Samsung and Volkswagen Crisis Communication

in Facebook and Twitter

A Comparative Study

Boyang Zhang

1,2

, Jari Veijalainen

and Denis Kotkov

Faculty of Information Technology, FI-40014 Unversity of Jyvaskyla, Jyvaskyla, Finland

Laboratory of Inductrial and Information Management, FI-33101 Tampere University of Technology, Tampere, Finland

Keywords: Crisis Communication Strategies, Volkswagen Emission Crisis, Samsung Galaxy Note 7 Crisis, Samsung

Washing Machine Crisis, Twitter, Facebook, Sentiment Analysis, SCCT.

Abstract: Since September 2015 at least two major crises have emerged where major industrial companies producing

consumer products have been involved. In September 2015 diesel cars manufactured by Volkswagen turned

out to be equipped with cheating software that caused NO2 and other emission values to be reduced to

acceptable levels while tested from the real, unacceptable values in normal use. In August 2016 reports

began to appear that the battery of a new smart phone produced by Samsung, Galaxy Note7, could begin to

burn, or even explode, while the device was on. In Nov. 2016 also 34 washing machine models were

reported to have caused damages due to disintegration. In all cases, the companies have experienced

substantial financial losses, their shares have lost value, and their reputation has suffered among consumers

and other stakeholders. In this paper, we study the commonalities and differences in the crisis management

strategies of the companies, mostly concentrating on the crisis communication aspects. We draw on

Situational Crisis Communication Theory (SCCT). The communication behaviour of the companies and

various stakeholders during crisis is performed by investigating the official web sites of the companies and

communication in Twitter and Facebook on their own accounts. We also collected streaming data from

Twitter where Samsung and the troubled smart phone or washing machines were mentioned. For VW we

also collected streaming data where the emission scandal or its ramifications were mentioned and performed

several analyses, including sentiment analysis.

1 INTRODUCTION

Since September 2015 at least two major crises have

emerged where major industrial companies

producing consumer products have been involved.

Volkswagen crisis became public on Sept. 18, 2015

when US Environment Protection Agency (EPA)

published its findings concerning cheating software

in certain VW diesel engines (for instance, of type

EA189) and required actions from Volkswagen

Group in the USA to rectify the situation. The

financial losses to Volkswagen due to the crisis have

been estimated to 18,2 bn euro and it will cut 30000

jobs (Times, 2016). It also withdraws from WRC

class (WRC, 2016) to cut costs, after it has won

world championships several times. The damages to

the reputation of the company have been substantial.

We have described the main events until June 2016

in (Zhang et al., 2016), where we also analysed the

crisis communication of the company in social

media around the issue.

Another somewhat similar crisis broke in

August-September 2016, when it turned out that the

new smartphone model Galaxy Note7 manufactured

by Samsung might catch fire or even explode

because of battery problems. This happened in a few

days and weeks after the model was introduced to

the market in the USA and elsewhere in August-

September 2016. In Oct. 2016, Samsung stopped

producing the troubled phone model and ran recall

programs in the markets where the phone had been

sold. Some estimates state that financial losses due

to the failed phone could be as high as 10 billion US

dollars (Mullen and Thompson, 2016). Damages to

the reputation of the company have been estimated

to be high as well.

312

Zhang, B., Veijalainen, J. and Kotkov, D.

Samsung and Volkswagen Crisis Communication in Facebook and Twitter - A Comparative Study.

DOI: 10.5220/0006301403120323

In Proceedings of the 13th International Conference on Web Information Systems and Technologies (WEBIST 2017), pages 312-323

ISBN: 978-989-758-246-2

In the social media era, organizations encounter

several challenges from public relations and

societies. In the presence of social media, continuous

monitoring is necessary, as social media has a

significant influence on the reputation and even

survival of organizations. This research provides

insights on the crisis communication strategies of the

two companies, VW and Samsung, as observed in

Twitter and Facebook communications on the

accounts controlled by them and on other accounts

controlled by ordinary users, various media and

other stakeholders.

In this paper, we answer the following research

questions:

1. What were the crisis response strategies of VW

compared to Samsung?

1.1 What kind of compensation measures did the

companies perform?

1.2 How quickly did companies reply to their

customers’ messages online?

1.3. What crisis communication strategies did these

companies possibly use?

2. What kind of customer feedback did companies

want to conceal (in order to protect their reputation)?

How many messages have they removed from their

FB accounts?

2 THEORETICAL FRAMEWORK

The scholarly discussion on crisis communication

has paid considerable attention to focal

organizations. In general, response strategies are

built based on various crisis types. According to

Meng (1992) a trigger event related to performance

in public relations, for example caused by unethical

action, can function as a catalyst to activate a crisis.

CSR challenges are crises when expectations

created are not delivered by an organization and can

thus result in a reputational crisis (Friederike Schultz

and Stefan Wehmeier, 2010; Timothy Coombs and

Sherry Holladay, 2015). The concept of para-crisis,

a crisis primarily existing in web discussions,

illustrates new ways of crises to manifest themselves

fast, making lack of responsibility in organizational

behavior widely known among the public (Coombs

and Holladay, 2012).

The Situational Crisis Communication Theory

(SCCT) (Coombs, 2007) is utilized in the sequel

while seeking answers to the research questions.

In order to accommodate various crisis types,

Situational Crisis Communication Theory (SCCT)

was developed by Coombs (2007) as a mechanism

to guide the response strategies. Its components are

shown in Figure 1. As an evidence-based framework,

SCCT theory is supposed to provide specific

response strategies in various crisis situations.

