I Analyzed 12,844 Tweets to Find How Top SaaS Companies Do Twitter Support

I Analyzed 12,844 Tweets to Find How Top SaaS Companies Do Twitter Support

PM Blogs

PM / PM Blogs 351 Views 0

I’ve coated our internal support process and analyzed SaaS support and success earlier than, however in relation to Twitter help, there’s not a lot knowledge on the market.

From what I’ve discovered on this research, I’d say that could possibly be as a result of it’s not one thing the corporate needs to do.

Is it a vital evil, caused by annoyed clients, or is it corporations assembly clients on their degree?

When you consider customers interacting with corporations on Twitter, you in all probability think about corporations making an attempt their greatest to settle down customers so indignant they should vent their frustrations publicly.

I obtained the thought for this research once I noticed the sheer quantity of injury management Outlook should do on Twitter due to their product, and got down to analyze different software program giants and the way they do Twitter help.

Since Twitter is totally open to the general public and may’t be edited, I assumed it was begging for a research on how SaaS corporations help their customers on Twitter. That’s for a few causes. One, as a result of nobody has revealed knowledge or written a submit like this earlier than. And two, as a result of I take pleasure in writing research on SaaS corporations, like this and this on pricing pages, and this on touchdown pages.

Right here’s my research into the effectiveness and makes use of of Twitter help at SaaS corporations.

Key findings:

  • Tweets from customers directed at help accounts are usually destructive/impartial in sentiment, however a better proportion of destructive tweets from customers typically correlates to a extra lively help account.
  • Goal (e.g. problem-solving) responses make up 15% of tweets to customers.
  • Damaging (e.g. apologetic, injury management) responses make up 15% of tweets to customers.
  • Constructive (e.g. grateful, useful) responses make up 23% of tweets to customers.
  • Impartial (e.g. requests for extra info) make up 62% of tweets to customers.
  • Probably the most apologetic firm is LinkedIn, which additionally receives probably the most adverse incoming communication and finally ends up offering the bottom quantity of goal help out of all analyzed corporations.
  • The least apologetic firm is Dropbox, which solely apologized 1.eight% of the time regardless of receiving the 2nd highest quantity of unfavourable tweets and tweeting extra subjective (much less useful) messages to customers.
  • Probably the most useful firm is Constant Contact, which supplies the very best proportion of goal (problem-solving) tweets.
  • The least useful firm is LinkedIn, which tweets each the 2nd highest proportion of negative-subjective (injury management) tweets and the bottom proportion of goal (problem-solving) tweets.

Methodology:

This research attracts from the Montclare SaaS 250 to seek out the highest 20 SaaS corporations and their Twitter help feeds. For corporations with no devoted Twitter account but in addition providing help on its basic Twitter, I used the overall Twitter for knowledge however filtered solely direct @point out tweets, which indicated they have been providing help on to a consumer. If the corporate solely broadcast from their account and by no means talked on to consumer I counted them as not having a help Twitter account.

Utilizing the t command line tool, I exported 1,000 incoming and outgoing tweets per firm (restricted by some accounts which had tweeted lower than the entire occasions) and loaded the info into Google Sheets.

Utilizing Aylien’s Text API, I did sentiment evaluation on all 12,833 tweets, which consists of calculating polarity (constructive, damaging, or impartial sentiment) and objectivity. I loaded the info into Airtable as a result of it’s easy to manipulate data, and I wanted views to divide tweets by sentiment, objectivity, and a mix of the 2.

To show probably the most often used phrases and phrases, I used Online-Utility.org’s Text Analyzer, which is a free on-line software.

Abstract of the highest 20 SaaS corporations and their Twitter help presence

Right here’s a fast breakdown of the highest 20 SaaS corporations, and their help exercise on Twitter:

(Zoom in)

Right here’s what you'll be able to study from this knowledge:

  • 13 out of the highest 20 SaaS corporations use Twitter as a help channel.
  • If an organization gives help on Twitter, they’ve been lively for round 5 years.
  • Throughout these 5 years, the typical SaaS firm analyzed tweeted 36,463 occasions to clients.
  • As a common rule, pure help accounts don’t comply with many customers. The typical is three,826.
  • Help accounts have a mean of 18,290 followers.
  • On common, help accounts tweet 16 occasions per day. The 2 most lively accounts — LinkedIn and Adobe — tweet 59 and 65 occasions per day respectively.

