Social Media Strategy for Increased Activity A case study at Ericsson Labs of how off-site activity affects on-site activity

MIKAEL

PALM

Master of Science Thesis Stockholm, Sweden 2010

Social Media Strategy for Increased Activity A case study at Ericsson Labs of how off-site activity affects on-site activity

MIKAEL

PALM

Master’s Thesis in Media Technology (30 ECTS credits) at the School of Media Technology Royal Institute of Technology year 2010 Supervisor at CSC was Christer Lie Examiner was Nils Enlund TRITA-CSC-E 2010:148 ISRN-KTH/CSC/E--10/148--SE ISSN-1653-5715

Royal Institute of Technology School of Computer Science and Communication KTH CSC SE-100 44 Stockholm, Sweden URL: www.kth.se/csc

Abstract Social media is defined as any form of online publication or presence that allows end users to engage in multi-directional conversations in or around the content on the website. (Coleman 2010) Modern technologies found on the Internet opens up new social ways of communicating and interacting between persons, this also effects how knowledge is received and shared. Back in 2007 (Dunn 2007) companies blocked Facebook for their employees because it was a time consuming activity, but recently more organizations realize the fantastic advantage in opening these channels. Social media opens up great opportunities for companies to gather their crowd and converse with them quickly, exchange ideas, brainstorm and receive feedback on their products. Since the corporate attitude changes towards being interested in social media, strategies for companies is in need of being defined. This master thesis at Ericsson Labs (Labs.ericsson.com) is delimitated to look at the strategies Ericsson Labs use for social media and how they should act. Ericsson Labs has been working since the fall of 2009 with social media and has seen great results in publicity. This paper focuses on strategies how to create maximum spin-off-effect from social media to traffic on the home page. In so doing a monitoring system were created to follow movements in the social media channels that were used. Carefully designed social media activities were performed and monitored during weeks of measuring. By applying different campaigns and activities in the social media channels, numbers were affected. This resulted in a theory of how to behave as a brand in social media and what content to focus on. The empirical study of the movements in activity created an idea of how to run social media campaigns and was added to the already existing strategy at Ericsson. This report suggest that content ought to have a high news value and a high core business value, and also that social media channels should be synchronized, or work in cooperation with each other, for maximum impact.

Sammanfattning Denna studie har genomförts på Ericsson Labs, en avdelning underordnad Ericsson Research i Stockholm Kista. Ericsson Labs är ett forskningsprojekt startat våren 2009 som går ut på att skapa, hantera och distrubuera APIÊer. APIÊerna används som små delar av program vilka externa programutvecklare kan använda som byggstenar för att skapa applikationer. Dessa applikationer kan användas till iPhone Apps eller andra mobiltelefonplattormar och datorplatformar. APIÊerna distribueras via hemsidan Labs.ericsson.com till hela världen. Genom att skapa ett nätverk utav utveklare runt världen och studera deras arbete, positionerar sig Ericsson Labs i framkant med förståelsen om vilken typ utav applikationer som det finns behov utav. Detta, tillsammans med Ericssons breda kompetens att bygga trådlös kommunikation, erbjuder utvecklare sätt att implementera dessa applikationstjänster i det mobila nätet. Ericsson Labs använder sig i huvudsak av sociala medier för att marknadsföra sina APIÊer. Twitter är den mest frekventa kanalen och sedan LinkedIn, även YouTube, Facebook, Amplify, SlideShare och den egna bloggen används för att sprida Ericsson Labs på webben. Denna rapport är resutltatet utav veckovisa mätningar i Ericsson Labs sociala medie kanaler eller offsite kanaler, och on-site nyckelvärden. Totalt användes 17 mätvärden. Under mätningsveckorna utfördes också veckovisa specifika aktiviteter för att på något sätt påverka Ericsson Labs roll på webben. Ett exempel på aktivitet var att vara extra aktiv på Twitter under en vecka, eller driva en kampanj på Reddit för att se om det har i effekt i det viktigaste mätvärdet „downloaded APIÊs‰. För att mäta tredje parts siter användes automatiserade sökverktyg bland annat Google Alerts. Resultatet visar att det är fulltmöjligt att mäta sociala medier, det går att påverka ens position på webben genom att ha tydliga mål i sociala medier. Rapporten påvisar vikten av att synkronisera ens närvaro på den sociala webben och förslag på hur kampanjer kan drivas genom sociala kanaler ges. De sociala nätverken går att använda som verktyg för att nå affärsnytta. En modell för hur det ska gå till presenteras i rapporten.

Acknowledgement I would like to thank a lot of people for helping me in this work. First of my Supervisors at Ericsson Labs in Kista, Tor-Björn Minde and Gerardo Montero for being a great support to me. You have had a key role in the writing of this report. Thanks Tor for believing in my idea and develop it along with me, and Gerardo for patiently reading and discussing results with me. I would like to thank Christer Lie my supervisor at KTH for the support. Also a big thank you to the people I interviewed at Ericsson Kristin Lindkvist, Ulrika Bergström, Richard Brisius, Olof Lundström and Karin Ronander, thank you for your time. A big thanks to my friends at The Amazing Society, Tobias Franzén and Peter Ragndahl for the advice and knowledge about social media strategies. A big thank you to Stefan Hyttfors at Wenderfalck for pouring out a fantastic two hours session about the future of digital media to me personally a cold winters day in early 2010. And a special thanks to Therese Reuterswärd form Electrolux for sharing valuable information about business strategies in social media and her willingness to help out in this research. My friends at KTH Carl Gillbland, Olof Hansen-Swärd, Samir Fors, Adam Gardefjord and Said El Shobaki thank you for the support during writing. I also have to thank my wife Alice for the support and patience with me when I was writing the report.

Table of Contents 1. Introduction! 1.1 About this report! 1.1.1 Ericsson Labs! 1.1.2 Thesis! 1.1.3 Hypothesis! 1.2 Background - Social media at Ericsson Labs! 1.3 Social Media - definition! 1.4 Purpose of this thesis! 1.5 Method for solving the task! 1.6 Delimitations!

2. Theoretical background! 2.1 Social media development! 2.1.1 Human behavior, why social! 2.1.2 Trust and relations! 2.1.3 Influence! 2.2 social media strategies (collecting existing ideas)! 2.2.1 Twitter and Facebook!

3. Method! 3.1 Introduction! 3.2 Measuring social media at Ericsson Labs - Monitoring chart and KPIʼs! 3.2.1 When numbers were monitored! 3.2.2 How numbers were monitored! 3.2.3 Onsite KPIʼs measured with in-house tools! 3.2.4 Defining activites during brainstorming session! 3.2.5 Defining KPIʼs from brainstorming meeting.! 3.2.5.1 The 17 KPIʼs monitored 3.2.5.2 Comments on KPIʼs 3.2.6 Listening to the web - Google Alerts and RSS! 3.2.6.1 RSS-feeds 3.3 Interviews!

