navy logo
Products
PRODUCTS
survey icon
In-Product Surveys
Capture targeted user insights right in your product
replays icon
Replays
Recreate and optimize user journeys across your product
teal icon of a survey with chapters
Long-Form Surveys
Measure UX at scale with advanced link surveys and AI analysis.
heatmaps icon
Heatmaps
Visualize user behavior in your product at scale
feedback icon
Feedback
Collect continuous user feedback at scale
ai recommendations icon
AI Insights
NEW
Sprig AI generates actionable product solutions
Features
integrations
Integrations
mobile icon
Mobile
star icon
AI Analysis
magic pencil icon
AI Study Creator
dashboards icon
Dashboards
Solutions
by Use Case
continuously optimize icon
Continuously optimize
Analyze your users’ experience through core flows
solve pain points icon
Solve pain points
Uncover emerging trends in your users’ behavior
improve conversion icon
Improve conversion
Understand why and how users are dropping off
save time and resources icon
Save time & resources
Know which new features are worth the investment
by TEAM
uxr icon
User Research
Maximize the speed and impact of your research
Design
Validate and get buy-in for designs with real user insights
pm icon
Product Management
marketing
Marketing
code icon
Engineering
star icon
Customer Experience
Templates
lenny template
Survey
Develop Product Sense to Build Great Products
lenny headshot
Lenny Rachitsky
View Template
arrow icon
feedback template
Feedback
Continuously Collect Product Feedback
favicon
Sprig
View Template
arrow icon
Optimize New Features
Replay
Optimize New Features to Enhance the User Experience
favicon
Sprig
View Template
arrow icon
templates
Template Gallery
Discover how our team and community use Sprig templates to inform product development.
View All
arrow icon
Customers
square nav photosquare left logo
Square uncovered 100+ actionable insights within the first 6 months
Read Now
arrow icon
ramp nav imageramp logo
Ramp created customer-centric products with Sprig AI
Read Now
arrow icon
classpass nav photoclasspass left logo
ClassPass improved usability and retention by optimizing core user journeys
Read Now
arrow icon
users icon
Meet our Customers
Learn how top companies leverage Sprig user insights to boost conversion, reduce churn, improve onboarding, and more.
View All
arrow icon
Resources
blog icon
Blog
Get expert advice on capturing product experience insights
event icon
Events & Webinars
Learn from past Sprig events & register for upcoming ones
help center icon
Help Center
Explore our knowledge hub to get started
in Sprig
video tutorial icon
Video Tutorials
Get a crash course in Sprig with our guided
video series
AI replay announcement text with a dashboard showing AI insights
New: AI-Powered Always-On Replays
Read Now
arrow icon
EnterprisePricing
Sign In
Book a Demo
navy logo
hamburger menu iconclose icon
Products
caret icon
Products
survey icon
In-Product Surveys
teal icon of a survey with chapters
Long-Form Surveys
feedback icon
Feedback
replays icon
Replays
heatmaps icon
Heatmaps
ai recommendations icon
AI Insights
Features
integrations
Integrations
mobile icon
Mobile
star icon
AI Analysis
magic pencil icon
AI Study Creator
dashboards icon
Dashboards
Solutions
caret icon
By Use case
continuously optimize icon
Continuously optimize your product & website
solve pain points icon
Surface & solve pain points
improve conversion icon
Improve conversion rates
save time and resources icon
Save engineering time & resources
By TEAM
uxr icon
User Research
Design
pm icon
Product Management
marketing
Marketing
code icon
Engineering
star icon
Customer Experience
Templates
Customers
Resources
caret icon
blog icon
Blog
event icon
Events & Webinars
help center icon
Help Center
video tutorial icon
Video Tutorials
Enterprise
Pricing
Sign InGet Started Free
Blogarrow icon
Product Insights
arrow icon
Analysis at scale: How Sprig leverages AI to supercharge user research
Product Insights

Analysis at scale: How Sprig leverages AI to supercharge user research

Written by Kevin Mandich | Jul 09, 2021

July 9, 2021

Analysis at scale: How Sprig leverages AI to supercharge user research

Thematic Clustering as a Research Goal

One of the richest sources of customer experience data is an open-text survey response. Historically this is also one of the most difficult data formats from which to extract meaningful insights, especially at scale.

When user researchers run surveys with open-text questions, a common goal is to group the huge number of responses into a small number of bite-size and actionable take-aways. The identified themes are shared with product stakeholders and play a critical role in determining how to improve the product experience. An example with some responses and a summarizing theme could be:

Response: “I’m lost, can’t really find anything easily in the product”

Response: “It’d be nice if there was a way to find users by name”

Response: “Please add a search field”

Theme: “Add search functionality”

Performed manually, this analysis takes the form of placing the responses into a large spreadsheet, reading through them to locate patterns, defining themes that represent actionable groups of responses, and then assigning all responses to one or more of these themes (a.k.a. “coding”).

