It was also the week when Twitter joined the fray.  The airline’s boss Willie Walsh criticised Unite joint leader Derek Simpson for tweeting live details of Saturday’s talks.  The row added to a wider debate on using the micro-blog in official meetings, and elsewhere, Coalition ministers have already been told to leave their BlackBerrys at the Cabinet office door.

So what’s happening?  And, what’s the real impact of the strike?  A bit of social media monitoring using Sentiment Metric’s system helped us to make sense of, and measure, the impact the strike was having on BA, Unite, and the airline’s passengers.  We looked at online news and social media.

We were able to look at buzz levels around the dispute, influencial sources in the discussion, what everyone was talking about and also look at whether comment on the strike was positive, neutral or negative.  We also added a human touch to our research to interpret and assess the valuable information generated by our system. 

Figure 1: Buzz volumes for the strike – 20/5 to 26/5

We monitored thousands of mentions, in real-time, across the BA discussion, and tailored our research to mentions of BA, Unite and the strike.  One of the first things we noticed was a spike in coverage on 23 May when buzz volumes peaked at 579 across all channels.

Our industry-leading custom reporting tool generated the top 75 lead topics for the strike over the reporting period.  It showed that the Unite union’s negotiations with BA to settle the dispute dominated coverage.  On 20 May, the High Court over-ruled an injunction that had stopped the planned walk-out by cabin crew.

And, the great thing about our system is we can track people, products, services, industry terms and over 300 types of other entities.  We are not limited to single word phrasing in the topic cloud.  So, as you can see in the chart below, we picked up topics such as ‘industrial dispute mediator’ and ‘High Court’ in the lead topics.

Interestingly, there wasn’t a mention of passengers in the lead topics but the effect of industrial action on the airline’s passengers was discussed.  Some human analysis across our system showed a focus on the disruption many passengers faced as the strike went ahead. (politics.co.uk 23/5)

Figure 2: Topic cloud for the strike – 20/5 to 26/5

Figure 3: Topic cloud for the strike on Twitter – 20/5 to 26/5

The custom reporting tool also enabled us to analyse the lead topics on any day, on any channel.  Two phrases on Twitter jumped out: the reference to a web campaign heating up and a ‘brutal web war’.

So, who was driving the discussion?

There was certainly a lot of noise on Twitter, which led the pack in terms of mentions with 49.8 per cent share of voice.  This was among the top ten influencial sources.  In terms of authority, our social media measurement system scored BBC News as highest with a 19.4 per cent share of voice.

Figure 4: Top influencial sources for the strike – by mentions – 20/5 to 26/5

Figure 5: Sentiment for the strike – 20/5 to 26/5

Over the reporting period, our system analysed over 1500 mentions of BA, Unite and the strike for sentiment.  The results across all channels were:

27.8 per cent of mentions were scored as positive
53.4 per cent of mentions were scored as neutral
18.8 per cent of mentions were scored as negative

There was positive sentiment to news that Unite won its appeal against a High-Court injunction on Thursday. (BBC News 20/5)

BA attracted positive sentiment on news that it had successfully run 65 per cent of 65 per cent of flights in May. (gather.com 23/5)

But, there was a negative tone to comments on BA’s balance sheet and there were many referencing to the fact that BA was a loss-making airline. For example, LA Times (25/5) 

A passenger post on Digital spy talked of the ‘buffoons’ at Unite taking at action at the ‘world’s worse’ airport – Heathrow. (Digital Spy 21/5)

We also noticed a spike in negative mentions on 23 May and we found that these were driven by syndicated comments from Unite joint general secretary Tony Woodley.  The union leader said there had been a “catastrophic breakdown” in relations as the deadline for the first strike neared.

Using the custom reporting tool, we could also drill down and look at sentiment on individual channels such as Twitter.

Figure 5: Sentiment for the strike on Twitter – 20/5 to 26/5

The spike in positive sentiment on 20 May was driven by chatter on Unite’s High Court victory.

The spike in negative coverage on 23 May was driven by chatter about the protestors that stormed the talks.

Talks on 26 May to end the next wave of strikes on 30 May were postponed until Friday. It is difficult to predict the final outcome of this bitter dispute. The question is will either side be declared a winner in the end?

He comments on the differences between automated and manual approaches, SentimentMetrics uses an automated approach and we are currently at an 80% accuracy which is considered good in the industry, however I sometimes get clients asking why something is negative when it should be positive which occasionaly happens.

Maybe we need to consider an human review stage in our service. The reason we haven’t so far is this would result in a delay in getting the data to the clients and also would inflate the cost of the service. At present our clients seem happy with the service but it is definately something to consider for the future.