4. Result! 4.1 Early strategies at Ericsson Labs (2009-2010)! 4.1.1 “The Recipe” by Tor-Björn Minde! 4.2 Ericsson Labs China! 4.3 Global social media engagement within Ericsson! 4.4 Results from the monitoring chart! 4.4.1 Twitter! 4.4.2 Ratio between Twitter following/followers! 4.4.3 Onsite KPIʼs! 4.4.3.1 Registered Members KPI 4.4.3.2 Key usage each week KPI

1 1 1 2 2 2 2 2 3 3

4 4 4 5 5 6 6

9 9 9 10 10 10 11 12

13 13

14 14 16 16 17 17 18 18 18

4.4.3.3 Visits to homepage KPI 4.4.3.4 Downloaded new API keys KPI 4.4.4 Hacked API during the measuring period! 4.5 Missing numbers in measurements! 4.6 Interview results! 4.7 Results from social media activities! 4.7.1 Physical event! 4.7.2 Boost weeks, Twitter and LinkedIn! 4.7.3 Release of new API! 4.7.4 Reddit campaign! 4.7.5 Code for cash! 4.7.6 Ideas 4 Aps! 4.7.7 Webinar! 4.7.8 New homepage design!

5. Analysis! 5.1 Proposing the Social Media Content Square! 5.1.1 News value! 5.2 The recipe! 5.2.1 Making the recipe even more effective!

6. Improvements and suggestions ! 6.1 Suggestions for Ericsson! 6.1.1 Synchronized campaigns! 6.1.2 Monitoring chart! 6.1.3 Campaigns with high core and news value! 6.2 Listening to the web! 6.3 KPIʼs to monitor for Ericsson Labs! 6.4 Further research! 6.5 General suggestions for improving this research!

7. Conclusion! 7.1 Was the thesis proved?!

List of References ! Monitoring ChartLabs ! Appendix!

19 20 20 21 21 21 22 22 22 23 23 23

24 26 26 27 28

29 29 29 30 30 30 30 31 31

32 32

33 37 38

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

1. Introduction 1.1 About this report This report is written as a Master Thesis at KTH, The Royal Institute of Technology, at the Media Technology Section. The report was done on behalf of Ericsson to evaluate the social media strategies at Ericsson Labs based in Kista, Stockholm, Sweden. A visit to Ericsson HQ in Beijing was performed June 25 to synchronize findings and as a part of understanding the global perspective on social media. The report was written during the period march 2010 to august 2010. This thesis is written as a report on the existing social media strategies at Ericsson Labs, a result of an empirical study on activities performed during spring 2010 and a suggestion for the future, with a purpose of finding out how to increase traffic to a webpage with the use of social media. 1.1.1 Ericsson Labs Ericsson Labs was founded the 27th of November 2008 and is an open innovation initiative for beta applications and beta API’s and tools. It enables for entrepreneurs and developers to create applications for the Ericsson platform. The site Labs.ericsson.com guides developers in forums and blogs to drive the software development forward in an open API involvement. The fundamental idea with Ericsson Labs is to understand what functions users need and what they use. With the unique connection Ericsson Labs has to the market through social media they are able to follow the development on the market closely. Together with the online community they reach it is possible to be in the very forefront of application development. The stand out feature that Ericsson Labs offer is an deep understanding of mobile communication infrastructure. Ericsson can provide knowledge and methods on how to solve the tasks on a deeper level than what operators usually are able to do. 1

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Therefore developing a service or application along with Ericsson Labs simplifies the process of bringing an innovation to reality. Ericsson Labs wants to reach out and engage influencers, attract developers, grow and maintaining a certain number of APIÊs and applications and foster innovation. With a vision of developing the future of applications together with developers around the globe fits in well in a modern way of thinking. Reaching out through social media is then the natural way of collecting and conversing with the target group that can actualize the purpose of Ericsson Labs. 1.1.2 Thesis It is possible to influence the activity on the webpage by using selected social media channels combined with strategically chosen content and activities. 1.1.3 Hypothesis A general trend in new media is the social engagement among end users. Through new media channels, commonly known as social media, companies can affect traffic to be directed back to the companyÊs homepage, and thereby generate activity.

1.2 Background - Social media at Ericsson Labs Since May 2009 social media engagement at Ericsson Labs has been performed. They had tried several social media channels for instance Twitter, YouTube and Facebook. Mainly the head of Ericsson Labs has run all social media channels at Ericsson Labs. Since Ericsson Labs is a part of the research department of Ericsson, early adopting is a part of the everyday routine. During 2009 and early 2010 the social media engagements had a few times generated a viral spin that brought a lot of traffic to their webpage.

1.3 Social Media - definition Social media is defined as any form of online publication or presence that allows end users to engage in multi-directional conversations in or around the content on the website. (Coleman 2010)

1.4 Purpose of this thesis

2

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

The purpose of this thesis is to evaluate how to increase traffic to a web page with the use of social media.

1.5 Method for solving the task In order to get full insight and knowledge in how to solve this task a thorough study of the existing strategy were performed, interviews with people working in the line of business and also literature seeking, mainly over the Internet, were deeply looked into. From this base of knowledge a number of activities were carefully designed, see section 3.5.4, to be performed during the spring of 2010 and then investigated and analyzed to create a more perfected strategy how to act in social media. The activities were well monitored through 17 Key Performance Indicators, KPIÊs, numbers, see appendix 3. These numbers were monitored each Monday during a monitoring period of nine weeks. Numbers were monitored in an monitoring chart, see Monitoring Chart and Chapter 3. Because of the difficulty in finding printed scientific research on social media strategies, due to its reasonably young nature, most references are from Internet links. Also knowledge about social media is often, naturally, posted on blogs and on the web. Basically a majority of the target group for social media strategies spends a lot of time on the web, and therefore knowledge on this topic is found on the web.

1.6 Delimitations This report is looking at Ericsson Labs involvement in social media. How to affect the ongoing conversation on the Internet to create a spin around the webpage Labs.ericsson.com. Therefore it is delimited to the division of Ericsson Labs at Ericsson Research.

3

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

2. Theoretical background A short explanation about human behavior, some existing strategies that companies today use, a framework for the content of this Master Thesis.

2.1 Social media development This section gives insight in why social media is able to develop itself. It talks about how people interact and the principals behind being social.

2.1.1 Human behavior, why social Humans are by nature social in the sense that they cling together. When humans meet they almost intuitively invent tools to communicate, its like a basic need for their existence. Thus an oral language is developed and body language, and social communities are formed. With this follows relations and also organizations. Communication is a foundation in societies and nations. The most common way of communication is obviously the use of the oral language. This evolved into letters. The alphabet combined with the printing press later became an extended arm for communication. Along the way the telephone, television and fax machine etc. were invented. (Groves 2005) Normally humans easily adapt and learn how to use different communication channels. A new invention appears to be important if it can offer communication with another human being and in doing so develop a relation, develop trust between individuals. (Grooves 2005)

4

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

2.1.2 Trust and relations Trust means being able to predict what other people will do and what situations will occur. If we can surround ourselves with people we trust, then we can create a safe present and an even better future. (Starker D. 2008) Trust means making an exchange with someone when you do not have full knowledge about them, their intent or the things they are offering to you. (Starker D. 2008) 2.1.3 Influence The science behind influence in sociology, communication, marketing and political science has long been studied (Rogers 1962; Katz and Lazarfeld 1955). In following influence trends, forecasts of the future becomes the basis for how businesses operate and how society functions - for instance, see observations on how fashion spreads (Gladwell 2002) and how people vote (Berry and Keller 2003). Studying influence patterns can help us better understand why certain trends or innovations are adopted faster then others and how we could help advertisers and marketers design more effective campaigns. (Cha, Haddadi, Benevenuto and Gummadi 2010) Historically it has been difficult to study influence patterns due to the problem in tracking back where the idea started. Traditional communication theories state that a minority of users, called influentials, excels in persuading others. (Rogers 1962) A more modern view of looking at the influence theory is that the key factors determining influence are 1) the interpersonal relationship among ordinary users and 2) the readiness of a society to adopt an innovation. (Watts and Dodds 2007; Domingos and Richardson 2001) Today in the social medium influence can be measured more easily since tweets and messages can be tracked back in time and thus its easier to find a way back to where it all started. Therefore science in how influence is created in social media is developing. (Cha, Haddadi, Benevenuto and Gummadi 2010)

