Context
Our client, a leading French multi-brand mutualist insurer, wanted to lift the conversion rates (direct and indirect sales) of its client contact centres, and reduce the rate of no-shows on branch appointments booked over the phone. Advisors were working with a single sales pitch, calibrated on what they perceived as the dominant client profile.
The problem
The dominant-profile hypothesis is statistically false. Research on decision-making in insurance (Bauer & Wätjen 2018) identifies five distinct decision models coexisting in any retail insurance client base. A risk-averse client, a bargain hunter and a spender need three fundamentally different, sometimes antagonistic pitches. A single pitch wins on one profile and loses on the others.
Above all, sales advisors show a systematic perception distortion: they massively overestimate the share of bargain hunters in their client base, because that is the profile that generates the most salient and most memorable objections (availability heuristic, Tversky & Kahneman 1973). The most-remembered profile is not the most-frequent profile.
Our approach
A six-phase project:
- Immersion and call listening in a regional contact centre, complemented by manager interviews.
- Scientific literature review on remote client relationships in insurance: 8 papers, including foundational work by Bauer & Wätjen (2018) on client decision models, Holbrook (2006) on perceived value, and Söderlund (2018) on perception of remote relationships.
- Client interviews to validate theoretical profiles in the field and identify observable verbal markers at the start of a call.
- Data science: generalised linear model to optimise the routing between contact centre and branch based on the detected profile, and predictive model on transformation probability by profile.
- Design of the "5 Fantastics" tool: operational typology of five decision profiles (bargain hunter, loyal, spender, risk-averse, indifferent), with distinct argumentation levers per profile and a "Golden Question" to surface the profile in the first minutes of the call. Advisor training, memo, follow-up grid.
- Test-and-learn on two pilot teams (retail and pros) over four months, followed by a debrief through structured interviews with 15 team members.
Results
5 profiles validated by advisors during the pilot ("What clients say is exactly what's on the sheet"). Quantitative confirmation of the perception distortion: bargain hunters are overestimated by a factor of 2 by advisors (perception biased by objection salience). Tool rated "indispensable" and unanimously endorsed by pilots. Shift from one sales routine to five distinct routines. Recommendation to integrate the tool into initial training and into the client relationship tool.
Advisor verbatims: "Frankly, it's not important, it's essential." / "Super useful." / "It forces you to ask the question and adapt your pitch."
What this case illustrates
An advisor's perception of "their" client is systematically biased toward the most-memorable profile, not the most-frequent one. This bias has been documented since Tversky & Kahneman (1973) as the availability heuristic. In a contact centre, this bias costs directly on transformation rate, because advisors calibrate their pitch on an imagined client base that over-represents objecting profiles.
The lever is not in product training nor in sales technique coaching. It is in a start-of-call profiling tool that restores the actual map of the client base.
Frameworks used
Availability heuristic (Tversky & Kahneman 1973), anchoring bias, loss aversion, framing effect, choice overload. See the glossary.
Frequently asked questions
How to identify a client's decision profile at the start of a call?
Through a calibrated question (which we call the "Golden Question") asked in the first two minutes. Its phrasing is designed to elicit the verbal marker of the dominant profile. The question is never direct ("are you price-sensitive?") because profiles would give each other away. It is indirect and projective.
What are the 5 client decision profiles in insurance?
Bargain hunter, loyal, spender ("American"), risk-averse, indifferent. This typology, derived from Bauer & Wätjen (2018), is stable across European retail insurance markets.
How does Krakn measure the effectiveness of a profiling tool in a contact center?
Through comparative test-and-learn on two pilot teams, over a duration long enough to neutralise novelty effects (minimum 4 months), and a debrief through structured interviews and analysis of transformation rates.
Why do advisors overestimate bargain hunters in their client base?
Because bargain hunters generate the most salient (price-related) and memorable objections. The availability heuristic (Tversky & Kahneman 1973) means advisors reconstruct a mental map of their client base from what they remember, not from an objective count. On this case, the perception bias was a factor of 2.
What is a "Golden Question" in B2C sales?
A single, calibrated, indirect question asked at the start of an interaction, that surfaces the client's decision profile without them being aware. Its phrasing varies by sector and product, but the principle is constant: extract psychological information without triggering the client's defensive posture. Want to equip your contact centre with a profiling tool? Let's talk →