Complete guide Twitter analytics
Upcoming SlideShare
Loading in...5
×
 

Like this? Share it with your network

Share

Complete guide Twitter analytics

on

  • 2,072 views

 

Statistics

Views

Total Views
2,072
Views on SlideShare
2,051
Embed Views
21

Actions

Likes
2
Downloads
13
Comments
0

1 Embed 21

https://twitter.com 21

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Complete guide Twitter analytics Document Transcript

  • 1. The Complete Guide to Twitter Analytics How to analyze the metrics that matter
  • 2. TABLE OF CONTENTS INTRODUCTION • 01 Defining the Statistics DEFINITIONS • 05 Engagement Retweets Mentions and @Replies Favorites Hashtags Potential Reach Potential Impressions Potential Reach versus Potential Impressions Response Rate Response Time Interactions per Person BASIC ANALYSES • 21 How to measure your account engagement How to measure your audience How to measured Share of Voice How to measure visual content How to time your Tweets ADVANCED ANALYSES • 34 How to optimize for site traffic How to measure customer service How to measure campaigns How to measure influence TOOLS • 51 Example Reports ABOUT SIMPLY MEASURED • 52
  • 3. INTRODUCTION Twitter is a green field for content marketers and social media managers. With more than 241 million active users, 500 million Tweets, and 2.1 billion searches every day, online marketers have an active and informed audience to engage with. But many ask questions like: “What’s the best way to engage my followers? How often do I need to tweet or reply to stay relevant? What’s the best way to leverage my brand’s Twitter account?” The answer lies in a few key metrics that you should be using to gauge your performance, all accessible through the Simply Measured Twitter Account Report. In this definitive eBook, we’ll walk you through the different Twitter metrics which are measurable through the Simply Measured Engagement Megaphone, and explain several specific ways to use them to create actionable Twitter tactics. Finally, we’ll give you the tools to do the analysis and reporting yourself. 3
  • 4. Defining the Statistics If you want to be able to effectively leverage Twitter actions, you must first understand what they are, how they work, and the ways they’re calculated. In this section, we’ll outline the actions, or the different ways that users interact with your brand. We’ll also give you insight into how they affect your brand’s Twitter visibility. Twitter Engagement Megaphone What started it? 207 Tweets Sent by @delta How many unique people engaged with your Tweet? How many times did these people engage? How many people could have seen these Tweets? How many impressions could have been generated? What happened? 1,337 18,750 28,793 53,493,020 127,884,074 Unique People Total Engagement Potential Reach Potential Impressions People that interacted with you on Twitter Organic mentions, @Replies, Retweets and Favorties Combined followers of people tweeting about your brand (6/1/13 to 6/30/13 Potential times served in all follower’s needs bit.ly Clicks 1.5 Interactions Per Person 2,853 Avg. Followers Per Person Engaging 12,741 Followers Added 2.4 Impressions Per Person Reached The graphic above is our Engagement Megaphone, which shows how content is amplified in social media channels. It provides an analytic breakdown of your brand’s engagement on Twitter, giving you an idea of how your Twitter efforts resulted in clicks or new followers. In fact, the Engagement Megaphone is the entire reason you create content on Twitter. You can reach massive audiences when your followers, influencers, and advocates engage, both spreading your content and increasing brand awareness. Great content leads to engagement and amplification. In turn, you increase your reach as more fans opt-in, which adds more fuel to the machine. With our Engagement Megaphone, there are metrics for each step of the process. Brands can use these to track and analyze their performance, and optimize and prove the value of social in their marketing efforts. The Engagement Megaphone gives you insight into how many unique people engaged with your content and how many times they engaged through mentions, @replies, and retweets. From this, we’re able to calculate the total possible users your content reached and how many potential impressions it generated. This calculation may seem complicated, so to help understand this better, we’ll break down and define each component that plays an active role in your brand’s Twitter visibility. 4
  • 5. DEFINITIONS Engagement In an industry that pairs web activity and bottom-line ROI with brand awareness and overall market penetration, engagement can be one of the best ways to demonstrate success and brand activity. It tells the story of who is talking to your brand, about your brand, and why. This is why engagement is one of the most important Twitter metrics used today. Engagement: @Replies + Retweets + Mentions + Favorites What does engagement mean to you? Engagement on Twitter accounts for every way followers can interact with your brand to make it show up in their timeline. It incorporates one-on-one conversations as well as promotion to their circles of influence. This interaction is what makes Twitter such a powerful tool. Brands of all sizes have the ability to converse with users, respond to their questions, and promote their message in real-time. The followers with whom you’re engaging present several opportunities: they can act as advocates for your company, provide feedback on products or services, purchase products, and help you better understand your customers. Understanding how your brand engages users on Twitter is the first step to learning, developing and growing your Twitter marketing campaigns. 5
  • 6. How is engagement calculated? Twitter Engagement Breakdown 3.5k Mentions @Replies Retweets Favorites Total Engagement 3.0K 2.5K 2.0K 1.5K 1.0K 500 3 8/ 26 /1 3 8/ 25 /1 3 8/ 24 /1 3 8/ 23 /1 3 8/ 22 /1 3 /1 21 8/ 8/ 20 /1 3 0 Engagement is the total of several components during the given report period: @Replies: When a user talks directly to your brand on the Twitter timeline by using your brand handle at the beginning of the Tweet. This will only show up in your feed and the feeds of users who follow you both. Example: “@SimplyMeasured Your reports rock!” Retweets: When a user directly shares your brand message with their audience. Example: “RT @SimplyMeasured: We’ve made some great updates to our Twitter Account Report. Check it out!” Mentions: When a user includes your brand handle, but not as a direct @reply. Example: “I really love that @SimplyMeasured charts are dynamic within Excel!” Favorites: When a user stars a Tweet from your brand without having to retweet or reply to the Tweet, or mention your brand. Many businesses focus on retweets and mentions because they have reach, appearing in the timelines of your followers’ followers, accessing a Twitter segment that may not be following your brand. This is not to say that @replies or favorites, which don’t have reach, are not worth looking at.. With thousands of Tweets being posted every second, examining which Tweets foster more engagement will give you insight into the types of Tweets that have the most impact. 6
  • 7. Retweets Retweets are a great way for users to let their followers know that they are actively engaging with your brand by republishing your content. A retweet is a repost of a Tweet sent by another user. These Tweets are marked with the retweet icon and include the author’s information, and the name of the user who retweeted the content. They are one of the most commonly used tools on Twitter and can be very helpful in identifying web trends, content that interests your readers or their followers, or Tweets that have the capacity to go viral. Retweets: A retweet is a repost of a Tweet sent by another user, marked by the retweet icon. Example: 7
  • 8. Mentions and @Replies The difference between mentions and @replies seems simple at first, but many business Twitter accounts aren’t leveraging them properly. Mentions and @replies have very different impacts for your brand, and both are extremely important when it comes to engagement metrics. Understanding the difference between mentions and @replies will help you determine how to best use both to your advantage - to help your brand’s Twitter account stand out from the crowd. Mentions: A Tweet including your @handle at any point other than the beginning. Example: “I really love that @SimplyMeasured charts are dynamic within Excel!” Mentions are when a user includes your brand handle, but doesn’t begin the Tweet with the @handle. These Tweets show up in your stream, the user’s stream, and the stream of anyone following the user. As we mentioned before, these Tweets have the potential to reach Twitter users who may not be following you. @Replies: Tweets that BEGIN with your @handle. @Replies are when a user talks directly to your brand by using your brand handle at the beginning of the Tweet in the Twitter timeline. This will only show up in your feed, the user’s feed, and the feeds of users who follow you both. Example: “@SimplyMeasured Your reports rock!” An interesting side effect of the difference in visibility for @replies and mentions is the .@reply (note the period before the @ sign). Twitter users use this method of engagement to overcome the less public nature of @replies. If a user wants their followers to see their replies to your brand, they will use an .@reply in place of a regular @reply. Because the Tweet starts with a period, it’s not considered an @reply, and will show up in their timeline and the timelines of anyone who follows them. In addition, because the Tweet starts with something other than your brand’s Twitter handle, these types of replies count as a mention. Example: “.@SimplyMeasured Your reports rock!” 8