5

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

2.2 social media strategies (collecting existing ideas) As the digital market opens up, feedback is more or less instant. Any information posted digitally on the web can be copied and pasted, downloaded and shared. This enables a new way of acting and thinking when moving as a brand on the web. When a marketing campaign is published companies has to take in consideration the immediate feedback from the crowd it reaches. (Studstill 2010) When browsing the web and collecting information about social media strategy one are soon to find out that the number of self-declared social media experts grows speedily. (Martin 2010) Having realized this its easy to understand why the quality of the blog posts explaining social media strategies differ in quality. The UGC, user generated content, principals behind the web makes knowledge evolve in the same rate as the people who participate in the conversation learn. Therefore time will sort out the real professionals form the self-declared experts. Like in any other professional line of business, the strongest and best ideas will survive and the online community will glorify itself as time goes by. (Wunsch-Vincent and Vickery, 2007) Ideas and strategies are easily shared on the web and the philosophy many social media involved people have is the one of free information. Thoughts spread on blogs faster than the printing press allow. New media and thoughts about new media seldom make it in paper form. Therefore it has been hard to find social media strategies in print. Also the fact that the new media is new makes it hard to find research in an academic level. (Lidman 2010) A general idea about the social media evolvement is that the end users tend to lean towards a social digital media. Research show that people spend more and more time on the social web. When target groups move in quantity in this sphere, it is important for companies to move along into the social web. (Lidman 2010) 2.2.1 Twitter and Facebook A common idea ran in 2010 by companies as shown by three Swedish companies and organizations are the „follow us on Twitter and Facebook‰ strategy. As seen at Telia, TV4 and Barncancerfonden. (Telia #1, TV4 #1, Barncancerfonden #1). The idea is simply to collect a lot of people and then have a conversation with them. A successful Facebook group is the Trysil fan page (interview S. Hyttfors). Trysil is a ski resort in Norway. Quickly the group collected over 45000 followers. By interacting and posting engaging content the discussion within the group is heavily

6

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

active. Companies engaging their end-users usually have an engaging product, for instance traveling agencies, niche companies and rock stars. What Twitter and Facebook does is enabling an simple channel for communication between people. The principal about online communication has been around much longer in forums around the web. The difference now is the availability, readiness in the target group and readiness in the infrastructure that enables this attraction of so many end users. (interview Montero) Still the Twitter and Facebook strategies are young. Most company strategies for social media started 2010 and knowledge about how to act is still under investigation. An American based company, online PR media, has done deep studies in Facebook versus Twitter strategies for companies. A condensed compilation of their original study is quoted below. (Moir 2010)

Summary of Business-to-Business marketing advantages: Twitter advantages- Twitter performed better in the general business to business marketing category due to its viral marketing benefits. Viral Marketing is a core business-to-business social media application for Internet exposure and expediency. The 140 character Twitter Re-Tweets (RT's) that include links to popular or relevant business content can increase branding, drive website traffic and multiply outside of the Twitter network. Including being streamed in Google 'Real Time' searches and featured on many third party resource platforms. The study also concluded that it's not even necessary to have a large group of followers to benefit from Twitter's viral marketing advantages. Facebook advantages- Facebook performed better in the business-to-business marketing category for select companies that are 'personalities'. These are companies that are already universally recognized by their name, product or service. As such they naturally attract large followings and these followers often take the time to share relevant or important content among themselves and their followers. Summary of Business to Consumer marketing advantages: Facebook advantages- Facebook performed better in the majority of Business to Consumer marketing categories due to its format. A format that allows for the inclusion of photos, multiple outbound links, videos, surveys, promotions, direct engagement and other business to consumer applications. Facebook also facilitates the building of a large network of followers in the shortest amount of time to help market a consumer product or service. Twitter advantages- Twitter performed very well in the Business to Consumer marketing category where mobile real time applications were vital. Mobile real time applications involve announcements that are time sensitive, distributed through a wireless network and intended for your target audience. 7

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Announcements of a spontaneous nature involving special offers, breaking news or notification of impromptu public appearances. Conclusion- The larger study concludes that in order to maximize your Social Media Internet marketing results it's best to create a presence on both. However, if trying to decide which one to engage first or where to spend the most amount of your available time- It's important to perform basic Internet Market research, apply the findings to your target audience and weigh the benefits of each platform. The larger study also concluded that SEO combined with Social Media Marketing will yield greater potential results. (Moir 2010)

8

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

3. Method Method for solving driving the thesis were divided into three parts. Literature research and interviews - for knowledge Activities - to affect KPIÊs Empirical study - monitoring in social media

3.1 Introduction As mentioned in chapter 1, this Master Thesis investigates how Ericsson Labs should act towards itÊs target group, and from those conclusions reach a general description for brands in their way of acting on the social web. An empirical study were made over actions and behavior in social media in order to increase traffic to the homepage. The base for this study is the monitoring chart, see chapter 9. From there conclusions were extracted. During ten weeks of the measuring period specially designed social media activities were performed to see what kind of affects it would have on the KPIÊs, Key Performance Indicators.

3.2 Measuring social media at Ericsson Labs - Monitoring chart and KPI’s A hot topic of 2010 within the social media sphere is the discussion of how to measure and monitor social media (Naslund 2010). In finding out how to measure social media, research had to be done to find the best way to monitor the activity outside the homepage. After trying out several different monitoring systems such 9

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

as, Silverbakk, Ubervu, Radian6, Aitellu, Socialoomph, Twitteranalyzer, Tweetmeme, Yahoo Pipes, Twitter Search etc, a conclusion were made that creating a custom made manual excel-type KPI-chart would do deed at this particular time. Mostly the measuring tools performed well, but when working with them they added a complexity in measuring numbers that were avoided by measuring manually. Even when measuring manually problems with databases occurred. For instance during one week of measuring, one third of the tweets made by EricssonLabs had disappeared, (see monitoring chart chapter 9) and a few days later it was all back to normal again. The monitoring chart consisted of 17 KPIÊs, see 3.2.5.1. 3.2.1 When numbers were monitored Numbers were monitored every Monday at lunchtime. And measurements were taken two weeks prior to the start of the activities. Also Google Alerts were used to scan the web for third party blogs etc. This to find as much published material as possible. In the past the biggest spin-offs had come from when social objects were spread widely. 3.2.2 How numbers were monitored A spreadsheet, see chapter 9, was created where all the numbers were documented. The 14 offsite Key Performance Indicators were mostly public numbers, displayed when visiting the relevant webpage. For instance the Twitter following number is found when visiting twitter.com/ericssonLabs and the same goes with the Facebook KPIÊs they are easily found when logged on to the Ericsson Application Award page. The four onsite numbers are found through Ericsson Labs internal statistics services. Therefore these numbers are hidden in the spreadsheet in this public report. 3.2.3 Onsite KPIʼs measured with in-house tools For the KPIÊs New Registered members, New keys registered and Key usage each week internal statistics tool were used. The writer was given access to the webpage, Labs.ericsson.com, for limited period after the measuring period in august 2010. This enabled access to the statistics tool used internally so all relevant numbers could be extracted. The visits to homepage KPI were measured through PIWIK that is used by Ericsson Labs. PIWIK is an open source analytics tool for web analytics. PIWIK is not open to the public and a specific password had to be used for access. 10

