitor the conversations which are relevant to your organization.
Monitoring, or ‘listening’ as it’s often called in social media, is an important and ongoing task. As I have said before, social media is not a broadcast channel for your marketing material, it is a place where people connect and have fluid conversations.
You obviously can’t monitor every post in social media so to start off you should complete an assessment to work out which people or organizations are engaging in content related to your own brand. You should determine who the influential users are. Influential users could be celebrities, other prominent figures, journalists or bloggers.
These people may have a large following and they may also be followed by other influential people. By performing an assessment of the conversations taking place in social media you’ll also be able to tell what topics resonate more with social media users within a specific network.
This valuable information can then flow into your content strategy to ensure that when you do post content and engage with social media users that your posts are relevant and of interest to your targets. By monitoring social media effectively you’ll be alerted to any potential issues or opportunities much more quickly, and in turn be able to promptly address the issue or exploit the opportunity.
A lot of social media is about timing, similar to the news. News that’s not current is no longer news, and people tend to engage on social media about things that are current and happening now.
Monitoring and listening tools
There is a very wide range of social media listening tools available on the market. Some of the popular platforms are Salesforce Radian6, ViralHeat, and Talkwalker, however, there are many, many more options available. Many social media management platforms include some analytics and listening functionality; some for free, some at a price.
The key things to consider when assessing which social media listening platform is right for you are your requirements, your current infrastructure, and your future ambitions. It may take time to implement and configure your chosen platform so you will need to document your requirements and perform an assessment of the tools available. Language is another important consideration.
If you operate globally and have multiple accounts that post in foreign languages, you’ll probably want to purchase a tool that can handle these complexities.
Some of the tools on the market only work with specific social networks or work more effectively with certain social networks than they do with others. If you want to perform historical analysis on trends over time, you may want to purchase a tool that gives you access to the so-called Twitter Firehose.
The Twitter Firehose guarantees 100 percent of tweets for a given criterion, whereas the Twitter Application Program Interfaces (APIs) impose limits on the amount of data that can be processed. An API is a set of routines, protocols, and tools for building software applications.
Developers can easily access the Twitter APIs for free whereas access to the Firehose can be very costly. In summary, it’s important to consider your monitoring requirements in advance to ensure that you make the right choice when purchasing a monitoring tool for your organization.
Performing social media analytics can be a time-intensive task as it often requires somebody to research the reasons for any significant changes or unexpected results. The social media landscape also changes quickly – the people that you engage with may change too, and new influencers may appear and others disappear.
To keep on top of this I recommend giving someone the responsibility for social media analytics to ensure that the task gets the focus it deserves. A little later we will look at the metrics and performance indicators that you might choose to track.
Monitoring employee use of social media
The question of whether employers can or should be able to monitor their employees’ social media usage is a complex one and often something that requires specific legal advice.
I’m not going to address the specific legal points to employee monitoring here as I’m not qualified to do so, however, we will explore some of the issues more generally. Some other legal or regulatory issues, such as the ownership of social media accounts or the use of social media in recruitment.
In short, it is technically possible to monitor employee social media usage, in the same way, that it is possible to monitor email or internet usage, and there are commercial tools on the market which will allow you to do this. I believe that the real issue, however, is one of ethics.
I question why an employer would want to monitor an employee’s social media usage, or why they feel that they have the right to in the first place. Up-to-date policies for social media or IT usage are less intrusive forms of controlling social media use.
Within these policies, you should clearly set out what is acceptable and what is not acceptable. While it is understandable that employers will want to protect their organization’s reputation, as well as protect their other employees from things such as cyberbullying, most of this can be done using policy controls.
One of the common reasons cited by organizations that want to monitor employee use of social media is to counter the loss of productivity. Some employers choose to block social media sites on company devices to try to stop their employees from accessing them.
Nowadays, however, most people have access to their social media networks through their personal smartphones, so I question how effective blocking social media on work computers actually is. Outside of work, of course, it’s highly likely that your employees will use social media and may indeed make posts related to their work.
Some employees have found themselves in trouble because, having phoned in to report that they are sick, they then post photos of themselves having a nice day out somewhere. If a post like this is reported to the employer and investigated, the employer may well be within its right to discipline the employee in question.