Figure 1: Components of SCCT theory (Coombs, 2007).

Organizational reputation refers to the reputation of

the organisation among the stakeholders and its

maintenance during a crisis. According to (Coombs,

2007), three factors in the crisis situation shape the

threat to the reputation of the organization. These are

initial Crisis responsibility (attribution), Crisis

history, and Prior relational reputation. Crisis

responsibility concerns the question, to which extent

the focal organization can be deemed responsible for

the crisis and possible (lethal) injuries for the people,

losses of property, product failures, etc. Crisis

Response Strategies refer to the strategies of the

focal organization that it can apply during the crisis.

An important part are of this component are the

crisis communication strategies towards

stakeholders, including employees, customers,

authorities, traditional media and social media

audiences. Other means include e.g. financial

compensations for customers who have experienced

financial losses, injuries, etc. Providing

compensation in the case of product or services

failure indicates positive intentions of the focal

organization, whereas apology uttered towards

stakeholders show that the organization is taking

responsibility for its actions and failures.

Emotions refer to the emotions of stakeholders

raised by the crisis and Behavioural intentions refer

to the behaviour of stakeholders directly or

indirectly towards the focal organisation. The arrows

in Fig.1 describe causal and other relationships

between the components (see (Coombs, 2007) for

further information).

Samsung and Volkswagen Crisis Communication in Facebook and Twitter - A Comparative Study

313

An organisation can be involved in a crisis at

least in three major ways: it can be a victim (victim

cluster in (Coombs, 2007)); be involved in various

unintentional accidents (accidental cluster), and be

involved in events that could have been prevented

by the organisation (preventable cluster). A typical

instance of the first cluster is an earthquake or

terrorist attack that damages organization’s

infrastructure. Technical-error accidents are typical

instances of the unintentional accidents. In the last

cluster an organization knowingly placed people at

risk, took inappropriate actions or violated

law/regulation. Volkswagen case clearly belongs to

this cluster, whereas Samsung’s cases would belong

more to the accidental cluster.

A crisis evolves during a certain period of time.

(Dougall, 2008), in agreement with Meng (1992)

mentions five crisis stages: early; emerging; current;

crisis; dormant. Stages follow each other in time

with growing intensity until the crisis falls into a

dormant stage. Simultaneously, issues become less

controllable by the focal organizations. Social media

provides a fertile ground for the spread of issues.

The process can be regarded as an issue contagion

process (Coombs, 2002). Therefore, social media

can fast turn an issue from an early to crisis stage

with a simultaneous high media attention. From the

organizational perspective, before an issues reaches

a crisis stage, response strategies could be utilized to

prevent the further spread of an issue (Dougall,

2008). This activity can also be called preventing the

crisis. An example might be that the sales of a

product or service are halted. The strategies applied

before an issue turns into a crisis can prevent it from

reaching the full crisis stage and thus decrease

reputational and financial losses. Alternatively,

neglecting para-crisis could cause giant financial

losses and require longtime recovery.

3 CASE STUDY DESCRIPTIONS

In this section, we compare the two crises and crisis

(communication) strategies the companies used. To

answer our research questions, we provide analysis

of the content the companies and stakeholders

published in and removed from the social media, in

our case Twitter and Facebook platforms.

3.1 Case Description: Samsung Galaxy

Note7 and Washing Machine Crisis

The global sales of Samsung Galaxy Note7

smartphones started in August 2016. Soon after the

devices were taken into use by customers, there were

multiple reports that the phone had overheated,

caught fire, or even exploded in August-September

2016. The company admitted the problem on Sept.

2, 2016. Samsung first explained that the defect

batteries were the cause for this problem. The first

solution was to replace the batteries leading to a

recall of 2.6 million Note7 smartphones. The formal

recall announcement in the US was issued on Sept.

15, 2016. After the recall and battery replacement

the same problem still persisted. The company

decided to stop manufacturing and selling Note7 on

Oct. 11, 2016. The phone was made non-operational

on Dec. 19, 2016 (Wiggers, 2016) by a software

update. The recall crisis has exerted fluctuations in

the stock market. Samsung’s shares lost $18 bn

value in October 2016. The public begins to pay

attention to how Samsung resolves the quality

control problems and crisis responses. Table 1 below

contains the main phases of the Samsung Galaxy

Note7 crisis.

Another problem, this time with the Samsung

washing machines, also emerged during autumn

2016. On Nov. 4, 2016 the company announced a

recall of 2.8 million machines that vibrated strongly

and partially disintegrated. 34 washing machine

models have been affected (Stieg, 2016). On Nov. 7,

2016 Samsung published apologizes in major US

newspapers (Washington Post, the New York Times

and the Wall Street Journal) for the troubled Note7

and the malfunctioning washing machines (Herald,

2016). The apology was signed by Gregory Lee, the

president and CEO of Samsung Electronics North

America. In the text Samsung promises to find out

what went wrong with Note7 and the washing

machines and publish the findings. For Note7 the

findings were published in January 2017. The design

of the batteries and their fitting have evidently been

the source of the problem (Pierre, 2017), whereas

the phone design has otherwise been intact.

3.2 Case Description: Volkswagen

The Volkswagen emission scandal has been

followed in Zhang, Veijalainen and Kotkov (2016)

from Sept. 2015 to June 2016. In this context we

present results based on a continued data collection

from Twitter and Facebook and also present

sentiment analysis results. There is a detailed

chronology of events (in German) until the

beginning of Nov. 2016 at (NDR, 2016a). We leave

out the details for brevity.