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

3.2.4 Defining activites during brainstorming session A brainstorm session was held at Ericsson Labs with the writer of this report along with Tor-Björn Minde. The purpose of this meeting was to define what to measure and what social media activities that could be done. Ericsson Labs employees were invited and also an external consultant, Tobias Franzén from the Amazing Society, attended. A list of 23 different guidelines for social media was listed on the whiteboard. List 3.1 included these activities; * read evaluate third party blogs * third party links / blog posts * focus on quality content *Important target group influencers *Important developer *Track the business emails *Look at related context for e/// Labs *Increase awareness in linkeind *Relevant content to talk about *Address trends and important things. *BBC tech review

* Improve conversation on forum at Ericsson Labs * Get more dialog (Twitter) * YouTube template (splash screen) * Be more personal *A new fb campaign (mashup Twitter feed) *Demo stuff on fb *Profiles on lab site / of Labs terms *Webinar on enabler *Email list (follow & write) *Reddit *Independent *Carnival of bloggers

list 3.1 brainstormed ideas

From list 3.1 nine activities were extracted, they were scheduled in do be performed during the measuring weeks. Physical event, web2.0expo San Francisco Twitter Boost Week Reddit Campaign API Release „Ideas 4 Apps‰ Facebook Campaign New Homepage Design Webinar LinkedIn Boost Week API Release list 3.2 Social media activities scheduled for the measuring weeks 11

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

3.2.5 Defining KPIʼs from brainstorming meeting. During the brainstorm meeting a list of 17 different KPIÊs were defined. The 17 Key Performance Indicators, KPIÊs, have 4 numbers that are on-site and 13 off-site-, or social media, numbers. The thirteen off-site numbers were carefully picked to cover the most influential social media efforts where Ericsson Labs were active. This mainly include Twitter numbers. Basically the general idea to figure out what activities in the „off-site‰ numbers will have a spin-off effect in the „on-site‰ numbers. All these KPIÊs were of course measured in numbers. A main goal of Ericsson Labs is to have Developers Keys active, so the development of applications is taken forward. Therefore it is important to study these numbers and se how they interact with each other. A lot of questions were raised at the time the KPIÊs were defined, what numbers would effect others and how. Some numbers might also be unnecessary. Historically there had been times when a social object, like a YouTube clip, did draw thousands of visitors back to the homepage during the week of its release. Other times marketing activities had been performed but no apparent peak in the numbers had been evident. This showing that there is a link between these numbers, but itÊs complicated to understand the principles behind it. 3.2.5.1 The 17 KPIʼs monitored Onsite Key Usage each week New key registered Registered new members *total Visits to Homepage Offsite Twitter Following Twitter Followers Re-Tweets "RT EricssonLabs" Re-Tweets "Twitter" @ericssonLabs D ericssonLabs #tweets EricssonLabs #tweets this week Facebook # Fans "code for cash Youtube Channel views LinkedIn #group members

(list continues on next page) 12

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Amplify followers Amplify sources Amplify Posts 3.2.5.2 Comments on KPIʼs Re-Tweets "RT EricssonLabs" and Re-Tweets "Twitter" are counted as one KPI. They are basically the same number but Twitter use two different systems for Re-Tweeting. Therefore it is counted in as two numbers. 3.2.6 Listening to the web - Google Alerts and RSS Some time were spent in finding the right tool to listen to trends on the web. There are several tools to follow topics and blogs that are being produced. The tool that were used specific during the measuring period were Google Alerts (alerts.google.com). The service does Google searches automatically and sends the findings to an email address. The word „Ericsson Labs‰ was used for listening on the web. Google Alerts were also used with the search term „Social Media Strategy‰ and „Social Media Monitoring‰ for continuous update in trends of these topics. 3.2.6.1 RSS-feeds Also RSSÊs were used to follow blogs specific in social media. An RSS-feed is an XML-based file format that is used to subscribe to web-content. Basically one can follow a website and have newly posted material sent to ones RSS-reader. The RSSreader receives the feed and displays it, so in reality it looks like you get and email from a webpage. The RSS-reader used for this report was Google Reader (reader.google.com). List of some RSS-feeds followed are in appendixes.

3.3 Interviews The writer found people working in the forefront of social media, marketing, internal communication, consultants, researchers etc. These interviews were performed in an open discussion about social media strategies. No specific form or questionnaire was used along with the interviews. The main question asked was their respective attitude and their knowledge about social media strategies. Nine distinct interviews were held with the purpose of this thesis. During the meetings notes were taken. (See appendix section 10 for interview listings.)

13

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

4. Result

In finding our what gives more activity on the homepage, Labs.ericsson.com, this are the results from the empirical study, interviews and literature searching.

4.1 Early strategies at Ericsson Labs (2009-2010) Ericsson Labs is a web based product and service. It attracts entrepreneurs and developers widely spread across the world. Naturally, Internet campaign initiatives by Ericsson Labs gather people with developed computer skills, perhaps this is a reason why such an immediate response were seen when Ericsson Labs started working in social media. fig 4.1 Usage explosion showing visits to home page

With the attitude inspired by the development during 2009 for Labs.ericsson.com pointing when Twitter channel started (The Y-axis is without one-liner; „Marketing is too out scale on request from Ericsson Labs but describes the important to be left to the visits to hompege) marketing department‰ (Godin 2008) Ericsson Labs started using social media to increase visits to Ericsson Labs homepage. The visits to homepage were quadrupled in four months, May to August 2009, See fig 3.1.

14

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

During the first quarter of 2010 Ericsson Labs were active in nine social media networks. These networks were carefully chosen out of hundreds of different networks because they reach wide and deep into Ericsson Labs target groups. A strategy formed internally was to divide the networks in two segments, namely Social Conversation and Social Objects (Inspired by J. Engeström 2003-2010). The first group mentioned contains networks mainly to converse with and the second group contains social networks where objects are created. These sites cover the biggest of the social media networks in the first quarter of 2010. Each site fulfill its specific purpose; Social Conversation; Facebook - for campaigns LinkedIn - for professionals Amplify - for comments Twitter - for conversations Social Objects; Slideshare - for slides YouTube - for videos Ericsson Labs blogs - for sharing developers content Enabler blogs - for external comments Del.icio.us - for relevant links Alongside with these social media networks, traditional events at hackdays and events has been used in combination with campaigns.

fig 4.2 Displaying how most social media content moves from Ericsson Labs

15

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Most of the traffic produced by Ericsson Labs in social media is published through the Twitter account @ericssonLabs, fig 3.2. Therefore Twitter is the most influential and important social media channel for Ericsson Labs. The account gathers about 100 till 200 new followers each week and the @ericssonLabs account is in the top 6 swedish Twitter company accounts. (Kullin 2010) 4.1.1 “The Recipe” by Tor-Björn Minde Distribution

Content

Sub Content

1/3

Core business related, Promotion, Branding, Company news etc

*Brand *Company *Labs

1/3

Business topic related, 3rd party blogs, External research etc

n/a

1/3

Socializing, Re-posting

n/a

table 4.1 Explains “the recipe” by Tor-Björn Minde

Head of Ericsson Labs, Tor-Björn Minde, has developed, a recipe for content in social media. Minde believes that by distributing one third of the content spread on core business related things, and one third with related to topic content, and also the last third of content posted being just social on the web, when doing this the content produced is not too pushy or overbearing neither is it too social. (Interview with Minde 2010) Over time this has been a successful strategy. Since numbers and results has been growing continuously with this recipe in operation.

