Client Case Study · Insurance & Financial Services

From feedback to focus: turning survey data into a customer experience roadmap

A leading southern African insurer had thousands of survey responses and no clear view of what to fix first. Key driver analysis gave them the answer, backed by statistics, not guesswork.

5,200+
survey respondents analysed
9
service touchpoints modelled
95%
confidence level applied
20+
significant drivers identified
The problem

Every attribute looked important. Not every attribute mattered.

Customer surveys across claims, withdrawals, and service centres captured 10–15 attributes each, alongside an overall satisfaction score, but with no statistical way to tell which ones actually drove that score.

9 touchpoints
Claims, next-of-kin, estate admin, withdrawals, multi-product enquiries, email and call centre, all surveyed separately.
120+ attributes
Rated across the study, far more than any team can action at once without a way to prioritise.
0 clarity
On which attributes were statistically proven to move overall satisfaction, versus simply "nice to have."
What the analysis found

The eight drivers that actually moved the needle

Multivariate regression, tested at a 95% confidence level, ranked every attribute by its true statistical pull on overall satisfaction, not by how it scored on its own.

Lifts satisfaction
Drags satisfaction down
Bar length = strength of statistical effect (t-statistic)
Making customers feel valued
Email Service Centre
t 7.50
Resolving queries within an acceptable time
Email Service Centre
t 5.90
Getting it right the first time
Email Service Centre
t 4.67
Handling requests within an acceptable time
Call Centre
t 3.62
Ease of interacting to resolve a query
Call Centre
t 3.44
Being kept up to date on claim status
Claims
t 3.09
Ease of finding where to submit a withdrawal
Withdrawals · negative driver
t -2.93
How enquiries were answered
Estate Administration · negative driver
t -2.48
Bars scaled to |t-statistic|; all drivers shown are significant at the 95% confidence level (p < 0.05). A negative driver means the attribute is significantly associated with lower satisfaction, a hidden friction point worth fixing.
What's working

Being kept informed and communicated with clearly beats speed and process simplicity as the top satisfaction driver across claims, next-of-kin, and estate administration.

Hidden friction

Email etiquette showed up as a significant negative driver across two separate multi-product touchpoints, a specific, fixable gap the client had not previously isolated.

Proof it works

Acting on the top drivers moved the scores

The client focused improvement efforts on the highest-impact call centre drivers. Tracked a year later, all three showed measurable gains.

Ease of interaction resolving queries
82
Baseline
88
Year later
+6 points
Handling queries within an acceptable time
79
Baseline
81
Year later
+2 points
Accuracy: resolved right the first time
79
Baseline
84
Year later
+5 points
How we get there

Our key driver analysis method

The same four-step process behind this result, adapted to your industry, your channels, and your data.

1

Collect

Pull together satisfaction ratings and every experience attribute, across as many touchpoints as you track.

2

Model

Run multivariate regression per touchpoint to isolate each attribute's true effect on overall satisfaction.

3

Prioritise

Rank drivers by statistical significance and direction, separating what helps from what quietly hurts.

4

Track

Re-measure the top drivers over time to prove the improvements are landing where they should.

Know what's really driving your customer satisfaction?

If your survey data is full of opinions but short on answers, let's find the two to five drivers that actually matter for your business.

Start a conversation: flowvisualpatterns.co.za
Flowing Visual Patterns · Research, analytics & data storytelling