However, if the organization had been secretly monitoring its employees without having received adequate legal advice, and without appropriate policies and controls in place, it gets trickier.
Privacy is a topic many people feel strongly about, so any tactics that may be seen to infringe someone’s privacy are likely to be met with hostility. Privacy policies are a way of setting out what an organization does with the data it collects about employees.
Another challenge that employers face in employee monitoring is complying with different laws around the world. If your company is global you will need to ensure that you comply with all the necessary local laws. There are, for example, considerable differences between what is acceptable in the United Kingdom and what is acceptable in some of the individual states of the United States.
After receiving independent legal advice, if you do decide to engage in social media monitoring of your employees, you should do so in an ethical manner. This means that you should be open about the purpose of the monitoring, and what you’ll be monitoring. Monitoring employees through deception is considered to be unethical and should be avoided.
An example of deceptive techniques for employee monitoring might be creating fake social media accounts that you use to request a connection with an employee for the purposes of monitoring them. In November 2008 a 31-year-old Swiss woman who worked for Nationale Suisse called in sick one day due to a migraine.
She reportedly accessed Facebook from the bed and was then fired by the company due to a breakdown of trust. The company claims that a co-worker who she was friends with her had noticed and informed management. But, the woman believes the company had created a fake account, made friends with her through it and used the account to spy on her.
In general, I would discourage companies from monitoring their employees’ social media usage unless there is a specific need to do so, where the employee is aware of the monitoring and where it is ethical. If you plan to implement a tool to monitor employee social media usage I strongly advise that you seek legal advice before doing so.
Metrics and performance indicators
Accurately measuring the return on investment in social media is difficult, but there are a number of metrics that can be monitored to measure the success of a social media programme.
KPIs fit into four main categories:
Reach. Reach describes how many users see the content that you post, which drives awareness over time.
Engagement. This category defines those actions performed by users where they are actively engaging around you or your content.
Influence. Those with authority have more influence than others. If you are seen as a leader in your field then you will have a greater influence.
Advocacy. When users actively support you and recommend you, your brand or your content to their own connections, this is classed as advocacy.
It’s important to monitor the success of your social media programme regularly and to report the metrics to social media working groups as well as the programme sponsors.
This can focus effort in the right place to improve poor metrics and help to ensure that focus is directed in the right place in any given time period. It may well be that, over the course of a year, certain metrics are more important than others.
For example, at the start of a marketing campaign, awareness may be more important than engagement; however, towards the end of the campaign, advocacy may be more important.
Often, the objective of social media posts is to drive users to your own website in order for them to either engage in some of your content or purchase products or services.
Social media metrics should not be looked at in isolation; instead, they should be reported along with broader website metrics. By linking the two metrics you’ll be able to get more insight into where your users are coming from and which posts are more successful than others.
By regularly reviewing the metrics, over time you’ll begin to understand which pieces of content drive certain actions. This information can be extremely useful and will go a long way to helping you reach your goals.
Social media doesn’t sit still, though. Your followers will change; posts that might have high engagement rates one month may not generate the same interest the next month. This may be due to outside factors, such as popular news stories at any given time.
This emphasizes the importance of having a robust social media monitoring programme in place, as discussed earlier in this blog so that you can constantly tweak your posts and hit your targets. The fact that a post gets X number of retweets means little on its own.
You need to analyze why that was, what it was that engaged the users, and which users were engaged by the content. Sometimes content is retweeted for the wrong reasons.
When mistakes happen and something gets posted by accident that causes embarrassment to a company, often it will be retweeted or shared by a huge number of people in a short space of time.
In isolation, it may look like your performance indicators for engagement are very successful. Until, that is, you delve into the detail and uncover the real reason for the high number of engagements.
Note that some metrics can be part of multiple categories and that the ultimate category is subjective.
With so much information available, it can be difficult to visualize performance. This makes it challenging to ascertain whether or not a social media programme is reaching its targets.
For those who are aware of social media and its importance, but don’t necessarily understand how each network works or how users are engaging with an organization, it can be quite intimidating to see a long list of metrics.
To bring the metrics to life I recommend using data visualization techniques and dashboards. This can be as simple as plotting a graph or chart using your metric data, or more complex, such as a tool which allows users to drill down into the data and manipulate it in different ways.