4.2 Ericsson Labs China The writer of this report visited Ericsson HQ in Bejing at June 25th 2010, a meeting with Ericsson Labs China division were held. Ericsson Labs is starting up in China. The aim is to „copy‰ the work that has been done in Sweden to reach out to the massive market in China. Due to the „Great Firewall of China‰ social media sites like Facebook, Twitter, Flickr and YouTube are blocked. Therefore the work of marketing Ericsson Labs China has to be done through chinese government approved channels. For instance renren.com is a Facebook copy with over a hundred million user accounts (2010) in China. In other words there are sites in 16

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

China doing the same type of job as the dominating social media sites Ericsson Labs are using outside China, one site equivalent for YouTube and another for Twitter etc. A phrase used at the Ericsson office in Beijing is CtoC, and when used is not referring to the common social media term Âcustomer to customer, but rather meaning ÂCopy to ChinaÊ. At times when good blogpost etc are posted on the @ericssonLabs Twitter account it has to be CtoC, copied to China, manually to reach the audience in China. This fact that information spread over the globe doesnÊt reach in to China shows the importance of studying social media principles deeper. In understanding how people act and communicate in social media one can move across platforms with strategies not only being bound to specific sites like Twitter etc. At one point UGCsite myspace were considered immortal due to its massive reach, over a 100 million accounts (Cashmore 2006) but itÊs user activity has declined dramatically over a short period. This again underlining the importance of being platform independent.

4.3 Global social media engagement within Ericsson During 2009 a big change in the core organization structure of Ericsson were designed. New CEO, Hans Vestberg is emphasizing an attitude in learning bottomup rather than a traditional top-down system used pervious. Top management at Ericsson is now pushing for web 2.0 solutions and this enables social media engagements in a new way. While interviewing the communication director at Ericsson, she stated clearly the situation with Ericsson. Ericsson only has a few main costumers, and Ericsson is mainly running business-to-business deals with them, therefore spending time on social media is not a priority for gaining business opportunities. These few main costumers are already in a close relationship with Ericsson on a day-to-day business. On the other hand for marketing and branding it is always right to be active in social media. Basically Ericsson are aiming to be in the forefront of social media development and are well on the way of reaching there.

4.4 Results from the monitoring chart The monitoring spreadsheet, chapter 9, stands as the main hub in this report. From the numbers documented in there conclusions are drawn. The monitoring chart was the tool used to find how activities in social media affect the main webpage 17

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

and the products produced by Ericsson Labs. Here is some of the most important results form the chart.

4.4.1 Twitter Looking at the Twitter numbers they were growing steadily each week. Twitter followers grew with an average of 105 listeners each week and with an impressive 204 following each week. With a number of 8100 following and 7632 followers, August 2010 puts Tor-Björn Minde in a top 20 position (Hillman, A 2010) on the Swedish Twitter account list and in a top 6 in corporate tweeters in Sweden (Kullin, H 2010). Publishing about 50 tweets a week thus making about 7 tweets per day, is a good number for staying in contact with the audience. There is no right or wrong in how many tweets to do each day (Sysomos 2010). 4.4.2 Ratio between Twitter following/followers The ration between Twitter following/followers gives an average each week of 0.949 showing the eager of the publisher to find new listeners and that he actively listens to the audience. Having just about the same number of followers as he follows.

4.4.3 Onsite KPIʼs

When looking deeper into the KPIÊs onsite. The registered members number increased with an average of 175.5 members each week. Mind you that Ericsson Labs were subject to a hacker attack during the measuring period. Several accounts were created by the hackers only to generate spam SMS. The number of API keys downloaded each week runs stabile and growing steadily in pretty much the same pace each week. This is when numbers are looked at excluding the hacked SMS Send and Receive numbers.

4.4.3.1 Registered Members KPI The onsite KPIÊs grew steadily each week during the measuring period. Registered members grew about half of the rate as followers on Twitter.

18

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

4.4.3.2 Key usage each week KPI Probably the most dynamic number of all KPIÊs measured is the „Key Usage Each Week (without „sms s/r‰)‰. It changed the most from week to week, differing more than 100% from the average from week to week. There is no apparent pattern related to the offsite numbers in how this KPI moves. This is probably due to the fact that developers test their applications during development independent to social media activities. Also when a new application is released containing an API. 4.4.3.3 Visits to homepage KPI This KPI were stabile and didnÊt move either up of down. At a first glance at the monitoring chart it looks as though no significant peaks in the „visitors to homepage‰ were performed. During the nine weeks of monitoring, the visits to homepage show that there has been no increase in visits. It pretty much kept itself unchanged. It peaks at 6.3% over the average visits per week and it also hits the bottom at 4.5% under the average visitors per week. Looking at these numbers in an Gaussian Distribution or normal distribution it gives to little of difference in order to give a result that can for sure tell us that our activity were the reason that the numbers changed. 4.4.3.4 Downloaded new API keys KPI Downloaded New Keys is a crucial measurement. It shows the number of new active users. This number was quite stabile neither growing or decreasing. Some weeks showed an 18% increase from the average on the measured week giving the best number. In opposition there was a minimum of 8% below average. There is no apparent pattern in the movement of this number. Looking in the monitor chart it looks like random selected numbers.

4.4.4 Hacked API during the measuring period During the measuring period the API; SMS Send and Receive were subject to a hacking attack. Basically parallel accounts were generated and took out SMS Send and Receive keys and sent SMSÊs. It was discovered before it went too far, though it is evident in the measuring chart. The last week of May the SMS API has 93.8% of the total usage that week. Keep in mind that there is 22 other APIÊs sharing the rest of the 6.2% usage. The API were shut down when discovered it being hacked.

19

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

fig 4.3 Each (colored) line represent an API Key. Y-axis represent the number of keys downlaoded each timeperod. Purple (the peak) is the API that got hacked in May 2010, and then turned off in early June, where the peak falls down to zero. This obviously had a big impact in analyzing numbers in the end. (The y-axis is confidential scale on request from Ericsson Labs)

4.5 Missing numbers in measurements Twice during the monitoring weeks there were „missing numbers‰, for instance measuring the total of YouTube channel views were 89335 and the following week there were a total of 88917 views. For some reason there were -418 views from the week before, an impossible measure. The number of total views can not decrease. The same thing happened with a Twitter measurement. One week a third of the total number of tweets were missing from the numbers. This opens up new questions how reliable the numbers really are, and when and how they are updated.