Since solely 65% of the highest 20 SaaS corporations supply Twitter help, is it truthful to say that it’s extra of a luxurious than a necessity? What do the businesses that use Twitter for help actually use it for?

What can the prevalence of Twitter help inform us about how and why it’s used?

For this research, I analyzed each incoming and outgoing tweets, which means the info set consists of tweets at the corporate, and tweets from the corporate.

I discovered that the sentiment of incoming tweets was, general, unfavourable/impartial:

What can we theorize the aim of Twitter as a help channel is?

  1. A strategy to help customers preferring Twitter over e mail or researching the corporate’s ordinary help channels
  2. A approach to have a devoted account to answer tweets from customers that require help, and direct these customers away from the primary account
  3. A way of injury management when coping with annoyed customers

Judging by the share of damaging incoming tweets to the analyzed SaaS corporations, it’s smart to conclude that injury management performs a serious position. A excessive proportion of impartial tweets signifies that questions (with out constructive or adverse sentiment) are sometimes tweeted at help accounts.

Primarily, the info says that customers use Twitter to complain, and to ask questions.

Let’s take a look at the breakdown of how corporations use their Twitter help channels, and what meaning.

The which means of constructive, adverse, and impartial sentiment in Twitter help

When analyzing the outgoing tweets, I checked out two most important knowledge factors: subjectivity, and sentiment.

Sentiment is the tone or angle of the tweet, and is calculated by Aylien. Aylien categorizes the phrases used and works out the context to find out whether or not the tweet is constructive, damaging, or impartial.

Destructive sentiment

Outgoing tweets with unfavourable sentiment are apologies and requests to take the dialog to direct messages or e-mail. They're principally responses to annoyed clients.

The extra adverse sentiment responses an organization makes, the extra ‘injury management’ they’re doing.

The most typical phrases and phrases utilized in destructive tweets to customers are:

To summarize the which means of adverse sentiment:

  • To apologize to the consumer
  • To get the consumer to take their points away from the general public discussion board of Twitter, and into personal messages

Constructive sentiment

Outgoing tweets with constructive sentiment are grateful. They're principally responses to acknowledge that a buyer’s drawback has been solved.

The extra constructive sentiment responses an organization makes, the extra possible they're to be fixing consumer’s issues immediately on Twitter as an alternative of pushing the consumer to direct messages/e mail.

The most typical phrases utilized in constructive tweets to customers are:

To summarize the which means of constructive sentiment: 

  • To thank the consumer for his or her suggestions
  • To thank the consumer for his or her query / bug submission
  • To acknowledge that the consumer’s drawback has been solved

Impartial sentiment

Outgoing tweets with impartial sentiment are sometimes requests. They're often direct responses to a buyer initially reporting a problem.

Impartial sentiment is usually indicative of a request for extra info, however with out an apology or thanks.

The most typical phrases in impartial tweets are:

To summarize the which means of impartial tweets:

  • Requests to direct message
  • Requests for extra info
  • No apologies or thanks

Dissecting the stability between subjectivity and sentiment

Because the knowledge’s out there and gives a deeper perception into the best way SaaS corporations help customers by way of Twitter, it’s useful to verify tweets towards mixtures of subjectivity and sentiment.

An goal tweet is usually a useful one, giving the consumer directions on methods to clear up their difficulty.

A subjective tweet is usually an apology, acknowledgement, request for extra info, or the corporate thanking the consumer.

Going deeper than simply pure objectivity or subjectivity, probably the most helpful classes are:

Subjective/Constructive: an acknowledgement of the issue being solved

From the info set I analyzed, I concluded that it's truthful to attract a correlation between the quantity of subjective/constructive tweets and the success price of that firm’s help efforts over Twitter. That’s as a result of they’re often indicative of the corporate rounding up the dialog and expressing their happiness to assist.

Listed here are two instance tweets with subjective/constructive scores:


Comments