  • 9. What do Mentions and @Replies mean to you? @Replies and mentions account for two-thirds of your engagement and demonstrate two very different things. @Replies tell a story of users looking to engage in conversation with your brand. Mentions are more of an endorsement of your brand. The key difference between mentions and @replies is where they appear – who automatically sees Tweets that use your brand handle. With mentions, your brand handle has the potential to be seen by a much larger range of users than it would in an @reply, allowing you to reach potential new followers. @Replies have a much smaller audience, but are important in building a strategy to retain followers through mutual engagement. An .@reply will be seen by your followers and your followers’ followers, and can help both gain and retain Twitter followers. Focusing on the types of mentions and @replies, the number of each, and patterns for each type can help you understand your audience and the relationship they have with your brand. This is important in the type of messaging you put out, the way you interact with various users, and the way you measure success. Twitter Engagement Breakdown 3.5k Mentions @Replies Retweets Favorites Total Engagement 3.0K 2.5K 2.0K 1.5K 1.0K 500 9 3 8/ 26 /1 3 8/ 25 /1 3 8/ 24 /1 3 8/ 23 /1 3 8/ 22 /1 3 /1 21 8/ 8/ 20 /1 3 0
  • 10. Favorites Favoriting is becoming an increasingly popular way to engage on Twitter. In fact, favorites have grown to represent a significant portion of the engagement mix on Twitter. Favorites are similar to Likes on Facebook. With a single click, you can engage with content to either bookmark, show your appreciation, or simply let the author know you’ve seen their Tweet. This has made favoriting an attractive form of engagement. Favorites are trending: We measured engagement for some of the top automotive brands on Twitter and found that favorites accounted for 8% of their total engagement. Twitter Engagement Breakdown 3.5k Total Engagement 3.0K Mentions @Replies Retweets Favorites 2.5K 2.0K 1.5K 1.0K 500 3 8/ 26 /1 3 8/ 25 /1 3 8/ 24 /1 3 8/ 23 /1 3 8/ 22 /1 3 /1 21 8/ 8/ 20 /1 3 0 Many automotive brands received nearly as many favorites as replies. In fact, in just one week @VW actually received 367 favorites, compared to only 276 replies. 4.5K Twitter Engagement Details Comparison 4.0K Mentions @Replies Retweets Favorites Total Engagement 3.5k 3.0K 2.5K 2.0K 1.5K 1.0K 500 0 @Honda @Ford @Toyota @NissanUSA @VW It’s clear that favorites represent a significant volume of total engagement, and tracking them ensures that brands are getting the complete picture of how users are choosing to engage with their brand. 10
  • 11. How Favorites are different: Favorites are a preferred form of engagement because they allow users to engage without having to retweet your content, provide feedback with a reply, or mention your brand. Users often want to engage without having to voice their opinion or broadcast something to their audience. Favorites benefit the user and send a signal to the author; as such, they’re the only form of Twitter engagement that doesn’t reach other users’ feeds.. The benefit for your brand: If favorites aren’t often seen by others, why should you care? Although favorites don’t create awareness the same way that retweets and mentions do, they are still another important indicator for measuring how your content resonates with people. If certain types of content are receiving more favorites than comments or retweets, you may need to deliver your content in a way that encourages more users to retweet and share. Percentage of Engagement by Content Type 60% @Replies Retweets Favorites 50% 40% 30% 20% 10% 0% Photos Videos Links Normal Tweets Measuring favorites will also give you a better understanding of your ability to engage your audience. It’s inevitable that a portion of your audience will be more interested in consuming content rather than engaging with it. Favorites show a more complete picture of how users of how users are consuming your content which can inform your decisions on how to best optimize your content on Twitter. 11
  • 12. Hashtags Another important method of engagement on Twitter is the use of hashtags – a form of earned engagement as opposed to the owned engagement we outlined above. They are one of the greatest tools social marketers have at their disposal. Yes, some marketers abuse the feature. Yes, there’s a relatively steep learning curve and some confusion when it comes to using hashtags as a tactic. But when it comes to campaign management, organization, and branding, it’s hard to top the simple power of the hashtag. Hashtags are terms used in Tweets (and now on Facebook and Instagram as well) that are searchable, clickable, and measurable. Hashtags: Clickable terms within Tweets that begin with the “#” sign What do Hashtags mean to you? Hashtags should be a part of any Twitter strategy, because they allow marketers to engage with users they otherwise wouldn’t be able to and to build branded campaigns of their own. You can use hashtags to build easily monitored campaigns in a variety of ways: Campaigns: When conducting a specific campaign, hashtags can be used to distinguish engaged Tweets and users. They can be searched to see activity and interest, and branded to promote your cause or company. For example, if Simply Measured wanted to promote a giveaway of a t-shirt and some stickers, we could use the hashtag #SimplySwag and ask users to tweet using the hashtag for a chance to win. Increased Reach: Hashtags of specific topics are often searched and monitored by marketers and users with specific interests. For example, if we tweet a link to our Twitter eBook with the hashtags #SocialMedia and #Twitter, we increase the chance of reaching social media marketers who are interested in learning more. Chats: There are countless regular “Twitter Chats” out there that use specific hashtags to allow users to organize the conversation. For example, if we wanted to host a regular conversation about social media measurement, we might choose the hashtag #MeasureChat. This allows us to promote a searchable term that allows users to view and interact with anyone getting involved in the chat or conversation. Discovery: When doing research, hashtags can be searched to discover interests, sentiment, attitudes, and demographics of the users engaging with the hashtag. Comparison: Different hashtags can be measured and compared to identify trends, growth, or disparity. This is important for recurring campaigns and competitive analysis. 12
  • 13. Potential Reach Potential reach is an important metric for any social media marketer. On Twitter, we aren’t just focused on engaging the people already following us, we’re trying to grow and expand our audience. Potential reach is one of the best ways to tell if we’re doing that successfully. Potential Reach: The sum of all users mentioning your brand + the sum of their followers. User A, who has 5 followers, posts a Tweet that mentions the @SimplyMeasured Twitter handle. In this example, we’ll calculate the potential reach of this Tweet for Simply Measured. User A • 3h I love @SimplyMeasured reports! + users mentioning your brand (@UserA) = 1 + sum of their followers (@UserA) = 5 Potential Reach = 6 What’s the potential reach of @UserA’s Tweet if it’s retweeted by @UserB, who has 9 followers? User A • 3h I love @SimplyMeasured reports! User B • 1h RT @UserA I love @SimplyMeasured reports! users mentioning your brand =2 (@UserA + @UserB) + sum of their followers (@UserA) = 5 + sum of their followers (@UserB) = 9 Potential Reach = 16 NOTE: Potential Reach will always include your brand followers since they are part of the audience you are reaching on Twitter. When calculating the reach of just a mention (like in the examples on this page), your followers are not being engaged so they are not part of the reach calculation for that specific tweet. But when we calculate the overall reach of your Twitter activity, your followers are added to the equation. If @SimplyMeasured has 20 followers, in example one the total potential reach will be 26 and in example 2 it will be 36. 13
