People tell you more when nobody can trace it

Banks and insurers run on small, specialist teams, where people stop saying what they think. Here is how an employee engagement platform protects them, in numbers.

4.7 / 5 on G2, Spring 2025

Two teams, one report
Anonymous survey · account view
Identity never attachedNot to an admin, not to a super admin
Retail lending
31 responses
7.4
Treasury
2 of 3 minimum
Not reported
The threshold
3
Report opens at
5
Every dashboard cut
3
Hard floor, never lower
Trusted by teams at
Cerence AIBayportAujan Coca-ColaAlumilRobertshawAstra Service PartnersGarrett
Replace

Most financial services teams are replacing a form

  • The employee ID field is the whole problem. Ask for a code so you can chase non-responders, and nobody believes it is anonymous again.
  • A form is often genuinely enough: one short survey by email, in a team small enough that everyone already talks. Test yours before you buy anything.
  • What changes is who administers it, not what it can do. On small specialist teams, belief is the difference between the answers you want and the ones you get.
The procurement and security path
What the employee opensTwo front doors
Protect

Where anonymity holds, and exactly where it stops

Protect people enough that they will be honest, and tell me who is unhappy. Both, in one breath. We attach no identity to an anonymous response.

  • Cut to a team, not a person. Business line, district, branch and tenure become filters. Groups under the threshold stay hidden.
  • Talk to them without knowing them, in a two-way thread from their comment.
  • Surface what needs attention. Sentiment and criticality raise the serious comments. No risk flag names a person.
  • Use a different survey where being known is expected: onboarding, exit, and Speak Up for conduct cases.
How anonymity is enforced, in full
The lineAnonymous survey · open a card

What you can see

What nobody can see

Who wrote it
Which response belongs to which person
Any cut below the threshold

CultureMonkey does not cross this line, for any role.

The right-hand cards do not open. There is nothing behind them.

Reach

Reaching the branch and field teams a bank runs on

  • Five channels plus shared links and QR. One primary channel per survey, reminders can use another. Answering always happens in a browser, never in the chat thread.
  • Their language, not yours: 100+ of them. A preferred language on the record renders automatically; otherwise they pick once and it sticks.
  • A forwarded link does not open. The passcode is matched to an active employee in that survey’s audience. It holds while the passcode step is on: employee-ID mode and skip-passcode bypass it.
Frontline and deskless surveys in depth
Responseslive
WhatsApp
District, East · response received
now
SMS
Field sales · response received
1m
Shared QR
Branch counter · response received
2m
Email
District, West · response received
3m
Teams
Collections · response received
4m

Field sales 41% against 88% at head office. The low number is the finding.

Hold

An organisation that keeps changing shape

  • Entities and brands as their own dimension. Business units and groups sit above team and location, so an acquired agency reports on its own, with its own logo.
  • One entity’s HR lead does not read another’s. A data scope built from teams, locations, units, groups and custom attributes, narrowed further per survey.
  • Reporting lines you did not draw by hand, from the manager field on each record. A manager who reports to themselves in your HR system does so here too, which is why the data check runs before launch.
Six entities, one sealedGroup account

Two different reasons a number is missing, in one grid. Neither is a masked or blurred value: a scoped admin sees nothing there, and a team under the threshold is not displayed at all.

See your own thresholds, org shape and rollout plan on a live walkthrough.

Release

Managers see it when you decide

  • Preview before you release. View as manager shows exactly what a named manager will see, before they have access. Off by default.
  • Choose what a manager sees: comments, heatmap, strengths, action creation, conversations, exports. Every flag is account-level and hits all managers at once. There is no per-manager override.
  • Stage it by granting the role later: HR in cycle one, managers in cycle two. The threshold is one account-wide value though, so you cannot hold department heads to 5 and line managers to 3.
What managers do with it once they have it
Release day, from the manager’s sideJ. Meyer’s inbox
CultureMonkey
Your team’s results are ready
now
Not sent
Open-text comments · hidden on this account
·
Not sent
Individual responses · do not exist to send
·

Two of the three notifications never fire. That is the release policy, seen from the receiving end, and it applies to every manager at once.

Compare

The number has to hold up for years

  • A flat average of every rating answer in scope, deliberately not the average of the driver averages. Open text and eNPS have their own reports.
  • Every rating question is 5-point. Eight preset label sets, no custom 6 or 7-point scale, no renaming the points. The reporting scale displays 0-5, 0-10 or a percentage.
  • Financial Services, Insurance and Banking are explicit benchmark industries, matched to your size band. Aggregated reference data, not live competitor data, and standard drivers and eNPS only.
How the benchmarks are built
You against the benchmarkYour score · size band · industry
ComparatorAccount setting
Industry · sizeFinancial Services · 1,001-5,000
YouSize bandIndustry
Client-facing pressure, a custom questionNo benchmark

Workload is the only driver where the size band beats you, by 0.9. Switch region and some rows go empty.

Interpret

A named human, not just an AI

  • The question set is researched, not improvised. Standard drivers come from a research-backed engagement framework.
  • Analysis is AI-assisted and correctable. An admin can change a tag the model read wrong, and the correction flows into the aggregate counts. AskCooper puts a question to your own data in plain English.
  • People Science is part of the offer. Confirm with us what is included in your plan and what is a separate engagement, so nobody is surprised later.
How the analysis works
From comment to actionOne cycle

The card moves across the board. The author does not move with it: every column reads Anonymous, and one tag was reclassified by a person this cycle.

