This week, I am using Google Maps and Reviews, as well as ChatGPT, as tools to explore customer satisfaction (and dissatisfaction) based on their online reviews. By focusing on common key words and themes in both positive and negative customer reviews, we can determine Microsoft’s top strengths and areas for improvement through their customer’s eyes.
Positive Review Analysis
Review Collection – Analyzing Positive Reviews
Analysis #1
Using Google Maps, I searched Microsoft and chose one of the headquarters locations in Los Angeles, CA. Navigating to the Reviews tab, I sorted by the highest ratings and gathered all 4- and 5-star customer reviews. After pasting these reviews into ChatGPT, I gave the prompt:
Prompt: “What are the customers most satisfied with based on the reviews?”
ChatGPT’s response:
Based on the reviews provided, customers are most satisfied with the following aspects of Microsoft:
- Facility and Office Environment: Many reviewers describe the space as “amazing,” “modern,” “creative,” and “inspiring,” highlighting the overall design and atmosphere.
- Events and Learning Opportunities: Customers appreciate attending seminars, tech events, and coding sessions, which create engaging and valuable experiences.
- Campus Size and Amenities: The large campus, multiple workspaces, outdoor areas, and unique features (like lounge spaces) contribute to a positive impression.
- Work Culture and People: Visitors note that employees are “very nice,” reinforcing a welcoming and professional environment.
- Innovation and Brand Appeal: The association with well-known products and innovation (e.g., software, gaming, and tech development) enhances excitement and overall perception.
These themes suggest that customers value the combination of a modern, engaging environment, strong learning opportunities, and the innovative reputation of the company.
Great information! But let’s keep going.
Analysis #2
Next, I asked ChatGPT to pick out key words from the positive reviews to determine some of Microsoft’s strengths.
Prompt: “Now, please extract positive keywords from those reviews, and analyze the frequency and associations of these keywords.”
ChatGPT provided the following results:
“Common positive terms and phrases include:
- Emotion & Impression: great, very good, cool
- Facility & Design: amazing, nice facility, modern, creative, inspiring, swanky, open space
- Experience & Events: seminars, tech events, coding, learning, visitor office
- People & Culture: nice people, welcoming
- Scale & Features: huge campus, multiple rooms, outdoor space, lounge area
- Brand & Innovation: Microsoft, Office, Minecraft, tech, starting point”
Analysis #3
Lastly, I prompted ChatGPT to provide suggestions from our findings so far.
Prompt: “What would be suggestions from the findings?”
ChatGPT gave a number of actionable suggestions for Microsoft:
- “Continue Investing in Facility Design and Experience – The office environment is a primary driver of satisfaction and should remain modern, creative, and inspiring.
- Expand Events and Learning Opportunities – Seminars, coding sessions, and tech events are highly valued and strengthen engagement with visitors.
- Enhance Campus Accessibility and Navigation – While the campus is appreciated, some feedback suggests it can be hard to navigate or access.
- Leverage Brand and Innovation in Experiences – Strong associations with products and innovation should be more actively integrated into on-site experiences.
- Maintain a Welcoming and Professional Culture – Positive interactions with staff reinforce the overall experience and should remain a priority.”
Now, let’s see how the negative reviews look.
Negative Review Analysis
Review Collection – Analyzing Negative Reviews
Analysis #1
Following the same steps, I sorted by the lowest ratings and gathered the 1- and 2- star reviews from Google Maps and pasted them into ChatGPT:
Prompt: “What are the customers least satisfied with based on the reviews?”
ChatGPT gave these results:
“Based on the negative reviews provided, customers are least satisfied with the following aspects of Microsoft:
- Customer Service and Support Accessibility: The most common complaint is the inability to reach a real person, with heavy reliance on automated systems, AI chat, and unhelpful support channels.
- Lack of Issue Resolution: Customers report that their problems remain unresolved despite repeated attempts, including no follow-ups or promised callbacks.
- Account Access and Recovery Issues: Many users struggle to recover accounts, reset passwords, or regain access, often without effective support.
- Product Functionality and Software Issues: Some users experience bugs or disruptive behavior (e.g., unwanted file creation, system issues) that negatively impact usability.
- Communication and Responsiveness: Customers feel ignored or dismissed, citing long wait times, no responses, and being redirected without solutions.
These recurring issues highlight significant dissatisfaction with support systems, problem resolution, and product reliability, leading to frustration and loss of trust.”
To simplify, we’ll look at some key words.
Analysis #2
Next, I asked ChatGPT to pick out repeating words from the negative reviews to determine Microsoft’s biggest areas for improvement.
Prompt: “Extract negative keywords from those reviews, and analyze the frequency and associations of these keywords.”
ChatGPT came back with this list:
“Common negative terms and phrases include:
- Customer Service Issues: no live agent, automated, AI chat, unhelpful, not helpful, helpless, terrible service
- Lack of Resolution: no response, no callback, unresolved, ignored, no solution
- Account Access Problems: locked out, cannot recover, reset issues, no access
- Product/Software Issues: faulty, bugs, unwanted files, technical problems
- Frustration & Trust: worst, horrible, joke, irresponsible, unprofessional
- Communication Barriers: overseas support, language issues, transferred repeatedly”
Analysis #3
Finally, I asked ChatGPT to provide suggestions from these findings.
Prompt: “What would be suggestions from the findings?”
ChatGPT’s suggestions:
- “Improve Access to Live Customer Support – The lack of real human assistance is the most frequent and frustrating issue for customers.
- Ensure Timely Follow-Up and Issue Resolution – Missed callbacks and unresolved cases significantly damage trust and satisfaction.
- Strengthen Account Recovery Processes – Account access issues are high-impact problems that require faster, more reliable solutions.
- Enhance Support Quality and Training – Unhelpful or ineffective support interactions contribute heavily to negative experiences.
- Reduce Over-Reliance on Automated Systems – AI chat and automated responses should support, not replace, effective human assistance.”
Ultimately, the customers want to know that a living person is handling their issue.
Reflection
Looking at both positive and negative customer feedback for Microsoft helps give a better understanding of the overall customer experience, showing what the company does well and where it needs to improve. These insights can be used to make more informed decisions that improve customer satisfaction, trust, and long-term success.