  • 14. What does Potential Reach mean to you? The potential reach metric allows you to quantify not only the users you engaged with, but also the followers of those users who may have seen your @handle or Tweet. This is important because the focus of social marketing is to expand your audience and promote your message to a wider segment of the population. The reach metric is a good indicator of the content that’s working to grow your audience and ultimately “reach” new people. This is a true look at the audience you have the potential to engage with. Potential Impressions Potential impressions have always been an important metric for advertisers. For traditional media like newspaper, radio, and TV, it has been one of the only metrics available to gauge success. Its relevance has stayed just as prominent throughout the advent of social media. Potential Impression: The total number of times a Tweet from your account or mentioning your account could appear in users’ Twitter feeds during the report period. It includes your Tweets, Tweets that mention your brand handle, and retweets of your content. What does the Potential Impressions metric mean to you? Potential impressions are an important part of measuring your brand impact. If content you’re creating has a viral impact – for example Tweets that earn a large number of retweets and @ mentions - your potential impressions are the quickest way to identify that trend, allowing you to focus your efforts on the content that is drawing the most attention. Potential Impressions Analysis Brand Tweets 800 User Tweets Total Potential Impressions 18M 14M Total Tweets 600 12M 500 10M 400 8M 300 6M 200 4M 100 2M 14 12 5/ 6/ 12 3/ 6/ 12 1/ 2 /1 30 5/ 6/ 2 /1 2 /1 28 5/ 2 26 5/ /1 2 24 5/ 22 /1 2 5/ /1 20 5/ 8/ 12 2 51 16 /1 2 5/ 14 /1 2 5/ 12 /1 2 5/ /1 10 5/ 8/ 12 0 5/ 5/ 6/ 12 0 Potential Impressions 16M 700
  • 15. What makes Reach different from Potential Impressions? Reach accounts for the possible number of people who may have seen your content, whereas impressions calculates how many times the people you’ve reached have seen your content. If you think of reach as how many screens your Twitter handle appears on, and impressions as how often your Twitter handle appears, it’s easy to get an idea of how far you can potentially spread your message and grow your brand. How are your Potential Reach and Potential Impressions calculated? Both potential reach and potential impressions are calculated based on mentions and @replies. When it comes to mentions, there’s an overall consensus on how to calculate that portion of potential reach and potential impressions based on the number of author’s followers: Potential reach and potential impressions for @simplymeasured Type Author Followers Potential Reach Potential Impressions Mention 10,700 10,700 10,700 Mention 10,700 10,700 10,700 TOTAL 10,700 21,400 User A • 3h I love @SimplyMeasured reports! User A • 1h Thanks @SimplyMeasured you guys are the best! The nature of @replies is that they show up only in the feeds of users who follow both handles (the original author and the @reply author). Unfortunately, there’s no consensus on how to calculate the potential reach and potential impressions for @replies; whether to assume no overlap (the authors share no followers) or assume some percentage of overlap (the authors share some of their audience). Based on feedback collected from our customers, our analysts were able to develop three different formulas for calculating this metric to accurately cover the widest spectrum of needs. At Simply Measured, we’ve focused on the most “conservative” calculation for @replies as the go-to model, because the majority of our research supports this method. 15
  • 16. In this calculation one @reply results in one person reached. This will likely give you a lowerthan-actual reading, but makes a fair assumption that @replies won’t generate the full reach of the author’s followers. @Replies are typically 1-to-1 interactions, where reach and impressions aren’t as relevant. Potential reach and potential impressions for @simplymeasured Type Author Followers Potential Reach Potential Impressions @Reply 10,700 1 1 @Reply 10,700 1 1 TOTAL 1 2 User A • 3h @SimplyMeasured your reports are the best! User A • 1h @SimplyMeasured you guys rock! If you want to assume some amount of overlap in followers, Simply Measured’s Twitter Analytics supports modified reach calculations at the push of a button. We can help you determine which solution is best for your brand. Knowing how to best calculate your potential reach and potential impressions will give you insight into your virality statistics and brand exposure. Imagine @UserA mentions the @SimplyMeasured handle in 2 separate Tweets that reach his 5 followers... User A • 1h User A • 3h Thanks @SimplyMeasured you guys are the best! ...The @SimplyMeasured handle appears twice on each of their news feeds. Wow! @SimplyMeasured your reports are amazing! User A • 1h Thanks @SimplyMeasured you guys are the best! User A • 1h User A • 1h User A • 1h News Source • 2h News Source • 2h News Source • 2h User A • 3h User A • 3h Weather tonight is cloudy with a chance of rain User A • 3h News Source • 5h News Source • 5h News Source • 2h News Source • 5h User A • 1h News Source • 2h User A • 3h User A • 3h News Source • 5h Wow! @SimplyMeasured your reports are amazing! User A • 1h News Source • 2h User A • 3h News Source • 5h News Source • 5h Tune in tonight at 8pm for the latest #worldnews Number of Potential Reach x Number of Appearances = Potential Impressions 5 followers x 2 appearances = 10 Potential Impressions 16
  • 17. Response Rate Response rate is becoming more and more important to social media teams. On Twitter, focusing on outbound promotion just doesn’t work anymore. Customers reach out with issues, questions, and requests on a daily basis. It’s our job as social marketers to act as first responders in these situations – answering questions, directing traffic, and sometimes just saying no. Whatever your policies and tactics, measuring customer service success boils down to a few specific metrics. Key among them is response rate. Response Rate: Total # of Mentions Replied to Total Mentions - Retweets What does Response Rate mean to you? Whether you’re a customer service rep, a community manager, or a marketing manager who handles everything for a smaller company, Response rate can help you identify successes and holes in your social strategy. Maintaining company image can be as simple as keeping your engaged customers happy. This means responding to the highest percentage of inbound @mentions and @replies. How is your Response Rate calculated? @fordservice Activity and Engagement Tweet Sent Retweets Mentions Excl. RTs 400 150 300 100 200 50 100 1/ 1/ 1/ 13 2/ 1/ 13 3/ 1 1/ 3 4/ 1/ 13 5/ 1/ 13 6/ 1 1/ 3 7/ 1/ 13 8/ 1/ 13 9 1/ /13 10 1/ /13 11 1/ /13 12 1/ /13 13 1/ /13 14 1/ /13 15 1/ /13 16 / 1/ 13 17 /1 1/ 18 3 1/ /13 19 1/ /13 20 1/ /13 21 1/ /13 22 1/ /13 23 1/ /13 24 1/ /13 25 1/ /13 26 / 1/ 13 27 /1 1/ 28 3 1/ /13 29 1/ /13 30 1/ /13 31 /1 3 0 17 0 Tweets Sent Retweets & Mentions 200
  • 18. Response rate is the number of inbound mentions of your brand handle that you’ve responded to, divided by the sum total of all brand mentions (excluding retweets). This equation gives you the percentage of inbound Tweets that you’ve responded to. With Simply Measured, this can be calculated using both our Customer Service Report and our Multiple Handle Customer Service report – designed specifically for brands with a dedicated customer service handle. Response Time Customers reach out to brands constantly. Whether it’s a question, concern, or gratitude that they are trying to convey. that they’re trying to convey, it’s our job to engage with them in a way that shows our company cares. An industry standard for customer service success. Response time, similar to response rate, is an important way to measure your social activity for both brand engagement and the customer service. Response Time: Time of @Reply to inbound Tweet Time of inbound Tweet What does Response Time mean to you? For many brands we’ve talked to response time is calculated in different ways. Direct messages and replies to Tweets that didn’t mention the brand handle (through a hashtag campaign for example) are just some of the factors that come into play. This can be a frustrating number for many brands that often don’t staff a Twitter account outside of normal business hours. A few dozen Tweets on a Friday that don’t see a response until Monday drastically skew your average response time. A recommendation that we make to Simply Measured users is to open the Excel version of the Simply Measured Customer Service Report and look at individual Tweets. This allows social media managers to identify and remove outliers, like the Friday evening mentions, to get a more realistic picture of their efforts. 18