In their words

Hear it from people leaders

Mollie Mock Kaufman, Director of Talent Development & Training at United Bank
CultureMonkey made it easy to automate our employee engagement surveys while still giving us the flexibility to customize them. Their real-time data analysis, combined with insights from the People Science team, helped us uncover key focus areas, align our strategy with industry benchmarks, and strengthen our leadership development efforts.
Mollie Mock KaufmanDirector of Talent Development & Training · United Bank
View case study
Heather Kane, Change Management & Engagement Lead at Robertshaw
QR codes, multilingual support, and open comment analysis helped our managers understand culture deeply and have more meaningful conversations, something that was not possible before.
Heather KaneChange Management & Engagement Lead · Robertshaw
View case study
Elizabeth Egan, Director of Talent Management & Organizational Development at Cerence AI
CultureMonkey is able to translate that accordingly, not just a translation you might get on Google, but one that is actually utilized in each language, understood as the question is intended to be understood. And that is done by the employee choosing the language they would like to take that survey in, which is a really big difference maker.
Elizabeth EganDirector of Talent Management & Organizational Development · Cerence AI
View case study
Proof

The science behind it is real, and recognized

15

drivers in the research-backed framework your engagement surveys score against

10M+

employee data points behind the benchmarks and sentiment models

4.7/5

average rating from HR leaders on G2

G2 High Performer, EnterpriseG2 Momentum LeaderG2 Fastest ImplementationISO 27001 certifiedSOC 2 Type IIGDPR compliant

Go deeper on our methodology, intelligence, and benchmarks, or start from the People Science overview.

FAQ

Employee engagement software for financial services: FAQs

Anonymity and trust

How anonymity is enforced, and what it can and cannot survive.

How do you keep an employee survey anonymous in a small team?
Two controls do it. A survey's report only opens once its response threshold is met, 3 by default on an anonymous survey. Separately, every dashboard cut is held to a minimum group size, 5 by default, raisable but never below 3. No filter and no role, including super admin, can display a smaller group.
Can we find out who wrote a specific comment, or who is at risk of leaving?
No. CultureMonkey does not attach an identity to a response on an anonymous survey, for any role, on any screen, and no risk flag names an individual. What you can do is cut results as finely as the threshold allows, filter comments by sentiment and criticality, and reply through an anonymous two-way conversation.
Can we reply to anonymous feedback without knowing who wrote it?
Yes. From a comment in the Feedback report an admin starts a conversation, delivered through a randomly generated, unguessable link. The employee reads and replies with no login and no account, labelled only as an employee, anonymously. Threads start from an existing comment, one per comment, and the option can be hidden from sub-admins and managers.

Reaching everyone

Branch counters, field sales and everyone without a work email.

How do you survey branch, field or non-desk staff at a bank or insurer?
On the channel they actually check. Each survey goes out on one primary channel, email, SMS, WhatsApp, Slack or Microsoft Teams, and reminders can use a different one. Counters and shared devices can be reached by a shared link or printed QR code. SMS and WhatsApp need enabling on your account; Slack and Teams need their integration connected.
Could someone outside the company answer our survey if the link were forwarded?
On a shared link or QR code the survey opens a verification screen first: the person's work email, phone number or employee ID has to match an active employee in that survey's audience, and nobody can answer twice. Two settings change this, employee-ID mode and skip-passcode, so the protection depends on leaving the passcode step on.
Can employees answer in their own language?
Yes. A survey is authored once and translated into the languages you add, which you can edit by hand. If an employee's record carries a preferred question language the survey renders in it automatically; otherwise they pick from that survey's languages once and the choice is saved. Anyone with neither set sees the base language.

Fit and evaluation

Where a form tool is enough, and what a platform changes.

Is Microsoft Forms or Google Forms good enough for an employee engagement survey?
Often yes, for one short survey by email where nobody worries about being identified. It stops being enough the moment people need to believe it is anonymous: a form you administer has no suppression, segmenting means handling the file yourself, and asking for an employee code is the thing that tells people it was never anonymous.
Can we change the rating scale to a 6-point or 7-point scale?
No. Every rating question uses a 5-point scale with one of eight preset label sets: Agreement, Satisfaction, Quality, Frequency, Performance, Importance, Focus and Happiness, plus an optional smiley display. Custom point counts and renamed points are not supported. What is configurable is reporting: scores can display as 0 to 5, 0 to 10, the default, or a percentage.

Results and rollout

What managers see, when they see it, and what happens next.

Who exactly are we benchmarked against?
Your industry and, where the data supports it, your company-size band. Financial Services, Insurance and Banking and Capital Markets are explicit benchmark industries; a regional filter covers EU, NA, LATAM, APAC and MEA, and eNPS carries its own benchmark. Two limits: benchmarks are aggregated reference data, not live competitor data, and custom questions have no external comparison.
How is the overall engagement score calculated?
It is a flat average of every rating answer in scope, deliberately not the average of the per-driver averages, so a driver answered by 2,000 people counts for more than one answered by 50. Answers are stored on a 1 to 5 scale and converted to your reporting scale. Comments and eNPS have their own reports, and thresholds still apply.

Security and commercials

The questions your reviewers and procurement will ask.

Where is our data stored, and do you have an EU region?
Yes. A European Union region is served from app.culturemonkey.eu alongside the primary environment; the documented regions are US and EU. Each customer's data sits in its own database schema, traffic is encrypted with HTTPS and HSTS, and passwords are one-way bcrypt hashes. If residency matters, confirm the region before provisioning, because moving afterwards is disruptive.
What happens to our data at the end of the subscription?
Retention and end-of-subscription handling are governed by your data-processing terms rather than a published product setting, so the answer that matters is the one in your contract: ask for it in writing during the evaluation. Separately, deleting a person pseudonymises their record while their answers remain inside past aggregates, which keeps historic reports stable.

Bring your org chart and your thresholds

A working session on your entity structure, your thresholds, and how you reach the branch and field network.

ISO 27001·SOC 2 Type II·GDPR·★★★★★ 4.7/5 on G2