Traditional media has played an important role in

both, but social media has also spread correct and

WEBIST 2017 - 13th International Conference on Web Information Systems and Technologies

314

not-so-correct information. In the sequel we will

mostly concentrate on the crisis communication

strategy of the companies at Twitter and Facebook.

Table 1: Samsung Galaxy Note7/Washing machine crises

story line.

Date

Issues

2 Aug

2016

The launch of Samsung Galaxy Note7 at a

media event in New York

19 Aug

The release of the Samsung Galaxy Note7 in 10

markets including U.S. and South Korea.

31 Aug

Shipment delayed by Samsung with quality

control problems.

1 Sep

Media reports reported Samsung was going to

launch the recall, and the starting of sales in

China.

9 Sep

Consumer Product Safety Commission of U.S

suggested to customers stop using Note7

15 Sep

Consumer Product Safety Commission of U.S.

officially publish the info about 1 million Note7

recall.

19 Sep

In Chinese market, Samsung announced that the

cause of the fire was external heating.

29 Sep

Samsung announced that over 1 million users

worldwide were using Note7 with safe battery.

1 Oct

The resuming of sales on new Note7s in South

Korea.

6 Oct

An U.S aeroplane was evacuated because of

smoke from Note7 on board.

10 Oct

Samsung adjusted the shipment of Note7 for

inspection.

11 Oct

Samsung permanently ended production and

sales of Note7 after it failed to fix the problems

on defective devices.

14 Oct

The department of transportation and Federal

Aviation Administration of U.S. officially

banned Note7 from all U.S. flights.

18 Oct

Samsung launched a roll out airport exchange

program to allow Galaxy Note7 owners to swap

their phones before a flight.

27 Oct

Samsung’s investigation states that it is

essentially a quality control problem.

29 Oct

Samsung held a media conference in China and

let the personnel kneel down to show the

gratitude.

4 Nov

Nearly 85% of the Note7 phone had been

replaced through the exchange program in U.S.

4 Nov.

Samsung announced a recall of 2.8 million

washing machines

7 Nov

Samsung published apologizes in major US

newspapers (Washington Post, the New York

Times and the Wall Street Journal) for the

troubled Note7 phone and the malfunctioning

washing machines

19 Dec.

Note7 phones will be inactivated

23 Jan.

2017

A report is released detailing reasons for Note7

problems

4 TWITTER AND FB DATA

COLLECTIONS AND

ANALYSIS

To answer research questions, we collected data

related to both crises from Twitter and Facebook.

We also applied sentiment analysis to the collected

data and analysed the obtained results. We also

studied traditional media analyses to answer the

research questions.

4.1 Twitter Data Collections and

Analysis for Samsung

From 15 September to 8 November 2016, there are

1081212 tweets collected which are related to

Samsung crises, the selection attribute for the

language is ‘en’, i.e. English. . Our data set

contained tweets also in other European and Asian

languages, but we decided to concentrate on the

English tweets. Based on the collection, sentiment

and topic analysis of the tweets is built with R-studio.

Figure 2: Sentiment analysis of collected tweets for

Samsung.

Another Twitter dataset was collected on Sept. 24,

2016 and on Nov. 11, 2016. The target accounts

where verified Twitter accounts controlled by

Samsung. These included @SamsungMobileUS,

@SamsungUS, @SamsungMobile, @SamsungUK,

@SamsungMobileME that tweet mostly or

exclusively in English, but there were also accounts

tweeting in other languages, like Korean. The

collection consisted of about 21000 tweets. The

oldest was sent on July 20, 2012 and the latest on

Nov. 10, 2016. About 12000 tweets were replies to

earlier tweets of customers. Among those 2617

20000

40000

60000

80000

100000

120000

negative neutral positive

Samsung and Volkswagen Crisis Communication in Facebook and Twitter - A Comparative Study

315

tweets were generated by a Samsung controlled

account above as a reply in English since Aug. 1,

2016. We could collect 1618 of the original tweets

Samsung had replied to between Aug.1, 2016 and

Nov. 10, 2016. This means that we could recover

only about 62% of the latter tweets sent to some

Samsung verified account. Those 1618 tweets with

both the tweet text and the reply to it by Samsung

along the timestamps and screennames are of

interest in this context.

The sentiment analysis reveals that majority of

all tweets from customers are positive during August

2016. If there are complaints, they mostly concern

the delays in delivery of Note7, or problems in some

other phone models. On Sept. 2, customers begin to

send more frequently tweets about global recall,

especially to SamsungMobileME (i.e. Samsung

Middle-East) and later to SamsungMobileUS etc.,

Samsung accounts begin to refer to the official

statement of Samsung at

https://news.samsung.com/global/statement-on-

galaxy-note7. The tone of the customers is mostly

questioning, “What should I do; when will a

replacement come? “etc. Starting Sept. 2, almost all

tweets concern Note7 recall issues for several weeks

during Sept. 2016. Some tweeters are impatient and

blame Samsung for bad service.

Looking closer, the first tweet concerning the

overheating of Note7 in our data set is from the user

linarhujer from 2016-08-26 20:29:53 and it reads:

'@SamsungUS @SamsungMobileME

@Samsung_Saudi what about the explosion of the

note7 when we charge it?