  • 19. How is your Response Time calculated? Response Time: In Minutes Avg Time Per Response Minutes Per Response (Goal) 90 400 80 70 300 60 50 200 40 30 100 20 10 0 2 /3 / 11 12 /4 / 11 12 /5 / 11 12 /6 / 11 12 /7 /1 11 2 /8 / 11 12 /9 11 /1 /1 2 0 11 /1 /1 2 1 11 /1 /1 2 2 11 /1 /1 2 3 11 /1 /1 2 4 11 /1 /1 2 5 11 /1 /1 2 6 11 /1 /1 2 11 7/1 /1 2 8 11 /1 /1 2 9 11 /1 /2 2 0 11 /12 21 11 /1 /2 2 2 11 /1 /2 2 3 11 /1 /2 2 4 11 /1 /2 2 5 11 /1 /2 2 6 11 /1 /2 2 11 7/1 /2 2 8 11 /1 /2 2 9 11 /1 /3 2 0/ 12 /1 11 /2 11 11 /1 /1 2 0 The simplest way to measure response time is by subtracting the time of the original inbound Tweet from the time of your response. Take this difference in time from each of your responses and average them for your average response time. If the numbers don’t add up or don’t represent your effort the way you expected, open the Excel file and identify the responses that are skewing your average. This can help you identify holes in your strategy, ways to optimize your responses, or reinforce what you’re already doing. Interactions Per Person There are many ways to measure the impact you have on your Twitter audience - whether through engagement, clicks, or favorites. One metric you can use to sum up your audience health is interactions per person. What are Interactions Per Person? Interactions Per Person: Total Interactions Total Audience What do Interactions Per Person mean to you? This metric is great for understanding how involved your audience is as a whole. It’s one thing to see a total number of retweets or @replies that you’ve earned through your social campaigns, but understanding how that number stacks up to your audience size can give you a much clearer idea of your social media success.. 19
  • 20. This metric is important in identifying the appropriate level of saturation and gauging whether the amount of content you distribute to your audience is too little or too much. Study this metric as you test your cadence to find the highest level of interaction per person, before it plateaus (a sign of over-messaging). The goal is to find the sweet spot where multiple impressions are making an impact and engaging your audience, without spamming them. How to measure Interactions Per Person: Twitter Engagement Megaphone What started it? 207 Tweets Sent by @delta How many unique people engaged with your Tweet? How many times did these people engage? How many people could have seen these Tweets? How many impressions could have been generated? What happened? 1,337 18,750 28,793 53,285,030 127,884,074 Unique People Total Engagement Potential Reach Potential Impressions People that interacted with you on Twitter People that interacted with you on Twitter Combined followers of people tweeting about your brand (6/1/13 to 6/30/13 Potential times served in all follower’s needs bit.ly Clicks 1.5 Interactions Per Person 2,853 Avg. Followers Per Person Engaging 12,741 Followers Added 2.4 Impressions Per Person Reached Interactions per person should be measured in context with your other KPIs. In the example above, @delta has earned 28,793 mentions, @replies, retweets, and favorites. Since their audience is 18,750 people, this equates to 1.5 interactions per person. It’s important to note that this metric isn’t saying each of those 18,750 people engaged 1.5 times, but rather giving a snapshot of how that total engagement relates to @delta’s audience size. 20
  • 21. BASIC ANALYSES Now that we’ve gone over all of the metrics definitions, we’re going to show you how to put those metrics to work to analyze your Twitter campaigns and make better decisions. We’ll be working out of our own software, but you can conduct the same analysis no matter which tool you use to pull Twitter data. Measuring Engagement One of the most important goals for social media marketers on Twitter is increasing engagement. Through simple interactions, engagement helps grow awareness of your brand and helps increase traffic to your social profiles and your website. Remember, we defined engagement as the sum of @replies, retweets, and mentions. These three Twitter actions directly amplify your brand’s share of voice as your brand’s handle reaches the timelines of users who may not be following your brand on Twitter for regular updates. To measure your engagement look at your audience and how you retain followers. Is your audience growing? Do these users follow you during promotions, but unfollow shortly thereafter? What is the industry standard for number of followers and how do you stack up? Now, look at how those followers and their circles respond to or share your content. Do you have a surprisingly low number of retweets for the number of followers you have? Are there certain times that are better for engagement? Have you recently published any Tweets that have had success in regards to mentions or retweets? We’ve provided a few different metrics you can use to benchmark your campaign performance and assess your engagement strategy. 9 ways to measure Twitter audience beyond follower count Follower count is an important metric for measuring the performance of a growing audience on Twitter, but too often it’s the only audience metric that marketers focus on. With each new follower gained, the characteristics of your audience change. Over time, substantial follower growth can result in very different type of audience and just measuring change in follower count can only tell you so much. There is more you can do to understand your Twitter audience and find more effective ways to grow your audience. 21
  • 22. 1. Measure your followers’ followers Your follower count is just the first step to understanding your potential to reach. The followers of your followers are your secondary network. They determine how much potential there is for sharing content downstream. Users Followers’ Follower Count Distribution 450 400 350 300 250 200 150 100 50 0 413 200 87 0-100 101-500 61 501-1,000 26 2,001-2,500 >2,500 Follower Count 2. Measure your competitors’ followers Comparing competitor audience growth rates to your own growth will give perspective on how effectively you’re growing your audience - letting you know whether you’re ahead of the game or playing catch up. Twitter Comparison: Total Followers Total Followers 250K 0.5% 200K 0.6% 207.6K 0.5% 150K 100K Growth Rate per Day 117.1K 72.8K 50K 0.3% 0.1% 0.4% 0.3% 51.8K 77.5K 0.3% 0.1% 0.1% 0 0.6% @sap 22 0.2% @ciscosystems @oracle @ibm @accenture
  • 23. 3. Track engaged users who are not yet followers Identify users mentioning your brand by name or retweeting your content who are not already following you. These users may be unaware of your handle or are following your followers. Create a connection with these users to encourage them to follow you directly. 4. Measure share of voice for online properties Follow influencers who engage with your brand or your competitors. Monitor their Tweets for opportunities to engage with them directly. 5. Segment your followers by location Are your social goals tied to specific markets? Segmenting follower growth by location enables you to measure your share of voice on Twitter with respect to regional markets. Follower Distribution By US Metro Area % of Followers % of Population Los Angeles, CA Chicago, IL New York, NY Miami, FL Atlanta, GA Houston, TX Boston, MA Washington, DC Las Vegas, NV Detroit, MI Dallas, TX Portland, OR San Antonio, TX Tucson, AZ Fresno, CA 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 6. Segment your followers by how they engage Segment your followers by how they engage with your brand. Create “archetypes” for followers who frequently retweet your content, and those who often mention you on Twitter. Social archetypes for these users will help you better understand who your followers are and why they engage with you on Twitter. 7. Compare your new followers to your old followers Significant changes in follower count mean that you have more to learn about who your followers are. Comparing new followers to your previous followers can help you determine the value of new followers gained and whether you’re retaining old followers as you grow your audience. 23
  • 24. 8. Measure engagement relative to your followers Is your engagement scaling as your followers grow? Engagement as a percentage of followers (engagement rate) shows how much of your audience you are able to engage. Change in engagement rate can indicate the value of fans gained or lost. Twitter Comparison: Total Engagement Total Engagement 3.0K 2,562 Engagement % of Followers 4% 3.5% 4% 2.5K 2.8% 2.0K 1,601 1,575 3% 3% 1,433 1.5K 1,178 1.3% 1.0K 1.5% 2% 1% 0.8% 500 2% 1% 0 0% @sap @ciscosystems @oracle @ibm @accenture 9. Know how active your followers are How frequently your followers tweet indicates how active they are as users on Twitter. The more active your followers are, the more likely they are to see and engage with your Tweets. Followers by Date of Last Tweet 33% 35% 30% 25% 20% 18% 20% 15% 11% 10% 10% 5% 5% 2% 1% 24 r > 1 Ye a Ye ar 1 La st s th La st 6 M on La st 3 M on th s th M st La W ee 2 st La on ks k W ee st La La st 24 H ou rs 0%