4.6 Interview results The open discussions about social media, both from in-house and external workers gave knowledge well worth the time spent. A general idea that sprung out of these meetings were the idea that social media will grow in the future. New media will have a more disruptive impact on society and for marketing. (see appendix 1: list of people interviewed)

20

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

4.7 Results from social media activities In looking at the nine activities there is a pronounced difference in the activities. Some are directly social and others lean towards the core business products. The activities were ran during consecutive weeks starting from the first week of May 2010, week 18 of the year, with the first specially designed activity starting two weeks later, week 20. 4.7.1 Physical event First out was an Physical Event at the Web 2.0 Expo in San Francisco. Ericsson Labs had a display at the fair and the expectations on this event were high. A couple of months earlier there had been a similar event held in Sweden and during that week a significant raise had been in the visits to the homepage. Looking at the numbers after this weeks Expo, no significant raise or peak were shown, instead a small but insignificant decrease in the visits to homepage were made this week. 4.7.2 Boost weeks, Twitter and LinkedIn Two boost weeks were performed; first out of the two were the Twitter boost week. During that week it was evident in measurement that the Twitter numbers increase. A list of the 25 most influential followers of the Twitter account was made so conversation could be focused on these people. Looking at the seven KPIÊs relating to Twitter all the numbers this week were up. ReTweets were up by 26% prior to the week before and @-mentions was up 60% compared to the week before. A peak in the Twitter account was made. When looking at the „Visits to Homepage‰ numbers this week, there is a small raise, 8%, from the week earlier but it is still not so significant. In fact this week didnÊt generate the most visits to homepage during the empirical study. The same tendency, but even more significant, was observed during the „LinkedIn boost week‰. The LinkedIn boost week was a simple but effective activity. Extra effort were put in to the Ericsson Labs LinkedIn group, posting relevant topics and links. Measurements show a 373% increase in „LinkedIn new group members per week‰ measurements. One possible reason why the Twitter measurements didnÊt increase as much during its boost week is that the Twitter account is active continuously in a different way than what the LinkedIn group is.

21

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

fig 4.4 Showing an analysis of the Social Media Activities. If an activity is positioned towards the left it has a higher value to the purpose of Ericsson Labs, and positioned to right itʼs an activity containing higher social value

4.7.3 Release of new API During the second week of monitoring social media a new API were released on the Labs homepage, network probe. The news of this release were posted on the social networks. This release had no evident impact in either the social media sphere or the onsite KPIÊs.

4.7.4 Reddit campaign Week five of the empirical study gave time for the Reddit campaign. Social Media expert at Ericsson Labs, Tor-Björn Minde made an effort in reaching a high ranking at Reddit.com by posting relevant and interesting blog links in the name of EricssonLabs. By having readers voting up the link the wider it spreads. Becoming famous on this site is a branding type of social activity. Ericsson Labs posted six links during this week. The biggest success was the link, „Learn HTML5: 10 Must Read Lessons‰. It got 201 votes in one month. Which is to be considered to be a success. The post generated a well-developed discussion of 58 comments. Looking at the visit to homepage numbers it was remaining the same level as all the other weeks. 4.7.5 Code for cash The results of the Facebook campaign „code for cash‰ showed no direct increase in members. The group only grew 1% during the 9 weeks of measurements. Keep in mind that there was no effort made to make the group grow and the „code for cash‰ campaign had its final day the in early June 2010, which was in the middle of the measuring weeks. 22

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

4.7.6 Ideas 4 Aps During the brainstorming meeting an Ideas 4 Apps-campaign were designed. This was meant to act as an Facebook group page where people that got an idea for an app could post their thoughts on the wall to reach out to developers etc. This would reach a new type of target group; The „new-thinkers‰ without programming skills. At the time when this empirical study took place the time and effort needed to run this project were not enough. It is a demanding job to gather a crowd in social media. Therefore this activity never took place. 4.7.7 Webinar Webinar, this activity never took place. It was suppose to be an informational seminar broadcasted on the web. 4.7.8 New homepage design New homepage design, were scheduled during the monitoring weeks. Ericsson Labs has been ran on an separate platform to the main ericsson.com webpage. It was in the pipeline to have the Labs webpage brought to the same platform as the main dotcom. The fact that Ericsson Labs has been on a separate platform, has enabled social media integration to be quick and manageable and therefore Ericsson Labs has been in the forefront of social media development.

23

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

5. Analysis In defining what gives a greater activity on the webpage, Labs.ericsson.com, this empirical study results in a couple of guidelines. There are a couple activities in spin-off-effects that are shown in the history of Ericsson Labs. The biggest spin-off were the short thirty second film explaining 3D-landscape API, were released on YouTube January 16 2010. This film has so far generated the biggest activity on the webpage. (see graph). It was posted on several famous blogs and spread widely across the web. Though 3D landscape clip on YouTube is, without doubt, the widest spread social object, the 3D landscape API still not the most downloaded. The API that tops the downloading statistics is the Mobile Location API with 51% of all downloads this year. 3D landscape is currently on a fifth place in most downloaded API, excluding SMS s/r. This indicates the theory that product that spreads widely in social media it does not necessarily be of best use for the target group. The 3D landscape video is to be considered as an entertaining video. The video of the 3d landscape had a high news value but as for usability for developers the grade were lower. Comparing the numbers generated during the empirical study week that were made along with this thesis, we can see that during those two months no significant spin or activity boost in the onsite KPIÊs is evident. When interviewing the Digital Manager at Electrolux, she mentioned that things spread over social networks is things that will give increased status among the group listening. This comes more apparent when analyzing status updates on social networks such as Facebook or Twitter. People do tweet things that draws attention and that in some ways creates a buzz. During the Twitter boost week, Twitter numbers increased and also during the LinkedIn focus week, LinkedIn numbers grew. This indicates that when social media tools intentionally used to gain activity numbers goes up. The fact that visits 24

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

to homepage numbers were not affected during the weeks monitored opens up a further question whether social media activity does have an impact on homepage numbers. Looking at statistics from early 2010, when homepage activity numbers have peaked, the peak is made in relation to a social media activity. It emphasizes a theory that there is a link between activity on homepage and social media activity.

fig 5.1 Social media content square

When studying the social media activities generating peaks, one conclusion is drawn, that when social media activity is related to the core purpose of the homepage it draws visitors to back to the homepage. This proposes a relation between content in social media objects and how it is spread and what kind of response it gets. Analyzing further it is shown that activities that have marketed products from Ericsson Labs that contained a „News Value‰ or „Social Value‰ were widely spread. According to Reuterswärd, that social object that promotes social prestige is passed on (Reuterswärd 2010). Therefore if an product does raise once prestige it will spread. The 3D landscape was a bit of a news breaking video and therefore it was passed on in the social web.