Please answer https://t.co/rtA96hXCXL'

2 days, 13 hours and 17 minutes later

SamsungMobileME replies:

'@linarhujer @SamsungUS @Samsung_Saudi Hello!

Please send us your contact information immediately.

We hope you are alright!'

Looking at the overall picture of the tweeting

behaviour of the company, the accounts below do

not directly help customers, but relay them to special

technical support or similar departments through a

direct message request. SamsungMobileUS has

replied to customers almost always in 24 hours,

average response time being about 5½ hours. Other

accounts have been slower, except SamsungMobile,

but the sample (6 tweets) is very small.

Table 2: Samsung reply activity in Twitter 1.8.-

10.11.2016.

Account of Samsung

Nr. of

pairs

Percentage

of replies in

24 hours

Avg.

resp.

time

SamsungMobileUS

980

5h26’

SamsungUS,

11h55’

SamsungMobile

100

0h18’

SamsungUK

410

1d20h

SamsungMobileME

136

1d9h

Concerning the streaming data set Figure 2, shows

that positive sentiments have prevailed every day.

There is no exception in this collection. As concerns

the sentiment of the customer tweets in Table 2, they

follow similar pattern; positive or neutral sentiment

is stronger than negative one.

When comparing the initial phases of the crisis

of VW in Sept. 2015 and Samsung in Sept. 2016,

Zhang et al. (2016), there were fewer tweets with a

positive sentiment in VW’s case. This raises the

question how response strategies are adapted to

different crisis situations.

In the context of Note7 recall in the U.S market,

Samsung has launched a compensation plan (http://

www.samsung.com/us/note7recall/) which contains

the refund and exchange program. Former Note7

users can choose between several options such as,

dollars bill credit under different conditions.

In the Chinese market, the recall plan was

different in comparison to US market (http://

www.samsung.com/cn/news/product-/galaxy-note-7

-recall-details). It offered less cash compensation,

blanked by the media announcement that the burning

cases were caused by external heating

(http://www.techtimes.com/articles/178492/2016092

0/samsung-note-7-explosions-caused-external-

heating-battery-china.htm). It had been interpreted

by local media as an irresponsible explanation.

Irresponsible behavior without effective measures of

the focal organization can turn an almost crisis into a

real crisis which causes financial and reputational

losses.

On Oct. 11, 2016, Samsung had ended the

production and sales of Note7 permanently after

failing to fix the problems that caused the

overheating and even fire or explosion of the

defective devices. The devices still in the hands of

customers were planned to become inactive and

useless on Dec. 19, 2016, after the last software

update.

WEBIST 2017 - 13th International Conference on Web Information Systems and Technologies

316

4.2 Twitter Data Collections and

Analysis for VW

There were three different data collections for VW

case. First, the tweets sent from VW controlled

(verified) Twitter accounts were collected as far as it

was possible on Nov. 15, 2015. We found 22

different accounts of this kind. We could collect

circa 18000 English tweets sent from these accounts

since Sept. 18, 2015. About 10400 of these were

replies to earlier tweets. We also collected these

earlier English tweets that were replied to by some

verified VW account during the above period. We

could not collect them all, but only about 6800

tweets, i.e. under 70 %. The rest were either

removed or protected so that they could not be

collected through the REST API of Twitter.

As concerns the replied-to tweets, a scan through

them reveals that the overwhelming majority

consists of complaints regarding bad service or bad

quality of the cars, or both. Roughly 50% of the

complaints are directed towards @VolkswagenIndia,

but also @VolkswagenSA gets similar complaints.

@UKVolkwagen gets more various questions. Only

a few tweets mention the emission scandal. The first

instance is a side remark in a tweet in Sept. 2015.

Later, during 2016 the recall arrangements are

mentioned more often. As an overview one can state

that English tweets sent towards verified accounts of

VW deal with technical or other problems of

individual customers with their cars and the services

offered by dealers and the sentiment is mostly

negative or at most neutral. This result was

established by scanning through the tweets manually

and reading their contents.

In addition, we collected streaming data since

June 23, 2016 with the Tweepy (www.tweepy.org)

predicate ['vw','VW', 'volkswagen', 'Volkswagen'

'scandal', 'reputation', 'dieselgate','vwgate',

'emission','fraud'] and later (since Nov. 17, 2016)

with ['vw','VW','volkswagen','audi', 'Audi',

'Volkswagen' 'scandal' 'reputation', 'dieselgate',

'vwgate' ,'emission']. The resulting raw data contains

90-95 % irrelevant data. The relevant data has been

selected by running a filter program after the tweets

have been parsed into database tables. This resulted

in about 360000 relevant tweets. These tweets can

be used to characterize the overall sentiment of a

wider public towards VW during the second half of

2016. This is, however, for further stydy.

We looked at the response times of VW in

Twitter. These tell about the attitude of the company

towards their customers concerns and are at the

same time an expression of the communication

strategy. There were 9627 English customer tweets

with a response from @VW (2321), UKVolkswagen

(2579), @volkswagenindia (2090), VWMiddleEast

(105), Volkswagen_NZ (11), VWcanada (288),

Volkswagen_MX (23), VolkswagenNL(15),

vwbelgique(5), VolkswagenSA (2053),

vw_france(5). The oldest reply was sent at 2015-08-

10 11:02:04 and newest at 2016-11-15 12:17:43.

Table 3 shows the response times to the

customer tweets by various accounts of VW. Over

50 per cent of the tweets are replied in 24 hours or

less, but the tail is rather long rendering average

response time of a month in most cases.