Once you have an agreed set of metrics and performance indicators you will report on regularly, then you can begin to design how they will be displayed. Many social media monitoring platforms come with ready-made visualizations; however, you are then constrained to the visual representations created by the vendor of the platform. To overcome this, you should build your own dashboards to extend the functionality and reports available within your chosen listening platform.
The first step to building a dashboard for your social media metrics is to define the data you will use. You want to get to an end result where you have a dataset or spreadsheet containing all of the data you will use to calculate your metrics.
Once the format has been established it should not change, unless you decide to tweak your reports at a later date, at which time you may need to update your dashboard too.
There are a number of good data visualization tools on the market, the main ones being QlikView, Tableau, and Spotfire. All allow you to build dashboards that you can use to display and manipulate data by applying filters, date ranges and so on.
You should consider how your data will be refreshed, meaning how often you will take a new cut of the data and load it into your dashboard. The advantage of building your own dashboard is that once created, all you need to do is load an updated dataset in order to refresh the dashboard.
It’s possible to get deeper insights into your audience by looking at splits in gender, location, age, and so on. Many of the common social networks provide analytics and allow you to export data about your posts.
These extracts often include information about the demographics of your audience, such as gender or geographical location, which is an extra insight that you can build into your dashboard.
In the pursuit of targets, it can be tempting to manipulate the metrics in some way to make it look as though performance is better than it really is. This is not only unethical, but it’s also a bad idea in the long term and may hamper your future success.
One of the most common ways that social media users, as well as companies, manipulate metrics in order to appear more successful is through the purchase of followers or likes.
There are a large number of people out there who will increase your total number of followers for cash. However, the new followers are likely to be ‘bots’ or fake accounts created solely for the purpose of increasing the total number of followers an organization has.
Firstly, this practice goes against the terms and conditions of most social networks, and engaging in it increases the risk that your account will be banned. Furthermore, it is relatively easy to spot accounts that have paid for followers, because they may have a suspiciously high number of followers in relation to the amount of content posted.
It’s also possible to analyze an account’s followers to gain insight, such as the geographical location of the majority of followers. The location can imply that followers have been purchased because many of the people who offer to increase your followers control thousands of accounts in specific countries – India, for example.
If your organization does not operate in India and you have a strangely large number of Indian followers, it’s a sign that followers are likely to have been purchased. So, why does this matter? Well, it shows that you engage in practices that are based on deceiving other social media users into thinking that you are more influential than you actually are.
This will likely cause a breakdown in trust between you and your target social media users, which may backfire and cause users to turn against you and your social media programme due to this unethical behavior.
Purchased social media followers are not real, and therefore will not engage with your content. Therefore, your metrics will suffer because your total number of followers will increase but the ratio between followers and interactions will decrease.
Building a strong social media presence takes time and shortcuts are likely to hamper your efforts and harm your reputation. By executing your strategy and engaging in ethical social media practices you are more likely to build a stronger social media following and, over time, reap rewards far into the future.
Documented operating procedures are an important element of good governance as they set out the processes that should be followed for certain tasks. For example, what is the process for creating a new user on your enterprise social network or social media management system?
Having processes like this clearly documented will make it easier for others to complete their tasks quickly and accurately. It also means that if a regulator or enforcement body requests information about how you manage social media risk you’ll be able to provide the documentation that they need easily.
You should also think about the processes for data, other than just those processes that require human interaction. Earlier in this blog we looked at data quality. By documenting processes that describe how data flows from one application or system to another, you’ll be able to pinpoint issues far more easily.
Having up-to-date documentation about your archiving processes will help you when the time comes that you need to access archived data quickly. Flow diagrams are a useful and common way of visualizing any given process. To design a process you should consider:
1. Your starting point.
2. What you will achieve in the end.
3. The steps that you need to take in order to get there.
A process will consist of a number of actions that need to be completed before you can move on to the next step in the process. There may be decisions to be made, and the decision will dictate subsequent actions.
The symbols are:
Beginning or end. This is the symbol to illustrate the start or end of your process flow.
Process/Activity/Task. An activity is something that must be completed before you can move to the next section of a flowchart; it could be something like ‘send email’, ‘delete post’ or something similar.
Decision. The decision symbol allows you to direct the flow chart in one or more directions, based on the decision. Decisions with a ‘Yes’ or ‘No’ outcome are most common; for example, a decision of ‘is the contract signed?’ would give two possible options.
