Skip to content
KatafactsBeta

Voice of the customer

VOC evidence log

VOC evidence log is licensed CC BY 4.0. Attribution: Katafacts (katafacts.com).

Customise with AISkill file ↓

1 · What it is

What it is

A VOC evidence log is the foundational Voice of the Customer artifact — everything downstream in this family (affinity diagrams, CTQ trees, segmentation work) starts from evidence captured here. Each entry is one quote or observation, with its source and how it was gathered. The log doesn't just collect quotes, it tags each one with a theme and tallies which themes come up most — turning a pile of individually true but disconnected notes into a small, ranked list of what customers are actually telling you, with the receipts still attached to every claim.

2 · When to use it

When to use it — and when not to

Use it when

  • You have real customer quotes or observations — from interviews, support tickets, surveys, gemba sessions, or win-loss calls — and want to know what they add up to, not just where they're filed.
  • You want a defensible, evidence-backed answer to 'what are customers actually saying' rather than the loudest anecdote in the room.
  • You're about to make a product, messaging, or process decision and want the voice-of-customer evidence behind it in one place before you decide.

Not when

  • You only have secondhand impressions of what customers think, not actual quotes or logged observations — go gather real evidence first; a log built from memory inherits every gap in that memory.
  • You already know the theme you're looking for and just want to confirm it — an evidence log works best as genuine discovery, not confirmation of a conclusion you've already reached.
  • You need to design a new research study, not tally evidence you already have — that's the VOC collection plan.

3 · How to fill it in

How to fill it in

Scope
Which customer population does this evidence cover, and over what period? A log with no stated boundary blends customers who don't belong in the same analysis into a set of themes that describes neither group well.
Evidence entries
Enter each quote or observation with its source, source type, and date. Use the customer's own words or a close paraphrase — a generalized summary defeats the point of evidence. No theme yet; that's drafted for you next.
Themes
Drafted from your entries, then genuinely yours to edit — reassign any entry's theme and the tally updates immediately. Unlike every other artifact in this catalogue, this stays editable after generation, because theme assignment is the real judgement call here.
Narrative
Drafted from the top themes — what the evidence, taken together, actually suggests.
Next steps
What will you do with this evidence, ranked by impact and effort, owned and dated? A next step not tied to a named theme is a coaching failure mode, not a real action.

4 · What good looks like

What good looks like

The example below is a B2B SaaS team's onboarding feedback: nine real quotes across four themes, a tally showing onboarding friction as the dominant cluster, and next steps that trace directly to specific themes rather than a generic 'listen to customers more' resolution.

Same example, as a downloadable xlsx workbook.

Download .xlsx

VOC evidence log

New-account onboarding — VOC evidence log

Dana Whitfield, Product Lead · 2026-02-24

Team: Marcus Lee, Support, Priya Nair, Customer Success · Sponsor: Alex Romero, VP Product

Scope

Covers customer-facing feedback about the first 30 days after signup, gathered across support tickets, onboarding interviews, and the quarterly satisfaction survey — for accounts that signed up in Q1 2026.

Evidence entries

Onboarding frictionTop theme

  • New account setup requires our IT team to manually configure SSO before anyone can log in — that took us almost two weeks.

    Acme Corp · interview

  • I couldn't figure out how to invite my team during setup, ended up emailing support to ask how.

    NRG Industries · support-ticket

  • The first week was rough, honestly — too many steps before we saw any value.

    Blue Harbor Inc · survey

  • Getting our data imported took way longer than the demo made it look.

    Fenwick & Co · interview

Pricing confusionTop theme

  • We almost didn't renew because we genuinely couldn't tell what the annual plan actually included.

    Ashgrove Ltd · win-loss

  • Sales quoted one number and the invoice had different line items — took a call to sort out.

    Redline Systems · support-ticket

Feature request: reportingTop theme

  • We'd pay more if there were a way to build a custom report without exporting to a spreadsheet first.

    Meridian Corp · interview

  • The built-in reports don't match how our finance team actually needs to see the numbers.

    Kestrel Group · gemba

Support responsiveness

  • Support got back to us in under an hour every time we asked — genuinely impressed.

    Blue Harbor Inc · survey

Themes

Top themes: Onboarding friction, Pricing confusion, Feature request: reporting

  • Onboarding friction4 · 44.4%
  • Pricing confusion2 · 22.2%
  • Feature request: reporting2 · 22.2%
  • Support responsiveness1 · 11.1%

Narrative

Onboarding friction is the dominant theme, and it clusters specifically around setup steps that take longer than the sales demo implies — SSO configuration, team invitations, and data import all came up as separate customers hit separate points of the same underlying problem: too many manual steps between signup and first value. Pricing confusion is the second cluster, both instances tied to a mismatch between what was communicated pre-sale and what showed up on the actual bill or plan page, not to the pricing itself being too high. Feature request: reporting is a distinct, product-shaped ask — customers want to build custom views without an export step — worth tracking separately rather than folding into onboarding.

Next steps

  • Audit the setup flow end to end and name the single longest manual step (SSO configuration is the leading candidate from these interviews) as the first target for either automation or a guided-setup wizard.

    Linked finding: Onboarding friction — the largest theme, spanning SSO, team invites, and data import

    Impact: high · Effort: medium · Owner: Dana Whitfield · Due: 2026-03-17

  • Add an explicit plan-inclusions summary to both the sales quote and the first invoice, so what's included is stated identically in both places.

    Linked finding: Pricing confusion — both instances traced to a quote/invoice mismatch, not price level

    Impact: medium · Effort: low · Owner: Marcus Lee · Due: 2026-03-10

5 · Common mistakes

Common mistakes

  • Summarizing or paraphrasing away the customer's actual words before logging the entry.

    The point of an evidence log is that every theme traces back to something a real customer actually said — a summarized entry can't be checked against the source anymore, which is exactly what makes evidence more trustworthy than an impression.

  • Treating every theme with any entries as worth acting on.

    Same vital-few discipline as every other artifact in this catalogue (master-plan §1.3) — the top few themes with real weight beat a flat list of everything anyone ever said once.

  • Splitting near-duplicate themes ("onboarding is slow" and "setup takes too long") instead of merging them.

    Inconsistent theme labels understate how big a real cluster actually is — two 3-entry themes that are really the same issue look like two minor patterns instead of one major one.

6 · What it connects to

What it connects to

upstream

  • Gemba walk observation form

    Observations captured on a gemba walk are exactly the kind of raw material this log tallies — feed them in as entries rather than letting them sit in a separate notebook.

downstream

  • Ideal Customer Profile / qualification standard work

    A named theme pattern — which customers say what, and why — is direct input to who an ICP should target or exclude.

7 · Where AI helps

Where AI helps

Judgement — stays yours

  • Deciding whether a drafted theme label is actually right, or reassigning it to something that fits better
  • Deciding which theme is worth acting on first

Analysis — AI helps

  • Drafting a theme label per entry from the quote's actual content
  • Drafting the narrative from the top themes
  • Drafting next steps that trace to a specific theme

Drudgery — automated

  • Tallying entries per theme and computing each theme's share of the total, with the arithmetic shown
  • Identifying the vital-few top themes
  • Exporting to xlsx in the house format

9 · Rate this kata

Rate this kata