  • 25. 6 ways to measure Twitter share of voice beyond brand mentions Share of voice is one of the most regularly used metrics for measuring social media performance, but are you really getting the full picture? There’s more to it than just brand mentions. There are ways to take share of voice a step beyond mentions to measure not just how much share of voice your brand has, but how different types of conversation impact your brand relative to your competitors. With data from our own social media reports, we’ll look at different ways (besides brand mentions) to measure how share of voice impacts brand perceptions, regional markets, site traffic, and more. 1. Measure keyword share of voice Measuring your brand’s share of voice against specific keywords provides context for how users discuss your brand compared to your competitors. Depending on the keywords being compared, you can gain insights related to specific product categories or brand attributes. This data is particularly useful for measuring the performance of initiatives aimed at building product awareness or shifting consumer perceptions of your brand. Keyword Share of Voice Within Brand Mentions @nissanusa @kia @honda @toyota ic Ec on om e bl lia Re st Co d ee Sp y fe t Sa rid yb H yle St SU V er Po w Se da n 100 90 80 70 60 50 40 30 20 10 0 2. Segment share of voice by location Share of voice by location can indicate which regional markets are most aware of and engaged with your social brand presence. Regional share of voice data can be helpful for distributing relevant content, planning location-based social media efforts, and measuring your brand’s ability to increase awareness in specific markets. 25
  • 26. 3. Compare potential impressions and reach Share of voice typically measures brand mentions as a percentage of total Tweets in order to gauge the level of consumer awareness. However, share of voice as a percentage of mentions doesn’t account for how many unique users are discussing your brand or how many followers they have. That’s why comparing potential impressions and reach are also important. Reach measures how many users were potentially exposed to Tweets mentioning your brand. Impressions can tell you how many times those users were exposed to mentions of your brand. Twitter Comparison: Impressions and Reach Potential Reach Impressions and Reach 120M Potential Impressions 103.3M 100M 80M 31.7M 17.7M 12.8M 20M 22.1M 4.8M 26.5M 3.9M V SU Se Po w da er n 0 yle 40M St 60M 4. Measure share of voice for online properties Segmenting Tweets that link to your website and the sites of your competitors can enable you to measure the share of voice responsible for driving traffic to your website. The share of voice for your online properties can serve as a performance indicator for how shareable your site content is and how well social media accounts and marketing efforts have been integrated with your site. Distribution of Links to Brand Sites All Links @Nissan @kia @honda @toyota 26 1 5/ 4/ 3 20 1 5/ 5/ 3 2 5/ 013 6/ 20 5/ 13 7/ 2 5/ 013 8/ 2 5/ 013 9/ 5/ 201 10 3 / 5/ 201 11 3 / 5/ 201 12 3 / 5/ 201 13 3 / 5/ 201 14 3 / 5/ 201 15 3 / 5/ 201 16 3 /2 5/ 01 17 3 / 5/ 201 18 3 / 5/ 201 19 3 / 5/ 201 20 3 / 5/ 201 21 3 / 5/ 201 22 3 / 5/ 201 23 3 / 5/ 201 24 3 / 5/ 201 25 3 / 5/ 201 26 3 /2 5/ 01 27 3 / 5/ 201 28 3 / 5/ 201 29 3 /2 01 3 13 20 3/ 20 2/ 5/ 5/ 5/ 1/ 20 13 70 60 50 40 30 20 10 0
  • 27. 5. Track hashtag adoption Many brands have hashtags they routinely use to provide context for specific conversation themes. Comparing share of voice for your brand’s primary hashtags with those of your competitors can reveal how successfully your hashtags are being adopted and reflect your brand’s ability to shape the way others discuss your brand. CES Year/Year Twitter Comparison: Days 0-1 Tweets from 2013 Tweets from 2012 Tweets from 2011 9.0K 8.0K 7.0K 6.0K 5.0K 4.0K 3.0K 2.0K 1.0K ( ) ( PM PM 0 :0 10 PM 00 8: PM 00 6: 00 4: 00 PM PM DAY 1 2: AM 0 0 :0 12 AM :0 10 AM 00 8: AM 00 6: AM 00 4: 00 2: 0 AM PM 0 12 :0 PM :0 10 PM 00 8: PM 00 6: 4: 00 PM PM 00 0 DAY 0 2: AM 12 :0 AM 0 :0 10 AM 00 8: AM 00 6: AM 00 4: 00 2: 12 :0 0 AM 0 ) All Times PST The above chart compares three consecutive years of Tweets using the Consumer Electronics Show’s branded hashtags. When trended over the same amount of time, we’re able to identify growth trends and determine the events that caused specific spikes and drops in activity. Hashtags are easy to track using the Simply Measured Twitter Activity Report or Stream Snapshot report and can be trended over time to measure volume and activity. This can be important for researching opportunity, identifying successes, and measuring causes of peaks and lulls in activity. This technique can also be effective for measuring the share of voice driven by hashtags used in TV advertising. 27
  • 28. 6. Identifying conversation overlap with competitors The percentage of overlap in competitor conversations measures how frequently your brand is mentioned in Tweets that also mention competitors. Conversation overlap can measure how uniquely your brand is being discussed. A small degree of overlap can indicate brand loyalty or unique product discussion, while a high degree of overlap signals frequent brand comparisons, which are typically seen between highly competitive brands. Overlap in Competitor Mentions vs. Share of Voice @Nissan Competitor Overlap 6% 38% 0% 10% 20% @honda @toyota 14% 34% Share of Voice 28 @kia 21% 30% 40% 41% 16% 50% 60% 29% 70% 80% 90% 100%
  • 29. How to measure visual content “Pics or it didn’t happen.” That’s the general rule when dealing with a friend who went fishing…or to Vegas. It might as well be the motto for social media too. Visual content has become more and more impactful as a way to reach potential customers, fans, and followers. But with so many options, services and types of visual content, how can you tell which works the best? Here are some quick ways to analyze your visual content and create a strategy based on past success. All of this analysis can be done using Simply Measured reports. Measure engagement by content type The starting point for your Twitter content analysis should be the type of content: photos, videos, or links. Measure the engagement as well as the number of Tweets you’ve sent for each content type. This will give you a solid understanding of what you should be focusing on. Brand Tweets by Content Tweets Sent 72 Photos Videos 25 Links 27 Normal Tweets 29 Engagement per Tweet 188.3 47.3 72.3 14.8 573
  • 30. Measure engagement by media type Next, identify which service helps your brand see the most success. If you’re posting videos you may ask if Vine clips perform better than YouTube. For photos, are Instagram pics as successful as Twitpics? Media Types in Tweets Engagement per Tweet Potential Impressions Per Tweet 2.3M 2.1M 20.9 2.1M 1.8M 27.8 14.8 4.5 Vine Instagram/Path 7 Posts YouTube/Vimeo 4 Posts 81.4 2.1M Other Photos 8 12 Posts Posts Other Links 30 Posts Measure competitor success by content type Finally, measure your engagement against your competition. Where are other businesses in your industry finding success? If they’re seeing much higher engagement with photos, you can perform a content analysis to find out what they’re doing with their photos that you might be missing. Type Comparison: Engagement on Brand Tweets (size of bubble = number of posts) Links Normal Tweets MOST ENGAGING CONTENT TYPE Photos 6 interactions [4.5% of interactions on all sent tweets] MOST COMMONLY POSTED s el ls rh co @ ac ho ice ot te r... ie ch o sid nc @ tc o @ hy at in sp g @ eb es tw e st er ... ... ld @ th nw or ay in n lto @ hi lid zz wo st ar @ @ m ar ri od ot bu tin in nl no lto hi @ ho tl Normal tweets 780 tweets [74% of all sent tweets] @ 30 Videos CONTENT TYPE PERFORMANCE e Engagement per post Photos 0.8 0.7 0.6 0.4 0.3 0.2 0.1 0.0 -0.2 -0.2 BEST PERFORMANCE: @marriottintl Photos 2 interactions [4.1% of interactions on all sent tweets]
  • 31. 6 types of analysis to time your Tweets Do you know how timing impacts the performance of your Tweets? There is no universal answer. The best times to tweet are unique to your brand audience and the type of content that you deliver. Because the lifespan of any given Tweet you send is limited, it’s important to determine when your audience is most active to find the best times to share your content. There are steps you can take to understand when to tweet and when to engage your followers through different types of content. 1. Analyze your top Tweets Analyzing your top Tweets can be a good starting point for identifying the best times to tweet. Identify which times are most commonly associated with top performing content. Viewing your top Tweets by time of day displays when your brand delivers its most successful content. Top Performing Tweets All Sent Tweets Top Performing Tweets Volume of Sent Tweets 16 14 12 10 8 6 4 2 PM PM 11 PM 9 10 PM PM PM PM 8 7 6 5 PM PM PM PM 4 3 2 1 AM AM AM AM PM 12 11 10 9 8 AM AM AM AM 7 6 5 4 AM AM 3 2 1 AM 0 All Times EDT 2. Measure organic mentions Measuring mentions of your brand, your handle, or hashtags, that are not retweets or replies of your Tweets can help you identify peak times for organic engagement with your brand. This approach gives you insight into user activity that isn’t biased by when your brand tweets. Measuring organic mentions over an extended time period, and taking care to avoid events that bias posting times, can reveal when users choose to engage with your brand. 31
