The airline business is famed for its brutally competitive nature. Usually this means competing on price alone, and we see spiralling price wars at the expense of a quality customer experiences.
British Airways has always cut against that grain. The airline has remained solidly profitable off the back of a brand focused on a high quality customer experience.
BA's focus on customer experience is one reason we are still ecstatic to work together. The British Airways Holidays team ( who provide a high quality, low friction way for customers to book their hotels, cars, and experiences) know the value of knowing their customer in detail and of gaining insight quickly enough to take action on it.
“In less than 5 minutes, we are now able to understand the drivers of our advocacy from over 100k reviews”—Head of Customer Service
British Airways Holidays came to us with 100,000s of customer reviews across multiple platforms.
The team wanted to understand what was driving positive sentiment so they could double down on what was driving brand advocacy.
Before we arrived, the insights team would take a few hundred customer reviews as a sample. And would spend a few hours manually labelling each one with a topic- such as ease, quality, trust, and locations.
There are inherent problems with manual labelling. Firstly, the time taken limits the analysis to just a sample. Leaving the worry that feedback is anecdotal, non-representative or missing a glaring issue.
Secondly, manual labelling is biased by the analyst doing the tagging. If one employee leaves and another takes over, it's likely that they understand reviews differently, making the topic subjective. This makes it hard to track evolutions in topics from one period to the next.
Finally, the length of time taken for the manual work detracts from other priorities. Meaning that some months may have missing analysis, resulting in an unclear picture of brand drivers over time.
SentiSum uses Natural Language Processing (NLP) to identify topics in customer data.
Whether it's customer support chats, customer reviews or surveys, our platform understands the meaning and labels it with a topic.
The BA Holidays team had a good idea of the drivers of brand advocacy and wanted to track both volume and sentiment of the brand drivers.
Our platform rapidly analysed over 100,000 reviews left by BA customers. In less than 5 minutes, upon a quick check of the dashboard, the BA Holidays team can track how many reviews discuss each topic and which topic contributes the most to positive and negative customer sentiment.
The BA Holidays team saved manual time, gained an objectivity that was previously unattainable, and no longer had to take a small sample size. They now have a detailed understand across all reviews, and have removed an arduous task.
BA Holidays also built SentiSum's insights into their way of working. Being very accessible and trackable, the insights fitted well as performance goals which the team could check against regularly to measure success.
SentiSum's notifications and daily email digests make the distribution of insights easy. So teams across multiple company functions having the same understanding of what matters to their customers. CX can now be much more of a team effort at BA.
"[SentiSum] allow us to quickly identify very recent feedback that indicates a customer experienced unexpected work taking place at a hotel that has impacted their stay…”Head of Customer Service
Before BA Holidays began working with us, they would regularly sift through hundreds of reviews using a keyword search for the complaints they wanted to avoid.
For example, if a customer booked a hotel that had unexpected building work, it could dramatically impact their stay. Leading to an unhappy customer that will not book with BA Holidays again.
The BA Holidays team spent a lot of time and effort trying to find hotels providing a bad customer experience, then helping them solve those problems, removing the hotels from listings, or upgrading any customers in those hotels.
With SentiSum, BA Holidays could capitalise on their desire to provide high quality customer experiences.
Machine learning techniques identified when specific customer issues were mentioned in reviews, flagging them for action by the BA Holiday team.
If particular topics arose in the reviews, they could then look at the individual reviews and identify which hotels were contributing to negative customer sentiment. This meant that British Airways could proactively work with the hotels to prevent further negative experiences.
With the keyword search used previously, only 15% of reviews were actually relevant. But, with SentiSum's NLP 95% of the reviews flagged to the team contained valuable information for improving customer experience.
Sentisum's platform turned customer data into something actionable. Our analytics provided a focus and speed for the British Airways team they couldn't previously achieve.