The Health Check Survey

Chuck Groom
Building VTS
Published in
4 min readMay 7, 2019

--

Our Health Check Survey is a tool for keeping pulse on team sentiment and getting ahead of critical problems. It’s an organization-wide web-based survey that’s regularly sent out asking team members how they are doing. Responses are not anonymous, but only managers may view individual answers. Conceptually, these surveys are an extension of regular 1:1s between an employee and their manager, but cover more breadth. Aggregate results are shared with everyone and can be trended over time.

At VTS, we started running these health check surveys 9 months ago. The results have been overwhelmingly positive:

  • Employees appreciate being able to give feedback on areas that might not come up in 1:1s.
  • Managers feel like they have more visibility into their teams and can get in front of problems sooner.
  • Leadership appreciates having easy-to-understand metrics about company sentiment.

Of course, companies should also conduct regular anonymous surveys that can go deeper and collect unfiltered critical feedback. VTS uses Culture Amp to conduct anonymous company-wide surveys twice a year; this gives us deep insight into our culture and morale, with benchmarks against the industry to put results in context.

Principles

The Health Check Survey is predicated on:

  1. The survey must be short and easy to take, and kind of fun.
  2. An individual’s answers must never be used against them; this requires a degree of trust with managers, and transparency about how results will be used.
  3. The results must be useful; we shouldn’t waste time asking questions that managers can’t respond to.

Our Health Check Survey

The VTS product-engineering-design team sends two health check surveys a quarter. This is a short Google form that takes about 5 minutes to fill out. The survey structure is:

  1. Introduction. Explain what this is, and that results are not anonymous. Ask for team (functional area) and role (engineer, designer, etc.).
  2. Role-specific questions. Based on the role selected previously, route to a different set of questions specific to that role. For example, we’ll ask engineers for their thoughts about topics like tech debt or pair programming.
  3. Team questions. We investigate teamwork, sense of group autonomy, and organization support.
  4. Individual satisfaction. These cover topics like personal work load, motivations, and potential.
  5. Rotating section. To keep things interesting we ask about a recent topic, such as a hackathon or desk move.
  6. Final thoughts. We end the survey with an opportunity to give feedback on the survey itself, and to share any issues that might not have been addressed.

Most questions are a multiple choice scale with a range of 1 to 5. There are a few open-ended responses, but those are optional.

Borrowing from Spotify

We borrowed the idea of this survey from Spotify’s Squad Health Check; and indeed, we copied many of their questions verbatim. As they say, “imitation is the sincerest form of flattery.”

The way Spotify runs their health check is quite different, however. They set up in-person team meetings with physical cards to prompt group conversation. This is perfectly reasonable, but we felt web-based surveys would be logistically much easier than setting up many meetings, and we wanted to allow people to respond privately about more sensitive topics.

Spotify’s “Awesome Cards” to use as health check indicators

Light Tone, Serious Topics

We followed Spotify’s lead to ask serious questions, but to anchor the responses with purposefully silly or melodramatic language. For example, here’s how we ask about someone’s current team:

We also insert silly clip art here and there to break the visual monotony. (Separately, note that some of our questions don’t fit a 5-point scale).

Reviewing Responses

Engineering managers may view individual responses. We appoint one manager to run the survey, and to export results to a Google sheet with some formatting and filters. This person also raises awareness of any worrisome trends or red flags. For example, if code quality is trending down we should discuss that with the management team; or if a person hates their physical environment we should prompt their manager to discuss that.

After each survey, we send an email to the leadership team describing the team health and noting important trends or response areas. Sometimes we need to go deep on the data and highlight response distributions or breakdown by team or role. We also send a broad email to everyone in the organization with a summary of results, but perhaps not with the same level of detail.

In this example, we may drill down on the result of one question to discuss not just the average number but distribution.
It’s useful to track the average of metrics over time. In this example (it’s contrived, not from VTS!) you can see the story of a team in trouble; even though code quality is up significantly, something started happening in January that’s hurting team cohesion and autonomy, as well as people’s sense of potential. Better investigate!

I would recommend this kind of regular health check survey to anyone. This is done in additional to regular 1:1s and anonymous feedback surveys, not as a replacement. Employees say this helps them feel better supported. And as a manager, I find having this tool is a kind of a safety net to alert me about things I may not see in my day-to-day.

--

--

Consulting CTO open to projects. I’m a serial entrepreneur, software engineer, and leader at early- and mid-stage companies. https://www.chuckgroom.com