  • 32. 3. Know your audience Knowing where your audience is located can help you identify when they’re most likely to be active online. Viewing a distribution of your audience by time zone can give you an idea of when they’ll be starting their day, taking lunch breaks, etc. The more broadly your audience is distributed, the larger window you have to post and the greater the need to tailor content that appeals to regional audiences. Followers Top Time Zones 40% 35% 30% 25% 20% 36% 13% 15% 10% 5% 0% 12% 8% (GMT-05:00) Eastern Time (US & Canada) (GMT-06:00) Central Time (US & Canada) (GMT-08:00) Pacific Time (US & Canada) (GMT-05:00) Quito 5% (GMT-04:00) Atlantic Time (Canada) 3% 3% (GMT-10:00) Hawaii (GMT-07:00) Mountain Time (US & Canada) 2% 2% 1% (GMT) London (GMT-07:00) Arizona (GMT+01:00) Amsterdam 4. Segment content types by time of day Segmenting engagement for key content types can help you determine when certain types of content are more likely to drive engagement – during different times of the day and days of the week. For example, you might find that posts with calls to action receive better response during peak times, but that content designed to entertain followers successfully drives engagement during weekends or off hours. 32
  • 33. 5. Monitor current engagement trends Distributing engagement by day and time shows your current engagement trends. This is an excellent way to determine which days and times are most effective for your brand to tweet. Viewing this data over weeks or months, excluding engagement outliers and paid Tweets, and making a point to experiment with when you post can help you avoid identifying peak times for engagement that are biased. Mentions and Brand Tweets by Day and Time Monday Tuesday Wednesday Thursday Friday Saturday Sunday Brand Tweets Mentions 1.6K 1.4K 1.2K 1.0K 800 600 400 200 0 60 50 40 30 20 10 0 12 AM 3 AM 6 AM 9 AM 12 PM 3 PM 6 PM 9 PM Time of Day 6. Compare competitor engagement Comparing a distribution of your competitors’ engagement by day and time will also help ensure that you don’t overlook opportune times to tweet. Examining the content strategies of your competitors can clue you in to successful post times you may not have experimented with, or help identify content types that drive engagement during times where you’ve had limited success. 33
  • 34. ADVANCED ANALYSES So you’ve got the basics down, but now you need to assess the ROI on your time spent engaging Twitter followers and building brand awareness. It’s time to look at how your social strategy impacts your overall bottom line by turning Twitter followers into paying customers. At this stage, you’ll need to consider your social strategy’s effects on your web funnel – how those Tweets drill down to measurable web actions like clicks or purchases. In addition, if your brand uses a customer service account, you’ll want to find out how quickly issues are resolved and where you can improve your social customer service strategy. In this section, we’ll show you how to use Simply Measured reports to tailor your analyses to your brand’s needs. 3 ways to optimize Tweets for website traffic You’ve undoubtedly spent hours crafting Tweets for your brand, but once they hit the feed you may wonder how to measure performance for posts designed to get users to your website? We all know that linking from Twitter to our website provides valuable cross-channel engagement, drives campaign objectives, and gets users to the point of purchase. Despite this, it’s not always easy or obvious how to optimize Tweets for website traffic. Let’s take a look at 3 measurement tactics you can use to start converting your Twitter engagement to website traffic. 34
  • 35. 1. Combine Twitter activity and your site’s conversion funnel in Google Analytics Tweets & Visits to Site 78 Link Mentions 2,581,260 Twitter Potential Impressions 1.19 Impressions per Person 2,167,743 Twitter Potential Reach 307 Visits to Site 0.01% CTR 300 Tweets with Link Visits 250 100 200 80 150 60 100 40 50 20 0 0 Total Visits from Twitter 0.02 Conversions per Visit 7 Goal Completions 120 Visits From Twitter 1,064 @ Mentions 3/ 5/ 13 98 Retweets Tweets with Links to Site 6 Sent Tweets 2/ 27 /1 3 2/ 28 /1 3 3/ 1/ 13 3/ 2/ 13 3/ 3/ 13 3/ 4/ 13 Twitter Activity Funnel Tweet Driving the Most Visits Visits from Other Sources Are you a data geek with a passion for writing and social media? You don’t want to miss this opportunity: http://bit.ly/xshu6G/simplymeasured... 307 27,129 36 Visits +1050% than Avg How do Tweets that resonate on Twitter funnel down to site visits and ultimately to conversions? Pairing Twitter referral traffic with Tweets linking to your website provides a much more powerful lens than looking at trended Twitter visits on their own. Combining Google Analytics with Twitter data allows you to see not just how much referral traffic you’ve received, but how many Tweets it took to drive those visits. Measure how sent Tweets are amplified throughout your network, expanding potential impressions and reach before funneling down to visits on your site. Since getting users to your site is just the beginning, configure goal completions to measure actions visitors took after reaching your site. What you have now is the ability to attribute site visits, campaign submissions, and purchases back to specific Tweets. You can see what content is responsible for helping your brand complete specific goals using Twitter. 35
  • 36. 2. Analyze content and calls to action based on website metrics The goal is to optimize content that draws users to your website. You optimize your content calendar to meet engagement goals, so shouldn’t you also optimize for site visits? Tweets that Drove Traffic Size of bubble = Avg Time on Site (in seconds) 5 Are you a data geek with a passion for writing... 4 Yup. RT @KevinSaysThings: Attention tech-sa... What’s your favorite Excel function? If you have... 2 59% of top brands are using instagram. Is yours... 1 Retweets 3 You guys, we’re hiring a content manager! 0 -1 Pretty much. RT @KevinSaysThings: Attention... 0 5 10 15 20 25 30 35 40 45 Visits Dive into the top Tweets to see what’s working. These are the Tweets linking to your site and the top content responsible for visits from Twitter. Analyze which audiences were targeted and what was included in Tweets that motivated users to click through to your site. Determine which calls to action work best. As you seek to take a holistic approach to your digital brand presence, start incorporating existing practices for mobile, search, and display into social. When driving users to your site, choose your keywords and calls to action carefully. Expanding these practices to incentivize site traffic will improve your outreach tactics moving forward. 36
  • 37. 3. Discover website influencers Who is driving traffic to your website? Identify which users link to your site and why. Top People Sharing Links to Your Site Top People Sharing Links to Your Site Top People Sharing Links to Your Site By Number of Tweets By Number of Visits By Visits per Tweet facetweet1388 seo1302 hoot1388 simplymeasured Cygfred TwitLimericks pigebak marramalho kmmpatke KevinSaysThings ninoscam gabejoynt dacort tabithagold resourcewater 26 24 13 6 5 5 5 4 4 4 4 4 4 3 simplymeasured KevinSaysThings otis dacort ElvisMoyaTTU owenblacker dixxieland AndyLloydGordon jakeludington 5ense KatHolmlund zdeluca adamshostack _vio_ paulabraconnot 73 14 12 6 5 5 5 4 4 3 3 3 3 3 simplymeasured otis ElvisMoyaTTU owenblacker dixxieland AndyLlyodGordon jakeludington KevinSaysThings 5ense zdeluca adamshostack paulabraconnot danbrookman mattisranting JuritaKruma 12 12 5 5 5 4 4 4 3 3 3 3 3 2 Understand where your traffic is coming from; segment earned versus owned visits. Are your own Tweets the primary driver for traffic to your site, or are others directing traffic for you? Your focus should be to create more engagement with your Tweets, encouraging others to contribute more content that links to your site. Explore user profiles that are successful at driving traffic, especially those that frequently share out Tweets linking to your site. Users who engage their audiences with Tweets to your site likely have audiences with particular interest in your brand. Discover what content resonates with these audiences, whether it’s products, promotions, or industry thought leadership. 37