Table 3: Volkswagen’s reply activity in Twitter

10.8.2015-15.11.2016 (major English accounts).

Account

Nr. of

pairs

Percentage

of replies

in 24 hours

Avg.

resp.

time

2321

36 d

UKVolkswagen

2579

29 d

volkswagenindia

2090

25 d

VolkswagenSA

2053

108 d

VWcanada

288

29 d

VWMiddleEast

105

54 d

4.3 Facebook Data Collections and

Analysis

To answer research questions 1 and 2, we collected

posts, comments and replies from Facebook groups

dedicated to Samsung (SamsungMobileUSA, https://

www.facebook.com/SamsungMobileUSA) and

Volkswagen (Volkswagen, https://

www.facebook.com/VW/?brand_redir=DISABLE).

We collected data twice to detect posts, comments

and replies that had been removed from the

mentioned pages.

4.3.1 Samsung FB Data Description

Samsung owns several Facebook pages, such as

Samsung Support, Samsung Mobile and Samsung

US. We chose SamsungMobileUSA, as the page is

related to Note7 and users are very active on this

page compared to other pages. Table 4 summarizes

characteristics of two datasets regarding Samsung.

We finished our first data collection from the

Samsung page on September 23, 2016. This data

collection included posts, comments and replies

published from August 8 till September 22, 2016.

Our second data collection finished on November 2,

2016 and included content posted from August 8 till

October 31, 2016.

Samsung and Volkswagen Crisis Communication in Facebook and Twitter - A Comparative Study

317

Table 4: Samsung FB data collection summary.

Date of

data col-

lection

Nr. of

posts

Nr. of

com-

ments

Nr, of

replies

Nr. of

Samsung

replies

23.9.16

113

81614

37430

1448

02.11.16

45299

18931

983

Figure 3: Comments and replies on SamsungMobileUSA

Facebook page.

Figure 4: Replies of Samsung on SamsungMobileUSA

Facebook page.

Figures 3 and 4 demonstrate the numbers of

comments and replies posted at different dates at

SamsungMobileUSA Facebook page (the first data

collection). Replies in Figure 3 include customer

replies and Samsung replies, while replies in Figure

4 include only replies of the company. The first

peaks (around August 2) on both charts corresponds

the release of Note7, the consecutive peaks are

caused by discussions and complaints related to the

Note7 battery problem. The number of replies from

the company changes proportionally, but it is much

lower than the number of replies of the customers;

the more comments and replies overall the more

Samsung replies.

4.3.2 Volkswagen Data Description

Table 5 summarizes characteristics of two datasets

regarding VW. Our first data collection finished on

April 13, 2016 and included content published from

September 17, 2015 till April 12, 2016, while our

second collection finished on November 23, 2016

and included content published from September 17,

2015 till September 24, 2016.

Table 5: VW data collection summary.

Date of

the col-

lection

Nr. of

posts

Nr. of

cust. com-

ments

Nr. of

cust.

replies

Nr. of

replies

13.04.16

17357

9552

103

23.11.16

16723

8488

145

Figure 5: Comments and replies on VW Facebook page.

Figure 6: Comments and replies on VW Facebook page.

Similarly to the Samsung case, Figures 5 and 6

demonstrate numbers of comments and replies

posted at different dates in VW Facebook page (the

first data collection). The number of replies at each

date is proportional to the number of comments at

that date. However, the relative number of VW

2000

4000

6000

8000

02-08-2016

05-08-2016

08-08-2016

11-08-2016

14-08-2016

17-08-2016

20-08-2016

23-08-2016

26-08-2016

29-08-2016

01-09-2016

04-09-2016

07-09-2016

10-09-2016

13-09-2016

16-09-2016

19-09-2016

Comments Replies

100

200

300

400

02-08-…

05-08-…

08-08-…

11-08-…

14-08-…

17-08-…

20-08-…

23-08-…

26-08-…

29-08-…

01-09-…

04-09-…

07-09-…

10-09-…

13-09-…

16-09-…

19-09-…

Samsung Replies

1000

2000

3000

4000

5000

17-09-2015

01-10-2015

15-10-2015

29-10-2015

12-11-2015

26-11-2015

10-12-2015

24-12-2015

07-01-2016

21-01-2016

04-02-2016

18-02-2016

03-03-2016

17-03-2016

31-03-2016

Comments Replies

17-09-2015

01-10-2015

15-10-2015

29-10-2015

12-11-2015

26-11-2015

10-12-2015

24-12-2015

07-01-2016

21-01-2016

04-02-2016

18-02-2016

03-03-2016

17-03-2016

31-03-2016

VW Replies

WEBIST 2017 - 13th International Conference on Web Information Systems and Technologies

318

replies is much lower than that of Samsung. Among

all the collected replies on the VW page 1% (the

absolute value is 103) of replies belong to VW,

while for Samsung this value is 3% (the absolute

value is 1448).

4.3.3 Facebook Response Time Analysis

To answer research question 1.2, we investigated

how quickly companies responded to customers on

Facebook.

Figure 7: Response time distribution on Samsung

Facebook page.

Figure 7 demonstrates the distribution of delays

between customers’ comments and Samsung replies

to them (the second data collection). The distribution

has the following characteristics: mean=5.46 hours,

standard deviation=10.65 hours and median=1.63

hours. Most comments (544) are replied faster than

in 3.7 hours. The quickest reply was published in 22

seconds after the comment, while for the longest

reply this value is 5 days. This shows that Facebook

is considered as a more important channel towards

stakeholders than Twitter, where the responses take

longer to generate.

