Are dashboards no longer useful or relevant?
This is probably my fifth or sixth article relating to Disneyland Paris. You may wonder if they sponsor me. Not really, but having an 8-year-old girl, I hope they will invite us for a weekend. I’m kidding.
The evolution of OpenAI, I am talking about the technical ones and not the saga of Sam Altman’s dismissal, is achieved quickly. These evolutions also led to further development of the original Disneyland Paris dashboard by offering new possibilities. Some innovations have prompted me to try new ideas to speed up the customer experience for my clients. I briefly summarize them below:
1. The Customer Journey Overview: have ChatGPT Commenting Charts
The latest version of ChatGPT, ChatGPT-4-turbo, is equipped with advanced image recognition and chat completion capabilities. As a result, we can utilize the model to analyze the top section of the dashboard which includes the Customer Journey Mapping analysis in detail. Then, we can request it to compare the Key Performance Indicators (KPIs) and provide us with a comment.
Please take a look at the comment that ChatGPT automatically generated after a few seconds. It’s quite impressive!
“Comment on star rating, sentiment, and customer journey phase
- Pre-Visit Stage:Sentiment: The majority is in the green zone (positive), followed by yellow (neutral), and then red (negative). This indicates a generally positive sentiment in the pre-visit stage.5-Star Rating: The average rating is 3.81 out of 5, which aligns with the largely positive sentiment but indicates room for improvement.
- Pre-Arrival Stage:Sentiment: There’s an increase in the yellow and red zones compared to the pre-visit stage, suggesting increased neutrality and negativity during preparations for the visit.5-Star Rating: The score dips to 3.19 out of 5, which is consistent with the increase in neutral and negative sentiments.
- Arrival Stage:Sentiment: Positive sentiments are dominant, but there’s still a substantial neutral and negative sentiment.5-Star Rating: The rating is 3.74 out of 5, which reflects the mixed but leaning positive sentiment.
- Park Experience Stage:Sentiment: Positive sentiment is quite dominant with a smaller proportion of neutral and negative sentiments.5-Star Rating: The score is 4.06 out of 5, reinforcing the positive experiences of visitors during their time in the park.
- Dining Stage:Sentiment: Positive sentiments still lead, but there’s a notable neutral and negative sentiment, possibly due to diverse dining experiences.5-Star Rating: A score of 3.80 out of 5 complements the mix of sentiments.
- Hotel Stay Stage:Sentiment: Predominantly positive, indicating high satisfaction with hotel experiences.5-Star Rating: A strong score of 4.38 out of 5 echoes the positive sentiment.
- Departure Stage:Sentiment: Again, majorly positive but with some neutral and negative feedback.5-Star Rating: The rating is 4.08 out of 5, suggesting that the overall departure process and reflections on the trip are positive.
- Branding:Sentiment: A mix, but with a substantial positive sentiment.5-Star Rating: Not explicitly given, but the presence of positive sentiment suggests a favorable view of Disneyland Paris branding.
- Post-Visit Stage:Sentiment: Highly positive, indicating great post-visit reflections and memories.5-Star Rating: A commendable 4.52 out of 5, supporting the overwhelmingly positive post-visit sentiment.
In summary, the sentiment remains positive across different stages of the customer journey, with some areas showing a mix of sentiments. The 5-star ratings align with the sentiment distributions, providing quantifiable feedback for each stage. Disneyland Paris shines particularly during the post-visit reflection and hotel stay but could potentially focus on the pre-arrival stage, where the sentiment and rating were the lowest.”
2. The CJ Stages and Touch Points Analysis
We have submitted the dashboard section solely devoted to the Customer Journey Stages and Touchpoints, just like we did with the upper part of the dashboard. We have asked for actionable insights to be extracted from the charts and for comments to be provided. I’ve included the results below.
Customer Journey Stages and Touchpoints Analysis
Let’s delve into an overview of the sentiment and touchpoints across different stages.
1. Pre-Visit Stage
Sentiment: Positive sentiment dominates at 58.2%, with only 10.6% being negative. This indicates a generally favorable perception before visiting.
Top Touchpoints: The most prominent positive touchpoints are website, price, and social media. While there is positive sentiment towards these touchpoints, there is also a presence of neutral and negative sentiments, particularly towards price and reservation, suggesting some mixed feelings.
2. Pre-Arrival Stage
Sentiment: A slightly reduced positive sentiment at 50.2% and a higher negative sentiment at 36.2% suggest mixed emotions during this phase.
Top Touchpoints: The main areas of focus here are transportation, website, and recommendation. These elements have mixed sentiments, with transportation and reservation notably showing substantial negative feedback.
3. Arrival Stage
Sentiment: Positive sentiments make a comeback at 64.2%, with negative feedback at 27.7%.
Top Touchpoints: Major touchpoints like queue organization, park entry, and luggage service show strong positive feedback. However, queue organization and ticket prices have noticeable negative sentiments, hinting at potential areas of improvement.
4. Park Experience Stage
Sentiment: The positive experience is strongly felt here, with a sentiment of 64.9%.
Top Touchpoints: Among the various attractions and amenities, shows, attractions, and food & drink stand out. Notably, food & drink and efficiency and effectiveness have split sentiments, with both positive and negative feedback.
5. Dining Stage
Sentiment: This stage has a positive sentiment of 53.2%, but a significant negative sentiment at 32.8% suggests there’s room for improvement.