  • 38. 4 ways to improve customer service measurement on Twitter In a recent study, we found that 34% of top brands now have a dedicated Twitter account for customer service. While Twitter programs are typically owned by the marketing/PR department, customer service is a key secondary activity. Many brands are investing efforts in solving customer problems via Twitter. The challenge is that when efforts are shifted from marketing to customer satisfaction, measurement strategies must also change to effectively evaluate performance. It is important that brands establish a customer service specific approach to measurement. We’ve outlined four strategies to help your brand do just that, using Delta Airlines as an example. 1. Measure end-to-end performance Brands need to ensure that they are measuring success throughout the entire process. Twitter Customer Service Workflow How many times was our brand mentioned across profiles? How many times did our CS Account send a reply? How many unique people did CS talk with? How many high volume issues were addressed? (4 or more responses) 28,334 4,104 2,263 92 Total Mentions CS Responses Unique People Complex Cases 91% 14.5 % 2,343 1.54 Percent of all responses were from CS. Percent of Total Mentions with CS Responses. Avg. Followers Per Person Engaging. Avg. Number of CS responses per user. 18.7 mins. Average Response Time Sunday Most active Day of Week 11:00 AM Most Active Time of Day Work flow and processes should be analyzed end-to-end – from customer demand to how efficiently issues were resolved. The visual above is a high-level summary that walks through the customer support process. This summary view provides a period snapshot that can be used for benchmarking customer service KPIs and optimizing overall performance. Quick response times and a high response rate are key to optimizing performance, but there is room for improvement if brands are experiencing a high volume of complex cases, or unexpected changes in the percentage of support responses being sent. 38
  • 39. 2. Capture complete coverage Often a brand’s Twitter presence consists of more than one branded Twitter handle. For example, a brand might have a primary marketing handle and a dedicated customer service handle. For big brands, multiple marketing handles might rely on one dedicated customer service account. Coverage by Account Total Mentions Total Replies from @DeltaAssist 25k 19.6k 20k 15k 14.5k 10k 5k 3,176 928 0 @DeltaAssist @Delta Regardless of how a brand is managing customer service, it is important to measure all of its efforts. The chart above displays total mentions for Delta’s primary marketing handle, @delta, and its customer support handle, @DeltaAssist. By measuring customer service activity across multiple accounts, brands can determine how support resources are being allocated and measure customer response times specific to each account. The 14.5K mentions of @DeltaAssist indicate that Delta customers are directing replies to the dedicated customer service handle and shows that many customer support issues are making it to their intended destination. 39
  • 40. 3. Shift context: focus on core customer service KPIs Measuring customer service performance on Twitter requires a different set of KPIs from those associated with marketing performance. Follower and engagement growth are standard marketing KPIs. However, support handles should view these metrics as potential red flags rather than account management success. When it comes to customer support, the goal is to respond to and resolve as many customer service issues as possible, as quickly as possible. The KPIs that matter most are Response Rate and Response Time. % of Mentions With @DeltaAssist Replies 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 12 12 9/ 0/ /2 /3 11 2 12 8/ /2 11 12 7/ 1 /2 11 11 12 /2 6/ 12 5/ /2 11 11 12 /2 4/ 12 3/ /2 11 11 12 /2 2/ 12 1/ /2 11 11 12 0/ 9/ /2 /1 11 2 12 8/ /1 11 11 12 7/ 1 /1 12 6/ /1 11 11 12 /1 5/ 12 4/ /1 11 11 12 11 /1 3/ 12 2/ 1/ /1 /1 11 2 12 0/ /1 11 2 /1 /9 11 2 /1 /8 11 11 2 /7 /1 2 /1 11 /1 /6 /5 11 2 12 /4 / 11 2 /1 /3 11 /1 11 /1 /1 11 11 /2 2 0% The chart above shows the percentage of total mentions that @DeltaAssist responded to in November. Keep in mind that not all brand mentions are customer-service related, and of those that are, not all warrant a response. The nature of customer engagement on Twitter varies for each brand. Brands should identify a target response rate and set goals for improvement by conducting ongoing monitoring and competitor benchmarking. 40
  • 41. Once a consistent response rate has been established, it can act as a good indicator of whether a brand has the resource to scale customer service to meet increases in customer demand. % of Mentions With @DeltaAssist Replies Avg. Time per Response 90 Minutes per Response (Goal) 80 70 60 50 40 30 20 10 12 12 0/ 9/ /2 /3 11 2 12 8/ /2 11 12 7/ 1 /2 11 11 12 /2 6/ 12 5/ /2 11 12 4/ /2 11 11 12 /2 3/ 12 2/ /2 11 11 12 11 /2 1/ 12 0/ 9/ /2 /1 11 2 12 8/ 11 11 /1 7/ 1 12 11 /1 12 /1 6/ 12 5/ /1 11 11 12 /1 4/ 12 3/ /1 11 11 12 2/ 1/ /1 /1 11 2 12 0/ /1 11 2 /1 11 /1 /9 /8 11 2 2 /7 /1 11 11 2 /1 /1 /6 /5 11 11 2 12 /4 / 2 /1 /3 11 11 /1 /1 /1 11 11 /2 2 0 Goals for response time must also be established through monitoring and benchmarking. Above, @DeltaAssist’s daily average response time is displayed for November. Response time measures a brand’s ability to respond quickly and is also an indicator as to whether it can quickly meet increases in demand. However, response time isn’t a standalone metric. Response time can falsely signal success if brands are simply responding quickly to recent issues and ignoring aging issues. Response Time and Response Rate must be measured in tandem when analyzing overall customer service performance. 41
  • 42. 4. Identify ongoing trends Aggregated data can be very useful for identifying ongoing customer service trends. Twitter Account Activity By Hour Business Hours Response Time User Mentions @DeltaAssist Replies 50 Influencer Tweet 4.5K 45 0 10 :0 0 PM 00 8: 00 6: 4: 00 PM 00 2: 0 :0 12 10 :0 0 AM 00 8: 00 6: 4: 00 AM 00 2: 0 :0 12 PM 5 0 PM 10 PM 15 1.0K PM 20 1.5K AM 25 2.0K AM 30 2.5K AM 35 3.0K AM 40 3.5K 500 Mentions & Replies 4.0K Response Time (In Hours) 5.0K The chart above displays @DeltaAssist’s account activity by hour in November. This trended view makes it possible to quickly identify whether customer service response scaled with brand mentions. It also clearly shows patterns in response time. The 6:00PM spike in user mentions deserves attention. Replies did not scale with the increased mentions, although response time was improved. Mentions peaked at 6:00PM after an angry customer Tweet was heavily retweeted. Improved response time indicates that @DeltaAssist was quickly engaging where it could, but likely avoided mentions it did not think were worth responding to. Identifying trends that capture complete coverage and focus on core customer service KPIs can be very useful for spotting issues and making decisions regarding resource allocation. 42
  • 43. How to tell if your Twitter campaign actually worked You’ve spent countless hours preparing, strategizing, and planning for your Twitter campaign. You put your plan into action, worked overtime making sure it was executed properly…and now you’re pretty sure it went well. But in an industry that focuses on measurable goals through specific metrics, pretty sure isn’t good enough anymore You’re going to want to know if your efforts were worth it. Did your campaign result in an increase in customers, better brand awareness, or more engaged users? These are important questions in determining whether your efforts were worth repeating. It’s important to conduct a “post-mortem analysis” to determine what you did wrong, what you did right, and what you can do next time to improve. As you dig into your Twitter analytics to unearth these insights, there are several key metrics that can help you better understand your campaign and how well it worked. Follower growth Total Account Followers Total Followers 6.0K 232K 230K 228K 226K 224K 222K 220K 218K 216K 214K Followers Added 5.0K 4.0K 3.0K 2.0K 1.0K 13 4/ 7/ 2 13 3/ 7/ 2 13 2/ 7/ 2 13 1/ 7/ 2 13 0/ 7/ 2 13 9/ 7/ 1 8/ 7/ 1 7/ 1 7/ 1 3 13 0 Followers Followers Added The easiest way to tell if your campaign had a meaningful impact is to look at follower growth and how it correlated with your campaign timeline. Did spikes in follower growth line up with key messaging from your campaign? Did influencer involvement drive a significant increase in new followers? There’s a lot of information that can be gleaned from follower trends, including something we call “campaign lift.” Was there a sustained increase in your follower growth rate after the campaign was over? Be sure to take that into consideration. If you saw a massive peak in fans and then a quick return to normal, you may need to reevaluate some of your campaign tactics. 43
  • 44. Engagement Brand Tweets and Engagement Brand Tweets Total Engagement 45 300 40 Brand Tweets 30 200 25 150 20 15 100 10 Total Engagement 250 35 50 5 13 1/ 13 7/ 3 9/ 13 7/ 2 13 5/ 72 71 / 13 7/ 2 3/ 13 7/ 2 1/ 13 7/ 2 3 9/ 7/ 1 7/ 1 13 7/ 1 5/ 13 7/ 1 3/ 13 1/ 7/ 1 /1 3 3 3 7/ 1 7/ 9 7/ 7/ 1 /1 7/ 5 /1 7/ 3 /1 7/ 1 3 0 3 0 While follower growth can give you some great insight and set a barometer for your campaign health, one of your main goals was most likely to get followers involved. Examine the engagement trends surrounding your campaign – both before and after – to determine your campaign lift for engagement. Set a wide sample set with at least twice the time of your campaign period to use as a benchmark for standard engagement. If users didn’t respond to your messaging, you need to understand why. What can you do better? If they did respond positively, you can analyze key points from your outbound content that kept them involved. 44