Figure 8: Response time distribution on VW Facebook

page.

Figure 8 demonstrates the distribution of delays

between customers’ comments and VW replies to

them (the second data collection). The distribution

has the following characteristics: mean=23.57 hours,

standard deviation=21.81 hours and median=27.0

hours. Most comments (55) are replied earlier than

in 27 days. The quickest reply was published in 2.7

hours after the comment, while for the longest reply

this value is 128 days.

The analysis of both distributions suggest that

VW seems to reply to customers’ comments slower

than Samsung due to the minimum, maximum

response times and response time form most

comments, considering that data collection in each

case was conducted in different periods of time after

the crisis. Our second data collection regarding

Samsung was conducted in two months after

Samsung recall, while for the VW this time is 1 year

and 1 month.

4.3.4 Facebook Removed Content Analysis

To answer research question 2, we conducted two

data collections for each company. We then

compared the collected content and detected

removed posts, comments and replies.

Table 6: Removed content.

Company

Nr. of

company

posts

Nr. of cust.

com-ments

Nr. of

cust.

replies

Nr. of

company

replies

Samsung

22 (19%)

52148

(64%)

8243

(22%)

773

(54%)

14 (35%)

2486

(14%)

2008

(23%)

12 (8%)

Table 6 summarizes numbers of removed posts,

comments and replies on Facebook pages of each

company. Posts can only be removed by the owner

of the Facebook page, while comments and replies

of customers could be removed by the owner of the

page or by the author of the content. All posts

removed by Samsung were about Note7, while some

posts removed by VW were related to special offers.

The preliminary analysis of the removed contents

indicates that there were some very negative

comments. Why the companies removed certain

posts and comments and replies is not clear. Perhaps

they have come to the conclusion that removing

them is better for their reputation?

4.3.5 Facebook and Twitter Topic Analysis

Table 7 contains the text modeling results from

100

200

300

400

112

Number of replies

Response time in hours

340

648

957

1266

1574

1883

2192

2500

2809

Number of replies

Response time in hours

Samsung and Volkswagen Crisis Communication in Facebook and Twitter - A Comparative Study

319

Facebook and Twitter collection. R-studio was used

to perform the text mining. Reply messages from

SamsungMobilUS were tested and extracted into

topics. The techniques are the same that were used

for the sentiment analysis.

Table 7: The most frequent topics.

Topic 1

Samsung

Topic 2

Samsung

Topic 1

Topic 2

Thanks

samsung

pretty

can

Galaxynot

galaxynote

check

volkswage

Smsngusnere

cal

support

point

please

Reach

exchange

danisordo

contact

Available

please

meeting

details

Can

check

warranty

will

Get

info

match

wrc

New

help

certainly

car

hey

half

satefy

Can

crankandpi

ston

new

Topic Samsung represents the replies of

SamsungMobileUS to users’ comments. Topic VW

represents several accounts that VW group has used

to reply to users. Because the word “galaxynot*” has

appeared both in topic 1 and 2 in the Samsung

stream collection, we assume that Samsung’s

response strategies were focused on the recall of

GalaxyNote7, whereas in the VW group, no specific

car series appears.

5 THE ABOVE CRISES IN THE

LIGHT OF SCCT

What is common to the crises, what different? First,

let us look at the responsibility aspects, i.e

attribution issues. Both companies can be blamed for

the faulty consumer products, so it is clear which

entity to blame for the problems. Volkswagen case

clearly belongs to the Preventable cluster, but

Samsung case belongs more clearly to the

Accidental cluster, although there are features that

would allow it to be classified also to the

Preventable cluster. With a more rigorous quality

control the problems with the shortcut batteries

could have been avoided or detected early enough.

One essential difference is that in the VW case

there was not an immediate safety hazard for the

drivers or owners, but rather, the emissions caused

an environmental hazard and indirectly possible

health problems for a wider population. The

products violated environment norms set in force by

the local authorities. In Samsung’ case, the

overheating, even exploding, phones posed a direct

health threat for the users and an indirect threat for,

for instance, fellow passengers on any flight with an

owner of a Note7. They also violated consumer

safety norms. The major difference is that VW has

deliberately designed and implemented the cheating

software into the car engines; it is not a mistake

some designers or production plants have made. In

Samsung’ case the company first blamed the battery

manufacturer for the overheating and explosions, but

it rather seems that there is a flaw in the phone

design (Mullen and Thompson, 2016). This is

certainly not a deliberate design flaw, but rather a

mistake and points to a flawed quality control inside

the company.

Whereas Samsung has admitted their failure to

meet high quality standards and apologized for the

troubles this has caused for the stakeholders,

especially for the consumers using the products,

Volkswagen’s strategy has been to deny the

problems as long as possible. As discussed in

(Zhang et al., 2016), the NOx emission cheating was

denied by the company almost a year, before the US

environment authorities made it public in Sept. 2015

and the company had to publicly admit the existence

of the cheating software and apologize for the harm

caused to the stakeholders. Later it has been

admitted by persons involved that the cheating

software development was started already in 2006

(NDR, 2016b). The resigned CEO of VW, Martin

Winterkorn, knew about the software evidently since

May 2014 (NDR, 2016b). Since spring 2016 there

have been reports that in some Audi gasoline and

diesel engines there is further cheating software;

some Audi models appear to have allowed the cars

to cheat in carbon-dioxide emissions when tested.