Manual operation. This symbol illustrates a step in a process which must be completed manually, not through automation. The manual review of a piece of content by a human.
Document. This symbol can be used to illustrate a document, such as a policy document or contract, usually the input or output of a process.
Data. Either an input or an output, the data symbol can be used to illustrate where data is received, such as receiving an email or generating a report.
I recommend that all of your most important or high-risk processes be documented in this manner. This way, you can quickly and visually see what tasks need to be completed in order to achieve a particular goal.
It’s also a much clearer way of documenting complex processes than attempting to type them out using words, particularly where there are multiple decisions or multiple possible outcomes.
New user process
One area of risk that can impact your organization is the operating procedures around user administration. In any organization, you will have joiners and leavers and it’s important to have documentation available that shows what tasks need to be completed in any scenario.
The processes around new joiners or leavers are likely to have significant overlap with the HR and IT departments, both of which will almost certainly already have defined processes and policies.
Many organizations generally have good documentation around new employee onboarding, such as ‘request security pass’, ‘request computer user account’, ‘request access to various different systems’ and so on.
Often, social media can be overlooked so it’s worth ensuring that you have robust processes that cover what tasks need to be completed when someone new joins your team. It may be that controlling access to social media accounts is your responsibility so it’s important to document robust processes to manage risk.
To do this, you might use a ‘role description form’ which sets out who the person is, what role they will play, how long they are expected to be in the role, what privileges or access rights they will have and so on.
This is important to ensure that they are given the right access because getting their access level wrong could expose you to additional risks. This is relevant to both enterprise social networks as well as traditional social media, for example where a social media management tool is used.
An advantage of having well-documented policies, procedures and governance, in general, is that it will greatly assist new joiners to your team. What seems obvious for you or your team may not be obvious to a new joiner. They won’t know anyone in the team, nor will they know the systems that you use or how your team works.
Checklist for new users
Have relevant role description forms been completed, reviewed and approved?
Have you determined which accounts the user will have access to?
Do you know what privileges the user will have within each account?
Do you know who will provision user access and when this will happen?
Do you know how the user will be notified of their account details?
Do you know what introductory training the user will receive, and when? Do you know if their role will require any additional training?
Has the user read and agreed to all applicable social media policies and guidelines? Has the user been provided with a copy of the policies and is their acceptance recorded?
Processes and procedures
Is the user familiar with the relevant processes and procedures they will need to follow?
Have you included examples of good and bad behaviours?
Does the user have a list of key contacts so they know who to reach out to for support?
Have you considered what other resources might be helpful for the user to refer to during the onboarding process?
As with the new joiner process, a well-defined and documented leaver process is essential. In fact, it’s more important than the new joiner process as it carries more risk.
If someone leaves your organization and their access to corporate systems or accounts is not revoked, they will be able to continue to access your confidential information after they have left. In the case of social media, they may be able to continue to make posts on your official social media channels, as has happened many times in the past.
I have heard stories where someone has left a company and everyone apart from one poor individual in IT has gone to the pub for leaving drinks. The person left behind has the arduous task of changing the passwords for all social media accounts because no management system was in place.
Of course, if you do use a management system you won’t need to reset all passwords, but you will need to ensure that you disable the leaver’s account to prevent them from accessing it. If there are any other group accounts that you use, however, you’ll need to remember to change the passwords manually.
You may also need to remove personal information about a leaver from your enterprise social network. They may have included personal details about themselves, including a photograph, on their profile. We’ve already covered data privacy and protection in Blog so you’ll know that it is your responsibility to keep personal information up to date and to delete it when it is no longer required.
This applies to your enterprise social network too, so you’ll need to include any tasks that need to be completed in relation to your enterprise social network in your leaver process documentation.
The data or content that the user-produced, such as discussions on the enterprise social network, probably won’t need to be removed because it’s likely that your employment contracts and policies include clauses which note that any content produced during the period of employment is owned by the employer. But, this is an important clause to check!
A question that has come up in the past is one of ownership of social media accounts. The question as to who owns corporate accounts, such as ‘@YouCompany’ or ‘@YourCompanyCustomerServices’ is fairly straightforward, but there have been instances where an employer has asked that an employee hands over access to their own social media account to their employer, or that they provide details of all connections to the company.
