On Feedback

Personal feedback: where to start?

6 minute read

There are many ways to gather feedback about our own work and skills. Here I want to introduce a very simple yet structured practice that can be done in any environment without any specialized tool.

I usually suggest to run this kind of feedback gathering at least once or twice a year but no more than four times. If you’ve never done it, I’d highly advise to give it a try.

Step 1: Identify People

The first step is to identify who you want feedback from. This step can vary a lot depending on your role, the kind of organization you work with, and your goal.

Here you should identify who you’d like feedback from. This shouldn’t be just your team (regardless if you’re part of one or leading them), but a cross section of people you have worked with. I’d run through these categories, in order:

  1. Your team — these are usually the people closest to you, and are more likely to give you more precise feedback. If you manage a large team, this includes all first level reports, but can include everyone beyond that too.
  2. Your boss — essential to get a higher-level perspective of how you’re doing.
  3. Your peers — these are people anywhere in the company, but that you recognize close to you in term of skills, background, or even just you resonated with them when you worked together.

I don’t think there’s a good number to aim for here, it really depends on your situation. Just to give a data point: in my recent years, in a company of 1400+ people and being a director in a business unit of about 50 people, I usually have had in the list about 20-30 people overall, and I got a response rate between 35 and 50% within a couple of weeks. This is just a single reference, try and find the balance that works for you and your organization.

Step 2: Survey

You should try to keep the questions very simple. Two for me are the foundation of any personal feedback:

  • What things should I do more?
  • What things should I do less?

There are many variations of these questions, so the specific language can vary, but you can intuitively see the sense of it: one tells you what’s good, one tells you what you should review and change.

Then you should add one or more special questions that are a deep dive on what you’re focusing on specifically at the time when you’re sending the survey. For example, a few years back I was working a lot to improve my feedback, and thus I added this question:

  • How do you find my feedback? Not just content, but also tone.

At a different time, I was looking more in general at the way I was communicating, so the question was:

  • How do you find my communication style?

In a period of change, with a lot of different threads of work happening at the same time, I instead asked this question:

  • From your perspective, what I should be paying attention to in the next few months?

Looking back there’s also a final question I’ve personally asked almost every time. I don’t consider it essential but I found it useful to help me identify better my value. This one is less about having actionable feedback, and more to help you grow as a professional in the longer term. The question is:

  • What’s in your view the biggest value I’m able to give to the organization?

In general, the combination of these questions is a minimum of two and a maximum of 4-5 questions. This is because the more question are added, the less likely it is for people to find the time to reply.

Step 3: Tokenize

This is a step a lot of people that have done text analysis of any kind will probably know really well. With tokenization I mean to break the qualitative feedback you received in single pieces, each with a distinct piece of information. You can do this in a normal text editor by copy and pasting to a new list, or use a spreadsheet pasting to individual cells, or whatever you find easier to use.

For example, this is a piece of feedback I received years back that helped me:

“I’ve seen you have a very good balance of standing your ground and allowing people to do something although you disagree. Now, this is my personal opinion (maybe I’m biased because I mostly agree with you), but you should stand your ground more”

And here’s how I broke it down:

  • “very good balance of standing your ground”
  • “allowing people to do something although you disagree”
  • “you should stand your ground more”

At this stage don’t rewrite or change the text, or try to interpret it. Just copy-paste it under a set of categories that loosely match the questions you asked — which means that at minimum you are going to have two categories:

  • Do More
  • Do Less

One more category you could consider adding is: appreciation. People tend to add in the feedback some extra positive things they think about you, so making these explicit can be very useful to help with your insecurities.

By the end of this you get a categorized lists of distinct data points, one token per row.

Step 4: Summarize

Reviewing a long list of text snippets even if categorized can be difficult, so the next step is to summarize them. You keep the same categories, but now you merge together the tokens that are basically saying the same thing, so it’s easier to have an overview of all the feedback.

When you merge try to keep as much as possible the words used in the various sentences, in order to introduce the least amount of personal bias as possible. You could also consider to keep a counter on each insight to note how many tokens said that same thing. I don’t consider this a hard number (7 vs 5 are roughly the same, but 1 vs 10 are not) but it helps in weighting them.

This creates a short-list of individual insights under each category.

Step 5: Find Insights

With the summarized shortlists, now you can review and try to find what are the top 3 things that seem most impactful to work on.

Then convert these 3 things to actions, and put these actions either in your future to-do list or in your calendar… in general, whatever works for you. This is for me a key step: knowing what to do sometimes gets lost in the day-to-day work so creating precise tasks makes sure the insights don’t get lost.

The way to assess if then the work was effective is just by comparing then with the next round of feedback months: did the feedback change? If yes, it means that you were able to improve, if no, you know you need to change strategy.

One consideration is about making this anonymous. I don’t think there’s a hard rule to say if the above should be done anonymously or not. The pro to anonymity is that people might give you more sincere feedback about things they would be worried telling you personally. The pro to having a name attached is that you can then have a conversation. Again here context is relevant: I personally always run the above anonymously because I’ve already a lot of other moments where I gather personal and direct feedback about my work, so I use this to get a different perspective. If you are organized differently, your decision here might be different too.

You don’t have to work for a company to do this. Even if you are freelance, or run a volunteer group, consulting, or any kind of mixed scenario, this can be done and can provide a lot of benefits. What changes is just the list of people you’re reaching out to, and how you’re introducing your request.

It also doesn’t have to take a lot of time. If you’re worried about the investment, just run one with 5 people and just the two foundational questions. For me what’s important is that you have an ongoing personal feedback practice. And if you already have one, I hope this article gave you some extra insights.