They press the real CO2 values lower in testing

situations, while they are much higher when the car

is in real operation on the road. The cheating

software was found in an Audi car during summer

2016 by California Air Resources Board (CARB).

The issue was published by CARB according

Washington Post (Smith, 2016) relying on Bild am

Sonntag report (in German) from Nov. 4, 2016. So it

seems that the overall strategy of VW is to admit

cheating only if caught. From the chronological

record of events in (NDR, 2016a), it is evident that

the highest management knew about the cheating

software for NOx at the end of August 2015, but

most probably much earlier.

The Samsung case shows similar but also

WEBIST 2017 - 13th International Conference on Web Information Systems and Technologies

320

different traits. They have claimed in China that the

problem with overheating comes from external

sources. They also launched Note7 in China, while

the recall was announced already in the US and

European markets. On the other hand, the defects of

the Note7 were evident for the consumers, while the

cheating software in VW and Audi cars was not at

all evident for ordinary car owners. Revealing its

existence required sophisticated emission test series.

In both cases various authorities have stepped in and

have played an important role in the development of

the crisis. In the VW case the US environment

protection authorities used their power. In Samsung

case the aviation safety authorities and airlines

prohibited to carry Note7 on board.

Both companies have suffered considerable

financial losses; Samsung perhaps 10 bn dollars,

VW 18 bn dollars. The duration of the crises might

be different. VW crisis still continues after 18

months, whereas Samsung crises might be over in 18

months. The financial losses will still have effect on

the companies for years. How long it takes for the

reputation to recover among consumers remains to

be seen.

6 CONCLUSIONS

The goal of this research is to analyse the crisis

strategies, especially communication strategies of

VW and Samsung during recent crises. The emission

fraud or cheating crisis of VW started on Sept. 18,

2015, when the US environment authorities made

public the existence of cheating software in certain

VW car engines that detected the test situation and

set the NOx emissions down to an accepted level.

The authorities had been in contact with the car

maker already since 2014, but the company had

mainly tried to deny that there is a general problem

and the issue was kept secret. Later in 2016 it has

admitted that the design of the cheating software had

begun already in 2006. During autumn 2015 further

car engines were found by US authorities to have the

same software installed as those first tested. During

the summer 2016 further software was detected by

US authorities that functioned in a similar way as the

earlier software. Thus, the crisis of VW has had new

sub-issues emerging from time to time and

corresponding peaks in the number of media reports

and social media messages. The crisis has cost VW

over 18 billion euro and losses in reputation. The

former is caused by fines the company has to pay to

the US government and other bodies, recall and

reparation costs of the vehicles, customer packages

and lost market share. VW shares have also lost

value on the stock markets which has made the share

owners angry against the management.

Looking at organizational tweets and replies we

attempted to explain what crisis communication

strategies were used in various situations. Issues

could turn into crises when activators occur, which

could be social media discourse, electronic word of

mouth, or other triggers. Before turning into a real

crisis, effective measures need to be taken to defend

and protect the reputation of the focal organization.

Otherwise, crises can exert an extremely negative

impact on organizations, which may cause financial

and reputational losses. In order to prevent the crises

to emerge, organizations should not underestimate

the power of online communications. Paying enough

attention to early stages of a crisis can safeguard the

functioning of a focal organization which requires

continuous monitoring of public opinions expressed

in social media.

In both these cases the crisis broke because of

deficiencies in the products and social and

traditional media spreading the information about

faulty products. Social media also seemed to follow

and spread the information.

Our hypothesis was that VW followed the crisis

communication strategy that nothing is admitted

before one is caught. In this case only authorities

could reveal the emission cheating. Thus, it is not to

be expected that the company would deny it towards

car owners or other stakeholders while interacting

with them at social media sites. They can only try to

comfort, repair and compensate in monetary terms

the damages for the stakeholders. VW has paid

compensations only to the US WV diesel car

owners, but not for other owners

Long response times to tweets or emails seem to

be a major issue in India, in addition to poor service

quality. These kinds of things are a problem for any

company and tend to cause customers to change to

another product.

In Samsung’s case the company admitted fast

that Note7 devices might be faulty. On the other

hand this was evident and it would not be viable to

try to deny the fact. At the same time also some

washing machine models had issues. The company

published in leading US newspapers an apology on

Nov. 7, 2016. VW also published a video on Sept.

22, 2015, where the CEO apologized for the

wrongdoing, but as far as we know, VW did not

publish apologies in newspapers.

Both companies have removed some posts from

their Facebook pages. Some comments and replies

were also removed. However, Facebook does not

Samsung and Volkswagen Crisis Communication in Facebook and Twitter - A Comparative Study

321

provide information on who removed these

comments and replies. All removed Samsung posts

were about Note7, while VW removed some posts

unrelated to their cars. These related to special

offers. Among removed comments and replies we

found some positive and negative messages. It

seems that the companies did not delete tweets from

their accounts, but our collection frequency was too

coarse in order to give a firm answer; a tweet might

have appeared and disappeared between collections

In both cases we could find positive, but also rather

negative tweet content towards the focal

organisation. On the other hand, the organisations

cannot remove the tweets sent by customers, only

the tweets they themselves sent, so this is

understandable.