As you can imagine, this is a painstaking process and certainly can’t scale easily beyond a few hundred responses. Automating this process can be a powerful way to increase the leverage of researchers and bring the survey life cycle from weeks down to hours. The ability to do this accurately is also one of the key differentiators between Sprig and other customer survey tools.

Existing Attempts at Automation

Many methods of grouping customer experience data exist in the industry, but they tend to lack the nuance required by product researchers and stakeholders to make informed decisions. Examples of existing methods include, in order of increasing complexity:

  1. Word and phrase counts and string matching
  2. Topic extraction and modeling
  3. Keyword extraction & post-processing
  4. Similarity matching of keywords (e.g. using a thesaurus, or maybe a neural network)

These natural language processing (NLP) methods are useful for a surface-level analysis, but they all have shortcomings. Take the following three sample responses received from users of a fictional web service:

“The subscription fee is a bit steep”

“The cost to change vendors might be too much for us at the moment”

“You should include a pricing plan based on usage”

The topic shared by all three responses is cost. However, none of the words associated with cost are the same, nor are the surrounding phrases. This eliminates word frequency counts, string matching, and keyword extraction as possible techniques, even if stemming or lemmatization techniques are used. Moreover, the intent of each response is different:

  • Response 1 is complaining about the subscription costs
  • Response 2 is referring to the switching costs of adopting a new tool, which includes time, effort, and other resources in addition to just monetary value
  • Response 3 is requesting a different subscription cost tier

Even if more sophisticated techniques such as topic modeling or neural network-based similarity scoring are used, grouping these three responses together would not make sense from a product perspective, and could adversely affect the decision-making process resulting from such an analysis.

How We Do It at Sprig

We employ a state-of-the-art approach to capture the nuance of themes seen in open-text survey responses. Instead of just counting words or identifying a topic, we also consider the context of the survey. This method is critical for capturing the user's intent in discussing the topic.

The examples in the previous section show how our approach disambiguates responses with similar topics but different meanings. How about situations where it's impossible to determine the user's intent from the response alone?

Take the example response: "User search." If the question asked was "What do you like about our product?" then the respondent is praising this functionality. However, that same response to the question "What could we improve?" implies that this functionality is lacking. Considering the question helps reduce the ambiguity that is common in survey responses.

We do this by utilizing deep neural networks, which capture the complexities of natural language. Our models, trained on millions of survey response data points, capture context from the entire response and the question asked to provide a complete picture of the respondent's meaning.

The End Result

All themes should have a topic and an intent to provide a clear and actionable takeaway. It's also important to identify an element of emotional response - sentiment - and a recommendation based on the urgency of the theme's responses. Using advanced machine learning techniques, Sprig generates an AI analysis for you quickly, accurately, and at scale.

Sign up for our newsletter

Get the best content on user insights, design, and product management delivered to your inbox every week.

linkedin icontwitter icon

Written by

Kevin Mandich

Engineering PhD turned AI expert. Previously at Agari and Incubit. Has over six years of industry experience building state-of-the-art machine learning applications. UCSD alum.

Related Articles

The Customer-Obsessed Product Manager's Playbook for 2025
Product Insights
Dec 18, 2024

The Customer-Obsessed Product Manager's Playbook for 2025

15 Best Product Management Software Tools: 2025 Modern Product Tech Stack
Product Insights
Dec 16, 2024

15 Best Product Management Software Tools: 2025 Modern Product Tech Stack

Click Heatmaps: Mastering Click Analysis On Your Product
Product Insights
Sep 10, 2024

Click Heatmaps: Mastering Click Analysis On Your Product

white sprig logo
Products
In-Product Surveys
Long-Form Surveys
Feedback
Replays
Heatmaps
AI Insights
Features
Integrations
Mobile
AI Study Creator
Dashboards
AI Analysis
Security Standards
Solutions
BY use case
Continuously Optimize
Improve Conversion
Solve Pain Points
Save Time & Resources
BY TEAM
User Research
Design
Product Management
Marketing
Engineering
Customer Experience
Templates
Customers
Resources
Blog
Events & Webinars
Help Center
Video Tutorials
Session Replay Guide
Pricing
Enterprise
Company
About Us
Careers
Sprig Service Agreement
Privacy Policy
Data Processing Addendum
Status
Compare
vs Qualtrics
vs Fullstory
vs Hotjar
vs Medallia
vs Pendo
Copyright 2025 Sprig, All Rights Reserved
linedkin logotwitter logo