Top Touchpoints: Key touchpoints like food quality perception, food quality reality, and restaurant atmosphere are highlighted. There’s a clear disparity between food quality perception and food quality reality, which can be a major focus area.
6. Hotel Stage
Sentiment: The hotel experience is largely positive at 65.8%, with only 14.8% negative sentiments.
Top Touchpoints: Touchpoints such as hotel rooms, hotel staff, and hotel price have received substantial feedback. While the first two are majorly positive, hotel price has a notable negative sentiment.
7. Departure Stage
Sentiment: Overall positive at 64.3%, but with some mixed feelings evident from the 23.9% negative sentiment.
Top Touchpoints: Tickets, overall experience, and queue organization are significant touchpoints. All of them, especially queue organization, display mixed sentiments, highlighting areas that might need attention.
8. Post-Visit Stage
Sentiment: Overwhelmingly positive sentiment at 81.6%, indicating guests leave with fond memories.
Top Touchpoints: Overall experience, friendliness, and customer service lead the feedback in this stage. All these touchpoints are heavily inclined towards positive sentiments.
9. Branding
Sentiment: A good brand perception is evident with a 71.1% positive sentiment.
While specific touchpoints are not highlighted for branding, the high positive sentiment suggests effective branding efforts by Disneyland Paris.
Conclusion:
Throughout the journey, the general sentiment is positive, indicating that visitors have an overall good experience. However, there are touchpoints, particularly during the pre-arrival and dining stages, where improvements can enhance customer satisfaction. This analysis can serve as a roadmap for Disneyland Paris to further refine their guest experiences.
3. Discover insights that are beyond human perception
Have a look at this chart:
Is it difficult to extract insights and provide useful information to the organization using this chart? Does it make sense to include it in our dashboard?
Let’s analyze the chart from a different perspective. Let’s have ChatGPT provide insights:
..:: W A R N I N G ::..
- Parade: This topic sees a distinct spike in negative sentiments around mid-2015 but then quickly fades away. Before and after this time frame, “Parade” doesn’t seem to be a significant concern for customers. This could indicate a particular event, change, or issue related to a parade during this period that didn’t resonate well with guests. Since it’s an isolated spike and doesn’t recur, it might have been a temporary event or problem that was resolved, or perhaps a specific parade that was introduced and then retired or modified due to feedback.
It’s always crucial for businesses to be alert to such isolated spikes in negativity, even if they’re temporary, as they can provide valuable insights into specific decisions or changes that may not have been well-received by the customer base.
As you can see, the model can quickly extract insights from complex visualizations without the need for human interpretation. We can also customize the analysis performed on specific visualizations.
4. WARNING: New Topic popping-up
The model can compare two charts and extract insights from that comparison. Let’s take a look at two word clouds. These are typically something nice to see in a dashboard, but they can be hard to interpret.
Let’s take a look at how the model analyzes the two clouds. In this case, the left cloud represents positive topics while the right cloud represents negative ones.
The tag clouds provided offer a comprehensive insight into Disneyland Paris customer reviews. Upon comparing the two:
- Common Themes in Both Clouds:“Attraction”, “ride”, “entertainment”, “experience”, and “overall” seem to appear prominently in both tag clouds. This indicates that these elements are core to the Disneyland Paris experience and elicit strong reactions, both positive and negative.Topics like “staff”, “park”, “ticket”, “dining”, “hotel”, “reservation”, and “crowd” also appear in both, highlighting areas that can swing either way in terms of customer satisfaction.
- Predominantly Negative Sentiment:Words like “inconvenience”, “disappointment”, “closure”, “difficulty”, “wait”, “pricing”, and “accident” stand out. These suggest pain points in the customer journey, possibly related to pricing strategies, attraction maintenance, and queue management.”Barrier”, possibly related to accessibility issues, is a noteworthy point of concern.
- Predominantly Positive Sentiment:“Joy”, “wonder”, “socialize”, “greeting”, “memory”, “appreciate”, and “memorable” stand out in the positive cloud. These evoke feelings of enjoyment and highlight the magic that Disneyland Paris can offer to its visitors.”Quality”, “cleanliness”, and “safety” also appear, emphasizing areas where the park is succeeding.
Recommendation: To enhance customer satisfaction:
- Address Negative Sentiment: Disneyland Paris should prioritize addressing inconvenience and wait times, possibly through better queue management and real-time updates to guests. Ensuring fewer unexpected ride closures and enhancing overall accessibility could also improve the negative sentiment around “barrier” and “closure”.
- Leverage Strengths: Build upon the magical and memorable experiences that the park offers. Continue maintaining high levels of cleanliness and safety, and perhaps introduce more events or attractions that evoke joy, wonder, and nostalgia.
- Neutralize Ambiguities: For topics appearing in both clouds, such as “staff” or “ticket”, it would be beneficial to deep dive into the feedback to understand specific issues or praises, allowing for targeted improvements.
By addressing these areas, Disneyland Paris can better ensure a consistent and delightful experience for all visitors.
CONCLUSIONS
The new functionality provided by OpenAI is proving to be extremely useful for Customer Experience Managers. In particular, the feature that allows for quick analysis of dashboards by visually representing parts of it and comparing charts is simply amazing. A few years ago, the idea of creating a “storytelling dashboard” was nothing more than a nice dream. But now, it has become a reality. However, the question that arises is whether we will still need dashboards in the near future.