The most severe issue for the companies are the

financial losses caused by the crisis. The cost for

VW will be perhaps 18 billion euros and for

Samsung 10 billion euros. VW withdraw from the

WRC class for the year 2017 and lay off 30000

people from its workforce. Perhaps the main positive

result is that VW promised to develop electric cars

and bring them to the market during the next 5-10

years. As far as we could observe, Samsung has not

announced a radically new direction in its consumer

product strategy.

In the future we would look more closely at the

possible relationship of the communication style,

and communication frequency of the companies at

various social media platforms towards the

stakeholders and the simultaneous financial

performance of the companies over time. Perhaps

there is positive or negative correlation. We could

find a small percentage of tweets sent to Samsung

where the customer wrote that he or she will not buy

anymore Samsung phones or any products. The

same is true for VW. Could this information be used

to predict the future sales in some product

categories? Another aspect is a more thorough

sentiment analysis and the relationship between the

social media user sentiment and financial

performance of the company.

ACKNOWLEDGEMENTS

Authors were supported in part by the Academy of

Finland, grant #268078 (MineSocMed).

REFERENCES

Coombs, W.T., 2007. Protecting Organization Reputations

during a Crisis: The Development and Application of

Situational Crisis Communication Theory. Corp.

Reput. Rev. 10, 163–176. doi:10.1057/

palgrave.crr.1550049.

Coombs, W.T., 2002. Assessing online issue threats: issue

contagions and their effect on issue prioritisation. J.

Public Aff. 2, 215–229. doi:10.1002/pa.115.

Coombs, W.T., Holladay, J.S., 2012. The paracrisis: The

challenges created by publicly managing crisis

prevention. Public Relat. Rev.

Coombs, W.T., Holladay, J.S., 2015 . CSR as crisis risk:

expanding how we conceptualize the relationship.

Corp. Commun. Int. J. 20, 144-162. doi:10.1108/CCIJ-

10-2013-0078

Dougall, E., 2008. Issues Management [WWW

Document]. Inst. Public Relat. URL

http://www.instituteforpr.org/issues-management/

(accessed 11.20.16).

Friederike Schultz, Stefan Wehmeier, 2010.

Institutionalization of corporate social responsibility

within corporate communications: Combining

institutional, sensemaking and communication

perspectives. Corp. Commun. Int. J. 15, 9–29.

doi:10.1108/13563281011016813.

Herald, T.K., 2016. Samsung runs apology ad over Galaxy

Note 7 recall [WWW Document]. URL http://

www.koreaherald.com/view.php?ud=2016110800011

6 (accessed 12.13.16).

Meng, M., 1992. Issue Life Cycle Has Five Stages. Public

Relat. J. 48, 23.

Mullen, J., Thompson, M., 2016. Samsung takes $10

billion hit to end Galaxy Note 7 fiasco. CNNTech.

NDR, 2016a. Die VW-Abgas-Affäre: Eine Chronologie

[WWW Document]. URL www.ndr.de/nachrichten/

niedersachsen/braunschweig_harz_goettingen/Die-

VW-Abgas-Affaere-eine-

Chronologie,volkswagen892.html (accessed 12.13.16).

NDR, 2016b. Hat Winterkorn Diesel-Problem richtig

eingeschätzt? [WWW Document]. URL

http://www.ndr.de/nachrichten/niedersachsen/braunsch

weig_harz_goettingen/VW-Skandal-Winterkorn-im-

Mai-2014-informiert,vw2674.html (accessed

12.13.16).

Pierre, 2017. Revealed: What went wrong with the

Samsung Galaxy Note7 [WWW Document]. URL

http://www.gizchina.com/2017/01/23/samsung-note-

7s-investigation-is-over-heres-what-happened/

(accessed 2.21.17).

Smith, C., 2016. Samsung formally apologizes for Galaxy

Note 7 and washing machine recalls. BGR.

Stieg, C., 2016. UPDATE: Millions of Samsung Washing

Machines Have Been Recalled [WWW Document].

Good Housekeep. URL http://

www.goodhousekeeping.com/appliances/washer-

reviews/news/a40642/samsung-washing-machines-

explode/ (accessed 12.13.16).

WEBIST 2017 - 13th International Conference on Web Information Systems and Technologies

322

Times, Fi., 2016. VW emissions scandal. Financ. TImes.

URL: https://www.ft.com/vw-emissions-scandal

(accessed 13.11.2016).

Timothy Coombs, Sherry Holladay, 2015. CSR as crisis

risk: expanding how we conceptualize the relationship.

Corp. Commun. Int. J. 20, 144–162.

doi:10.1108/CCIJ-10-2013-0078.

Wiggers, K., 2016. End of the line: Galaxy Note 7 to be

disabled in the U.S. starting December 19 [WWW

Document]. Digit. Trends. URL

http://www.digitaltrends.com/mobile/samsung-halts-

galaxy-note-7-shipments-phones-catching-fire/

(accessed 13.12.2016).

WRC, 2016. VOLKSWAGEN PULLS OUT OF WRC-

URL: http://www.wrc.com/en/wrc/news/october-2016/

vw-withdraws/page/4046--12-12-.html (accessed

25.2.2017).

Zhang, B., Veijalainen, J., Kotkov, D., 2016. Volkswagen

Emission Crisis – Managing Stakeholder Relations on

the Web: SCITEPRESS - Science and and Technology

Publications, pp. 176–187. doi:10.5220/

0005892401760187.

Samsung and Volkswagen Crisis Communication in Facebook and Twitter - A Comparative Study

323