The legalities of this are outside the scope of this blog and are points that should be assessed by qualified legal professionals. However, I strongly question an employer’s perceived ‘right’ to access an employee’s social media account, let alone to ‘hand it over’ to the employer.
First, it is a breach of the terms and conditions of most social networks to give anyone access to your account (for example, through password sharing), so by giving your employees access, you would be in breach of these terms.
Second, if you are worried about the effects of an employee leaving and taking their contacts with them, there are other measures you can put in place to minimize the impact of the employee leaving.
For example, you could mandate that any professional contacts, whether they originated from human interaction or social media interaction, be logged on the company’s contact management system. That way, the company would still hold the details of the key contacts, which would minimize the effect of someone leaving.
My view is that an organization has no rights to any of its employees’ personal social networking accounts or connections, and it seems unethical to think otherwise. You should seek independent legal advice if this is a particular concern.
Things change: people come and go, new products are launched; old products are discontinued; new software replaces old software. The process for making changes to IT systems, policies or processes needs to be tightly governed to ensure that any new or existing risks are assessed and managed appropriately.
It’s likely that over time your social media policies and procedures will need to be refreshed and updated. The best indication that it might be time to update your policies or procedures is if you start to receive requests for clarification or when new technologies are adopted at your company.
When this happens, you will need to ensure that new changes are proposed, validated, reviewed, approved and then take effect at an agreed date. Your own process for change management may differ and may be more or less complex, however, it’s important to have a change management process defined and documented so that you don’t make changes that introduce new or heightened risks to your operating environment.
All documentation should include a change history that details any changes that have been made to a document since its inception, who made those changes, who approved them and when they took effect. This is a useful audit trail as it will allow you to see the history of changes and, if required, to easily look back to a previous version.
You should have a change process defined for updates or amendments that you make to your social media platforms, such as your enterprise social network. If you have opted for an off-the-shelf package with absolutely no customization, you may choose never to implement custom changes.
However, if you do want to make changes, it’s likely they will require some form of development in order to add or enhance some functionality. The first stage is to document your requirements. It’s advisable to define a standard change request template that can be used for all changes as it should ensure that all of the information required by your developers is included.
When dealing with developers or development teams, it’s vital that your requirements are documented thoroughly and accurately. Developers will usually follow requirements precisely – if you include obvious typing mistakes.
For example, they will be included in the finished product. It’s therefore important to review requirements thoroughly, as making additional changes can be very costly.
Once the developers have received the requirements they will follow their own processes, such as the development of technical requirements. They will then develop the change you requested and, once complete, will perform what’s known as ‘unit testing’.
Unit testing is testing which the development or technical team complete to try to identify and fix any bugs before they release it for testing by you.
You will then perform what’s known as User Acceptance Testing, or UAT. Normally you will develop testing scripts based on your requirements which detail, step by step, how you will test the new functionality.
Your testing should also include ‘regression testing’, which is the term used to describe testing the wider functionality of your application to ensure that the new change has not had an adverse impact on another part of your application or network. Any bugs that you uncover should be documented and sent back to the developer who will address the issue and release another change for you to test.
Once testing is complete, you will sign off the change. By doing this, you are confirming that you are happy with the changes and that you are ready for the change to ‘go-live’, which means for it to be released to your production (or ‘live’) environment.
The final step will be completed by the developers, who will agree on a schedule for when the ‘go-live’ should take place. The application or network may incur some ‘downtime’, which is the time that your application or network might be unavailable while the updates are made.
The documentation around application changes is important as it shows how an application has changed over time and, should you run into issues later, it can be useful to review when trying to work out the root cause of any seemingly mysterious issues.
In this blog, we looked at the stakeholders who have an interest in social media governance and methods for engaging them through working groups or committees. We then looked at some of the mechanisms and controls you can implement to manage risk and support your projects, such as moderation.
We covered the different types of data which you might capture or analyze, and looked at some of the difficulties in analyzing unstructured data or sentiment. Monitoring is one of the hotly debated topics, which we covered in detail, before moving on to the assessment and measurement of social media. The metrics you implement will depend on your requirements and your goals.
Finally, we covered the importance of operating procedures and how they can be documented.