  • 45. Web traffic Twitter Activity Funnel 73 95 882 38 Sent Tweets Retweets @ Mentions Link Mentions 2,575,528 Twitter Potential Impressions 1.18 Impressions per Person 2,185,125 Twitter Potential Reach 62 Visits to Site 0.07 Conversions per Visit 0.12% CTR 12 Goal Completions This is our favorite feature of Simply Measured’s Twitter Analytics suite. With the Twitter Traffic Report, you can tie your Twitter data in with your Google Analytics data to see how your brand efforts during your campaign translated to actual website visits and even which goals (from your Google Analytics goal set) were completed. This is the most concrete way to tell if your campaign worked. If you can determine the actual site traffic driven from your efforts and, in turn, the goal completions driven by your Twitter activity, you can determine a base-level ROI of your Twitter campaign. While it may not reflect residual brand awareness and later referral traffic, it can set the tone for future campaigns. 45
  • 46. Influence Most Engaged Users billxaruto RaulElcheVettel moreira_osman DamienSeuzaret stinglove002319 PierrickLM lirrycupcake Myll_Erik Rhino_219 RedBullBOS SmalexTheSkater AllejanDR0oo redbullw1ngs VulgarDaClown Gabearchambault Most Followed Users 54 40 33 31 29 28 27 26 25 23 22 22 22 21 20 RyanSheckler MaciBookoutMTV BigBoi John_Wall GameOfThrones JonahLupton GoPro marinapajon DefJamRecords brentnhunter XGames abc_es redbullracing AntonioMartez BrooklynNets Top Users by Klout Score 2.4M 1.2M 748.4K 726.9K 650.4K 546.8K 544.6K 530.1K 393.1K 369.5K 317.2K 299.5K 253.7K 242.6K 242.5K DefJamRecords SPIEGELONLINE GameOfThrones exclaimdotca Applebees BrooklynNets XGames pizzahut BigBoi Jason redbullracing Dolby KeystoneMtn AtlanticStation SnowboardMag 89 89 88 88 88 88 87 87 86 86 86 85 84 84 83 When running any kind of campaign, your goal isn’t to simply reach people – it’s reaching the right people. With Twitter, you can easily see whether or not you did. How influential was the audience you engaged? Not only can this be valuable insight for your campaign’s success, it can give you targeted goals and opportunities for future ones. 46
  • 47. Impressions & Reach Impressions vs. Reach Total Impressions Total Reach 500K 450K 400K 350K 300K 250K 200K 150K 100K 50K 3 5/ 30 /1 3 3 /1 5/ 27 /1 3 /1 5/ 24 3 /1 5/ 21 3 18 5/ /1 15 5/ /1 3 13 9/ 12 5/ 13 6/ 5/ 13 5/ 3 3/ /1 30 4/ 5/ 3 3 /1 4/ 27 /1 3 /1 4/ 24 3 21 4/ /1 3 18 4/ /1 15 4/ /1 3 13 9/ 12 4/ 13 6/ 4/ 3/ 4/ 4/ 13 O The two metrics that tie in the audience of your followers – impressions and reach – can be a great way to tell if you were successful. Benchmark against standard impressions and reach, discover the key reasons for increases during your campaign, and set the tone for future activity. 47
  • 48. 7 ways to measure influence on Twitter Twitter is a goldmine for social marketers. The audience is there, the conversation is rich, and the power players are actively engaging. What do we mean by power players? That depends on a lot of different factors: Your product, your goals, your engagement strategy, etc. But it boils down to one key factor that we often ignore: Influencers. Let’s take a look at some of the different ways you can measure influence on Twitter. Measure your most engaged followers Which users are interacting with your brand regularly? These are your brand ambassadors. They’re promoting and engaging with your content in ways that no one else is. It’s important to focus on the types of Tweets that engage them. There’s a lot to be learned there. Look at which types of activity engage your most active users, and utilize them to drive an even larger audience for your brand. Most Engaged Users JetBlueNews LFANTBRAND Runway1R Yaerospace SteveMAbrams WanderingAramean Mdholidayz yankees368 VivaJazzOrlando LonesTarLou tvladeck autiglobetrot htctest02 JeanetteJoy DaniQT314 48 151 53 30 28 28 26 25 23 21 20 18 17 16 16 16
  • 49. Identify your most followed followers Which of the users that you’ve engaged have large audiences? What did you do to get them involved with your Tweet? These users have the potential to influence large audiences beyond your own. By identifying them, you can foster that relationship and increase the types of content that they’re interested in. If these influencers are engaging with different content from the rest of your audience, it’s important to identify what separates them from the rest of the pack. Most Followed Users GMA TheSamSlater HarvardBiz Jillzarin FortuneMagazine AmericanAir McHypeMedia TheRealRaye MILKTYSON CNTraveler theoverheadwire chrisborgan DanKimRedMango kanygarcia fodorstravel 1.8M 961.9K 732.6K 547.6K 542.4K 375.6K 374.1K 292.6K 226.3K 224.7K 217.8K 210.1K 206.7K 179.6K 170.2K Analyze the visits driven by users Top 10 Users Driving Visits to Your Site Size of bubble = # of Tweets simplymeasured 1.2K rsarver 1.0K randfish 800 bahoo 600 MissBerry206 400 BessemerVP 200 Visits 1.4K KevinSaysThings 0 0 50K 100K 150K 200K Followers 250K 300K 350K jaybaer Audience Stats 2,020 39 1.24 Avg. # of Followers Avg. Klout Score Avg. Tweets per Person Traffic Stats - Top Users by Visits 1,633 By Top 10 Influencers 1,357 All Other Users KristinEIDE Influence is more than just the number of times people reply to your Tweets. If your primary job is to engage potential customers on Twitter, there’s a good chance you have some bottom-line goals to meet. By tying Google Analytics into your Twitter reporting, we’re able to break down the individual Twitter users driving site traffic for your brand. 49
  • 50. Break down the people Tweeting links Just like you want to identify the people amplifying your Tweets, you want to know who is amplifying your website content. By identifying users sharing the most links to your site, you can create engagement campaigns to thank them for acting as ambassadors, create content geared towards them and their audience, and focus your efforts on earning their continued involvement. Top People Sharing Links to Your Site By Number of Tweets ptnovaodessa IfUSeekLeo McBoessio pt_novaodessa Bjorn_ds isqais20 AlbionKryeziu nodessa elgrodo Joe2596 tipsJP _justpp liesatrosewood PigglyBigglyBoy earthtoalliee 13 10 6 5 5 4 4 3 3 3 3 3 3 3 3 Measure the Tweets driving the most visits Identifying the users who have the greatest reach with the least effort is a crucial step in the influencer relations process. If a user has an audience so engaged that they can Tweet a link to your site once and get a huge response, you’ll want to focus on increasing the amount of content they’re willing to share. By identifying the top visits-per-Tweet, you can identify these users and engage them further. Top People Sharing Links to Your Site By Visits per Tweet i999y _CallMeJohn start_write DonaBibes carofjx Far0s hollyabingd0n CasperDeWinter bellaale thijsvdvaart ItsMeKillian Feadare prezlinator mickhinds REVEALERblog 50 26 20 20 17 16 16 16 14 14 12 12 11 12 11 10
  • 51. TOOLS Example reports Understand the performance of your brand’s Twitter account. Answer questions about Twitter Account performance and engagement, as well as followers, sent Tweets, retweets, mentions and engagement trends. Dig into the most effective Tweets and the impact they’ve had on your account and your audience. Twitter Account Report (JetBlue) Twitter Follower Report (StarwoodBuzz) Twitter Audience Analysis (Tide) Multiple Twitter Channel Analysis (IPG Agencies) Vine Tweet Analysis (chrisbrogan) Benchmark your brand’s performance against competitors Analyze the audience, content and engagement trends of any Twitter account, allowing you to understand your competitive position, relative performance, and market share compared to your competitors. Twitter Competitive Analysis (Hotel Chains) Measure your customer service efforts on Twitter Dive into individual user Tweets, track response rates and times, and analyze a dedicated customer service handle in relation to your main brand account to understand your customer service effectiveness. Twitter Customer Service Report (Ford) Multiple Twitter Customer Service Report (Ford) Compare your efforts on Twitter against those on other social networks Measure your efforts on Twitter in context with Facebook, Instagram, Tumblr, YouTube, Google+, and more. Compare audience size & growth, as well as post engagement across all major networks. Complete Social Media Snapshot (RedBull) 51
  • 52. ABOUT SIMPLY MEASURED Simply Measured is a fast-growing team of data geeks dedicated to making the world of analytics and reporting a better, more beautiful place. Our goal is to put the tools to understand business data in the hands of business users. We think reporting should be simple, beautiful, and accessible for everyone – not just data scientists. Our software streamlines the process from data to deliverables and eliminates the countless hours spent on everyday reporting tasks. We do this by putting cloud data sources at your fingertips, providing a marketplace of best practice reports, and generating beautiful deliverables on the web, in Excel, and in PowerPoint with a couple of clicks. Want to try Simply Measured? Request a Free 14 Day Trial Copyright © 2010–2014 Simply Measured, Inc. All Rights Reserved.