25

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

5.1 Proposing the Social Media Content Square During this research a formula for content on social media has been shaped. fig5.1 5.1.1 News value The writer suggest that social objects spreading on the web is due to the „News Value‰ the higher it gets the more likely the object is to spread on the web. For instance if one post something that is uninteresting for the listeners than it most likely wont be re-posted. A problem with a formula that contains „News Value‰ is that „News Value‰ is an subjective axel. What is a piece of news for one person might not be important for someone else. „News Value‰ depends on a lot of factors. During interview with Digtital Manager at Electrolux it was mentioned that a piece of news is depending on closeness, presentness and effects on direct fig. 5.2 Figure showing the maximum point in how to act in environment to the people social meida involed. The fact that there is a news value parameter in social media makes it unpredictable and difficult to grasp. Knowing exactly what the audience wants is key in creating a viral social object. Still this is a difficult forecast to predict. Through the results of measurements it is evident that spending time on social tools, talking or writing to people and networking with people generates increased numbers in the area where time is spent. For instance were there a 300 plus percentage increase in members joined in the LinkedIn group the week some effort were spent there. The same pattern is found with Twitter. Since the Twitter account has been continuously active at Ericsson Labs, followers continuously join the account. Also during the Twitter boost week numbers peaked and more people joined. From this a conclusion is made; being active on social networks generates activity on the network. Basically the administrator of the account is the driving force for keeping the account alive. ItÊs a simple conclusion, if thought of in extremes, if the account never were updated, nobody would visit. In opposition to this conclusion there is also a peak in how much one can tweet before tiring once 26

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

followers and subsequently lose followers. The activity formula on Twitter is depending on a number of factors, for instance celebrity level, content etc.

5.2 The recipe Referring back to the Social media recipe, see 2.4.1, research of this report gives that when activities are done in the „last‰ third of the recipe, which is the social part, numbers goes up in that area. Conclusions from this report indicates an add to the recipe strategy. A social object combining all three aspects of the recipe gives the biggest impact on the web. This means that an object containing a high „core value‰ that also is packaged with a high social value will spread widely and than appear in third party blogs etc. An object like that would contain all the parts of the social media recipe at once. Distribution

Content

Sub Content

1/3

Core business related, Promotion, Branding, Company news etc

*Brand *Company *Labs

1/3

Business topic related, 3rd party blogs, External research etc

n/a

1/3

Socializing, Re-posting

n/a

table 5.2 Displaying The Recipe by Tor-Björn Minde

In order to create this kind of social object it has to start from the business platform. In other words the products has to have a launch strategy which include all parts of the recipe. Basically news spreads in an unconventional way in comparison with traditional media where one sent press releases to printed press etc. Today news are launched directly to the target group through social media or new media. A product is launched on the web and the spread over the social network. Then as a final part of the marketing process the product launch does reach the 3rd party blogs. When a social object is picked up by an acknowledged blog often the spin is of, and it reaches back to a bigger social crowd, it gets reposted and new blog posts are produced again.

27

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

5.2.1 Making the recipe even more effective In the launching process all effort in social media should follow the concept of what is being released. In other words having all social objects referred to and posted synchronized could effect the trend on the web, and then raise the interest of the product being released.

28

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

6. Improvements and suggestions Results from this paper inspired for ideas for the future in how to act, here are some ideas how to improve and how to act in the future.

6.1 Suggestions for Ericsson -Synchronized campaigns -Monitoring Chart -Start strategies from the bottom 6.1.1 Synchronized campaigns It is proved in the monitoring chart that if effort is put in numbers goes up. And also that depending on content media is spread widely. Adding this together creates a new type of strategy for Ericsson Labs yet not tried. If content with a high news value is spread parallel over multiple channels, the message about new products may spread faster. For instance the launching of the 3D-landscape API could be launched in different ways in all social media channels. This means that the recipe only include 3D-landscape related articles and posts. In this way Ericsson Labs as a company could perhaps start a trend or follow a trend that is seen on the web and emphasize that trend by posting and talking and responding to conversations already running in that topic.

29

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

6.1.2 Monitoring chart The main hub for this report has been the monitoring of the social media activities and engagements. From the excel chart with numbers a lot of knowledge about how trends and costumers react and act on social media. Therefore the first suggestion is to have a monitoring system that enables overviews and trends. From here a lot of knowledge can be extracted and be a great learning and guide for future strategies. 6.1.3 Campaigns with high core and news value Taking it from the start to create a product with high core and news value would create a product that spreads easy on the social web. Making outstanding products to position the products of Ericsson Labs in a unique and innovative position would make the products spread wide and far quickly.

6.2 Listening to the web Research was made to find an easy way to listen to tends on the web. The tool that were used in the end for this report was „Google Alerts‰ and „Google Reader‰ those are two common, easy to use and free tools that still can be very powerful. In social media it is absolutely crucial to listen carefully to the web, what is being published and produced.

6.3 KPI’s to monitor for Ericsson Labs Obviously not all KPIÊs were as useful as others. The general principal behind what KPIÊs to measure or not is simple, basically measure was the work is put in. If we look at the Facebook measurements and the Amplify measurements they donÊt move that much during measuring periods basically because no specific work were put in to grow those numbers. And a lot of effort were put in to Twitter and to LinkedIn and numbers are growing visibly the weeks effort were put in. Therefor a monitoring chart should use the KPIÊs were the work is affected. Specific for Ericsson Labs the chart suggest to keep the onsite numbers and the Twitter and LinkedIn numbers and then to use Amplify, Facebook and YouTube for campaigns.

30

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

6.4 Further research The writer suggests to do further research in the field of social media strategies. Due to the young nature of social media the strategies quickly evolve and the conclusion of this report needs to be tried with an empirical study again to show the validity of the results given in this report. Does synchronized campaigns give higher activity on site? Is it possible to increase the core and news value in the social objects?

6.5 General suggestions for improving this research To start of, a lot of activities were planned and only four of them were carried out fully. This leave the monitoring chart a bit weak in its observations. Still this four activities gave result in the raised KPI numbers in its respective area. A suggestion from the writer of this report is to try out an opposite thesis. Basically to prove that if nothing is done in social media, numbers will go down in the onsite KPIÊs. Then the impact of the social media engagement perhaps would be more defined. Still to perform a research project like that would be harmful to Ericsson Labs but still it would answer the question whether social media activities does have an impact on the fundamental homepage.

31

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

7. Conclusion

This rapport shows that if effort is made to increase the activity in social media channels the activity in the social media channels does increase. It also indicates that if the content sent out in social media is solemnly social activity it is most likely to stay in the sphere where it was put in. In other words if a Twitter admin is being social it is most likely that no activity is sent back to the webpage. From events prior to the time when the measuring period of this report took place it is shown that spins can be created from the effort of planned activities. Still it is difficult to trace exactly the reasons why the spins take off. It is then proposed for further research that if an synchronized effort in parallel campaigning in multiple social media channels is performed that it increases the possibility of raising activity on the webpage. This indicates that a traditional push strategy is working on the social web, rather than pull strategies, that are widely suggested in the social media-marketing sphere. Perhaps social media strategies are not so different from any other marketing strategies.

7.1 Was the thesis proved? It is possible to influence the activity on the webpage through selected social media channels combined with strategically chosen content and activities. (see Chapter 1.1.1 thesis) It is shown that if specific effort is made in a specific social media channel then activity in that specific channel is increased. And it is suggested that the content that is spread in the specific social media channel determines the likelihood of the chance of spreading that activity back to the webpage.

