We’ve written about crisis communication quite a bit over the years, especially how a comprehensive social listening program can help identify issues before they become bigger problems. Now that we’ve added sentiment analysis to our Social Suite, we wanted to explore exactly what sentiment data can do for crisis communication across all stages.
Got a question or something we missed? Find us on Twitter @UnionMetrics.
Contextualize a crisis
The first thing sentiment analysis can do is help you contextualize a crisis: Was it a lot of negative tweets or fewer, very angry and prolific people? When you start a new Tracker, it will backfill with sentiment data about 30 days. So while we can’t at this point learn from an older crisis with sentiment specifically (you can use keywords and more in Echo), you can look back a month and try to contextualize a crisis that’s unfolding now or beginning to unfold.
Is there a moment that stands out that you could have reached out to someone and fixed a problem before it snowballed? Look for moments like that vs. an overall downward trend in sentiment that still most likely stemmed from something you can pinpoint: A particular polarizing tweet, a campaign that went in a new direction from what your audience is used to from you, maybe even a redesign for your brand’s visuals.
In our Social Suite, a quick glance at the top of the sentiment report in a Tracker will show you generally if there’s more positive or negative sentiment overall, mapped out over time as it changes. Then you can look more closely at the intensity breakdown, plus notable positive and negative tweets.
From there you can note important things: Was there a brand advocate defending you through the worst of it? How can you thank them? (We’ll discuss this more in the next section.) Does it make sense to reach out to some of the angrier tweeters once things have cooled off? If they’re willing to talk to you, it could go a long way toward understanding what went wrong, why they were so angry with your brand and how you can avoid similar situations in the future.
Identify the positive
If you’ve identified anyone who was defending your brand through a crisis, you want ask a few questions: Are they brand advocates you’ve already identified, or new ones? Are they current or former employees? How can you reach out and thank them in a way that makes sense for your brand and your relationship with them?
You may also want to think this through theoretically before a crisis ever happens and make it part of your crisis comms plan. Then- even if the situations don’t match up perfectly- you’ll already be a few steps ahead.
See the speed of sentiment shifting
If you’re using our analytics, you can analyze how fast or slow sentiment changed over the course of the crisis using these features in particular from our sentiment analysis:
- Change. The change percent shows you how much higher or lower the sentiment score is for the current time period relative to the previous time period. If your Tracker is new, this will show n/a until there’s data in the previous time period to show.
- Average daily score. This shows the average daily sentiment score for the time period you’re viewing. The chart shows the intensity breakdown on the right x-axis – red for negative posts, gray for neutral, and green for positive posts. The line shows the average daily sentiment score on the left x-axis.