Quantitative-Qualitative Customer Synthesis
Combine funnel evidence with customer context before explaining behavior
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 97%
This synthesis method pairs comprehensive behavioral measurement with continuous qualitative research. Teams track entry, exit, fall-off, conversion, repeat booking, sentiment, social conversation, and supply-side behavior, then share those signals across functions. They also gather external customer samples, including video that reveals context and emotional cues. Leaders step back from the metrics, question their own assumptions, and ask what the customer is actually thinking at that moment. When the numbers appear unintuitive, qualitative evidence can reveal that the customer is solving a different problem or occupying a different mindset than the company assumed. The output is a contextual explanation of behavior that can guide product and marketing choices more reliably than either dashboards or anecdotes alone.
Origin
Extracted from Marketing Against The Grain, where Hiroki Asai explained how Airbnb combines funnel analytics with external qualitative research and leadership judgment.
Core principles
- 01Behavioral data reveals what happened but may not explain why
- 02Qualitative evidence restores the customer's situation and mindset
- 03Leaders must remain skeptical of their preferred interpretation
- 04Customer understanding requires both scale and context
How to run it
- 1
Measure the journey
Track acquisition, funnel movement, fall-off, conversion, repeat behavior, sentiment, and relevant marketplace signals. Share the evidence across all functions shaping the experience.
Pro tip Connect measurements across moments rather than treating each page or campaign independently.
Watch out A narrow metric can reward a local improvement that harms the wider journey.
- 2
Gather outside voices
Continuously recruit representative people outside the company and collect qualitative feedback. Use interviews, written responses, and video where appropriate.
Pro tip Video can expose hesitation, emotion, and context that flat responses omit.
Watch out Internal employees are not a reliable substitute for external customer samples.
- 3
Raise the altitude
Step back from the team's immediate work and inspect the evidence with healthy skepticism. Separate what the data shows from what the team wants it to mean.
Pro tip Ask leaders to state which assumptions would change their interpretation.
Watch out Deep involvement in the work can make familiar explanations feel self-evident.
- 4
Reconstruct the mindset
Identify what the customer is trying to do, notice, or avoid at the moment when the behavior occurs. Use qualitative evidence to explain counterintuitive data.
Pro tip Describe the customer's immediate priorities in plain language.
Watch out Do not force behavior into the company's internal hierarchy of importance.
- 5
Synthesize and act
Form one account that explains both the measured behavior and its human context. Use it to prioritize a product, communication, or marketing response.
Pro tip Keep alternative explanations visible until further evidence separates them.
Watch out Correlation and a compelling customer story do not automatically establish causation.
In the wild
Airbnb may see data that conflicts with the team's initial expectations. Leaders combine the metrics with qualitative evidence and recognize that the customer is thinking about a different concern at that moment.
→ The team develops a more credible explanation of behavior and avoids optimizing for an internally important but customer-irrelevant detail.
Common mistakes
Treating the dashboard as the customer
Metrics capture behavior but not the full situation producing it. Pair quantitative patterns with direct qualitative evidence.
Using anecdotes without scale
A vivid interview can overrule a broader behavioral pattern without justification. Interpret qualitative findings alongside representative data.
Protecting the preferred explanation
Teams often identify with their work and resist contrary evidence. Leaders must interrogate their own thinking openly.
Is it for you?
Best for
It is best for marketing and product teams diagnosing conversion, retention, sentiment, or unexpected customer behavior.
Not ideal for
It is not ideal as a substitute for controlled experimentation when a specific causal effect must be proven.
From the transcript
“We mix that though with the work that we're doing.”
“We'll take a look at that quantitative data, but then we'll step back and look at the work with kind of a qualitative customer eye.”
“But really, you just need a lot of qualitative insights from people outside the company.”
From the episode
The $86B Marketing Playbook Behind Airbnb w/ Global CMO Hiroki Asai
Hiroki Asai