32

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

List of References Barncancerfonden #1: Följ oss på facebook (Follow us on facebook) Available: http://www.barncancerfonden.se/Om-oss/Aktuellt/Folj-oss-paFacebook/ Downloaded: 21.09.10 Berry, J., and Keller, E. (2003) The Influentials: One American in Then Tells the Other Nine How to Vote, Where to Eat, and What to Buy. Free Press Cashmore, P 2006 MySpace Hits over 100 million Accounts Available: http://mashable.com/2006/08/09/myspace-hits-100-millionaccounts/ Downloaded: 21.09.10 Cha, M, Haddiadi, H, Benevenuto, F, and Gummadi, K,P, 2010: Measuring User Influence in Twitter: The Millione Follower Fallency. Coleman, A (2010) Glossary of Online Marketing Terms Available: http://www.onlinematters.com/glossary.htm Downloaded 21.09.10 Dougherty, H (2010) Facebook Reaches Top Ranking in US Available: http://weblogs.hitwise.com/heather-dougherty/2010/03/ facebook_reaches_top_ranking_i.html Downloaded 21.09.10 Dunn, J (2007) Worried companies block Facebook Av a i l a b l e : h t t p : / / w w w . c o m p u t e r w o r l d . c o m / s / a r t i c l e / 9 0 3 1 8 4 0 / Worried_companies_block_Facebook Downloaded 21.09.10 Engeström, J 2003-2010 Blog Available: http://www.zengestrom.com/ Downloaded: 21.09.10 Gladwell, M. (2002) The Tipping Point: How Little Things Can Make a Difference. Back Bay Books. Godin Seth (2008) at Business of Software. 33

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Available: http://network.businessofsoftware.org/video/seth-godin-on-whymarketing-is Downloaded: 21.09.10 Groves, C (2005) Wilson, D. E., & Reeder, D. M ed. Mammal Spiecies of the World (3rd ed.). Johns Hopins University Press. ISBN 0-801-88221-4 Hillman, A 2010 Top 100 Twiterholics based on Followers in Sweden. Available: http://twitaholic.com/top100/followers/bylocation/Sweden/ Downloaded: 21.09.10 Katz, E, and Lazarsfeld, P. (1955) Personal Influence: The Part Played by People in the Flow of Mass Communications. New York: The Free Press (via Cha, Haddiadi, Benevenuto, Gummadi, 2010) Kullin, H 2010, Top 20 Swedish Brands on Twitter Available: http://www.kullin.net/2010/03/top-20-swedish-brands-on-twitter/ Downloaded: 21.09.10 Lidman (2010) Onlinestrategier ISBN13: 9789186539283 Recito Förlag Martin, T 2010. Why Social Media Experts Hate Being Called Experts Available: http://www.socialmediatoday.com/SMC/198798 Downloaded: 21.09.10 Moir 2010, Social Media Marketing Strategy Study: Twitter vs. Facebook Business Applications http://www.onlineprnews.com/news/30315-1271088215-social-mediamarketing-strategy-study-twitter-vs-facebook-business-applications.html Downloaded: 28.10.10 Naslund 2010, 5 Social Media Topics to Watch Available: http://www.radian6.com/blog/2010/07/5-social-media-topics-towatch/ Downloaded 03.12.10 Rogers, E. M. (1962) Diffusion of Innovations. Free Press. (via Cha, Haddiadi, Benevenuto, Gummadi, 2010) Saleem, M (2010) Visualizing 6 Years of Facebook 34

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Available: http://mashable.com/2010/02/10/facebook-growth-infographic/? utm_source=feedburner&utm_medium=feed&utm_campaign=Feed: +Mashable+ (Mashable) Downloaded 21.09.10 Silfwer Jerry (2010) Social media strategies for a wired PR world #eurocom Available: http://www.doktorspinn.se/2010/02/01/social-media-strategies-fora- wired-pr-world-eurocom/comment-page-1/ Downloaded: 21.09.10

Starker D. (2008) What is trust M.Sc., P.G.C.E., Dip.M., FRSA Available: http://changingminds.org/explanations/trust/what_is_trust.htm Downloaded 21.09.10 Studstill 2010, The Power of Immediate Feedback Available: http://www.psfk.com/2010/08/the-power-of-immediatefeedback.html Downloaded: 27.10.10 Sysomos 2010, How many tweets is too many? Available: http://blog.sysomos.com/2010/06/11/how-many-tweets-is-toomany/ Downloaded: 21.09.10 Telia #1: Möt oss på Twitter och Facebook (Meet us on Twitter and Facebook) Available: http://blogg.telia.se/battreaffarer/2010/08/31/mot-oss-pa-facebookoch-twitter/ Downloaded: 21.09.10 TV4 #1 : Följ TV4 på Twitter och Facebook (Follow TV4 on Twitter and Facebook) Av a i l a b l e : h t t p : / / w w w . t v 4 . s e / 1 . 1 5 8 5 5 7 1 / 2 0 1 0 / 0 4 / 0 7 / folj_tv4.se_pa_facebook_och_twitter Downloaded: 21.09.10 Wikipedia #1 (2010) List of Social Networking Websites Available: http://en.wikipedia.org/wiki/List_of_social_networking_websites Downloaded 21.09.10

35

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Wunsch-Vincent and Vickery, 2007, Working Party on the Information Economy Available: http://www.oecd.org/dataoecd/57/14/38393115.pdf Downloaded: 27.10.10

36

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Monitoring ChartLabs Onsite numbers on public version of rapport is confidential on request from Ericsson Labs.

37

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Appendix Appendix 1: List of people interviewed: Name

Position

Contact

Bergström Ulrika

Marketing Communications Manager

[email protected]

Brisius Richard

Director Marketing Communications

[email protected]

Fransén Tobias

Consultant at The Amazing Society

[email protected]

Lindqvist Kristin

Communications Director

[email protected]

Lundström Olof

Head of Ericsson LabsLabs China

[email protected]

Hyttfors Stefan

Consultant at Wenderfalck

[email protected]

Minde Tor-Björn

Head of Ericsson Labs

[email protected]

Montero Gerardo

Research Engineer at Ericsson

[email protected]

Ragndahl Peter

Consultant at The Amazing Society

[email protected]

Reuterswärd Terese

Digital Manager at Electrolux

[email protected]

Ronander Karin

Director, Internal Communications

[email protected]

Thunberg Marianne

Marketing Communications Manager

[email protected]

Appendix 2: RSS-feed list: Writer

Web-address

Title

Jerry Silfwer

www.doctorspinn.se

PR-consultant

Disruptive

www.disruptivemedia.se

Entrepeneurship, Web 2.0

Jesper Åström

jesperastrom.com

SEO, Social Web, Facebook tutorials, Google Analytics

Fredrik Stenbeck

sixx.se/nextgen

Next Generation Internet

Social Media Today, LLC

socialmediatoday.com

Tech Crunch

techcrunch.com

Whats hot, web 2.0

Mashable

mashable.com

social and digital media, technology and web culture

38

Master of Science Thesis - KTH 2010 - Mikael Palm - Social Media Strategy Ericsson Labs

Appendix 3: List of 17 KPIÊs measured:

Off-Site

On-Site Key Usage each week

Twitter Following

New key registered

Twitter Followers

Registered new members *total

Re-Tweets "RT EricssonLabs"

Visits to Homepage

Re-Tweets "Twitter" @ericssonLabs D ericssonLabs #tweets EricssonLabs Facebook # Fans "code for cash YouTube Channel views LinkedIn #group members Amplify followers Amplify sources Amplify Posts

39

TRITA-CSC-E 2010:148 ISRN-KTH/CSC/E--10/148--SE ISSN-1653-5715

www.kth.se