A Step-By-Step Guide To Building Your Startup’s First Performance Review Cycle
Okay, so, we had a tough time creating this guide. “Why”, you ask? Brilliant question!
Well, because we’re torn. Not because we don’t believe in performance reviews, but because we do. We just hate how most companies actually run them.
That’s why we’ve broken this guide into 2 (admittedly chunky) parts:
- Why (most) performance reviews processes suck, and
- How to build one that, well, doesn’t.
Enjoy.
Part 1: Why Performance Reviews (Mostly) Suck. Especially at Startups
Let’s be real: performance reviews are the New Year’s resolutions of the workplace. Everyone starts with good intentions, but when it comes time to put in the work, things fall apart. And honestly? It’s usually not any one person’s fault (nobody wants to fill out 101 lengthy feedback forms).
But why do performance review processes kind of suck, especially at startups?
1. No One Knows Why They're Doing It
Are reviews about development? Promotions? Raises? A legal checkbox exercise? If you’re not clear on the goal, your team definitely won’t be.
2. Startups Overbuild Too Soon
You just hired your 25th employee, and someone rolls out a 6-step performance process that takes 4 months and asks managers to answer 27 questions per employee. Chill.
3. The 360 Trap
Self + peer + manager + upward reviews = process bloat. Most teams aren’t ready for this yet. 360 reviews work at some orgs, but not all.
4. Bad Questions = Bad Feedback
Questions like “Would you grab a drink with them after work?”. Ask better questions.
5. Feedback Goes Nowhere
If it’s not followed up with a convo or a growth plan, what’s the point?
6. Premature Comp Tying
If your comp philosophy and framework is still a work in progress (which it likely is if you’re sub 100 people), tying it to review scores is dangerous. More on that below.
7. Inconsistent Manager Quality
Some write novels. Some recycle last year’s review. Nobody knows what they’re doing.
8. No One's Been Trained
Giving and receiving feedback is a skill. If you don’t teach it, don’t expect it to be good.
9. It Takes Too Long
A 3-4 month performance review process is too long. You’re a startup, not a Fortune 500 goliath.
10. “Everything Is Made Up and the Points Don't Matter”
“Whose Line Is It Anyway” reference? Anywho…
Some managers hand out perfect scores like candy. Others never give more than a “Meets expectations”. Then HR or Finance curves everything to fit a made-up distribution. The result? Inconsistent, arbitrary scores no one trusts.
Part 2: How to Build a Startup-Friendly Performance Review Process (That Doesn’t Suck)
You don’t need to copy-paste what Google or Deloitte does. You just need a process that’s lightweight, clear, and actually useful for your team.
Caveat: The below assumes you’re a fast paced org. If things move a bit slower at your company (enjoy that, by the way…) you may or may not be able to get away with a process that’s a bit of a heavier lift and more involved.
Here’s a step-by-step guide:
Step 1: Define the Purpose (and Stick to It)
First thing first, answer this:
Why are we doing performance reviews in the first place?
Common (valid) reasons:
- Support employee growth and development
- Align expectations and give feedback
- Calibrate performance across the team
- Start building a culture of feedback
What to avoid:
Trying to do everything (all of the above + comp planning, promo decisions, values alignment, competency frameworks, etc.) all in one go. Pick 1–2 core goals and be transparent about them.
Note: Reviews can inform compensation or promotion conversations, but that’s very different from tying scores directly to raises or bonuses. More on that below.
Step 2: Choose the Format (Simple > Fancy)
For small teams (1-100ish), skip the bloated 360s. You don’t need self + peer + manager + upward feedback.
Start with something simple:
- A downward/manager review
- Optional self-reflection
- A 1:1 conversation to talk about it all
Step 3: Identify What You’re Measuring
Let’s make this simple: if something’s important at your company, bake it into your reviews from the start.
This doesn’t need to be its own separate step in the process - it should just show up in the questions and categories you include.
Here are a few examples:
- If company values matter
→ “Give an example of how this person demonstrated -insert core value- this cycle.”
→ Or use a short rating scale for each Value (1–3 works great). - If customer impact is your north star
→ “What tangible client outcomes did this person drive in the past X months?” - If team collaboration is key
→ “How has this person contributed to team success or unblocked others?”
Startups often skip this part and end up measuring... vague feelings? Name what matters and reflect it in the structure.
Step 4: Build a Lightweight Feedback Form
Use a mix of:
- Thoughtful, open-ended (qualitative) questions
- Simple, rating-scale (quantitative) questions
Open-ended questions:
See Step 3 above That is, map out the things you want to measure and ask hyper-specific, behavioral-based questions about them:
- “What specific outcomes or projects did this person drive or contribute to this cycle?”
(Focuses on tangible impact and deliverables.) - “What new skills, behaviors, or approaches has this person demonstrated since the last review?”
(Highlights growth in how they work, not just what they do.) - “What’s one area where this person could focus to level up their impact or effectiveness?”
(Keeps development tied to performance, not just personality.) - “How has this person contributed to team success — unblocking others, sharing knowledge, or collaborating effectively?”
(Makes “good teammate” concrete and observable.)
Rating-scale question:
Decide on a rating scale. “What scale”, you ask? The answer to that is complex and will depend on a few things. As a good rule of thumb, we typically recommend something simple and lightweight, like a 3-point scale:
- 1 - Struggling (Below Expectations)
This person is having trouble meeting expectations or delivering reliably. They likely need additional support, coaching, or clearer direction.
Example: Missed key deadlines, unclear communication, inconsistent follow-through. - 2 - Solid (Meets Expectations)
This person consistently meets expectations. They deliver quality work, are dependable, and contribute positively to the team.
Example: Hits goals, works well with others, follows through without much oversight. - 3 - Exceptional (Exceeds Expectations)
This person regularly goes above and beyond. They have a strong impact, support others, and create value beyond their core role.
Example: Took initiative on a critical project, mentored teammates, or improved a key process.
Skip complex scales: Complex scales (even the classic 5-point scale) seem precise, but mostly just lead to a sea of 3s and endless debates about what makes someone a 4 vs. a 5. It turns feedback into a numbers game instead of a real conversation.
Step 5: Have the Review Conversation
Don’t just collect docs and call it a day. Set up a dedicated 1:1 for every review.
Agenda:
- Walk through the review feedback together
- Ask how the employee felt reading it
- Co-create or revise a growth plan for the next X amount of months/quarters
Keep it human.
Step 6: Follow Up (or It’s a Waste)
After the conversation:
- Share a short summary and co-create a growth plan
- Check in on those goals regularly during weekly 1:1s
Step 7: Run a Retro
Remember, the first iteration is your MVP, so it probably won’t be perfect. Make 1–2 small improvements with each cycle by asking folks things like:
- What worked well?
- What felt clunky or confusing?
- What would you change next time?
Rollout Tips
Roll it out with intention. Try a combo of:
- A live walkthrough (e.g. during All Hands)
- Slack reminders + nudges
- Drop-in office hours for questions
- A Notion hub with templates, FAQs, Loom videos, and deadlines
What about tooling?
- If your budget allows and you’re planning to grow, we recommend investing in a dedicated performance tool. There are a ton out there (Lattice, Culture Amp, or 15Five - to name a few). These tools often come with additional perks, such as built in AI, advanced filtering, and other related modules like engagement surveys, 1:1 templates, growth plans, and more.
- If your budget is tight or you’re not scaling aggressively, you can totally DIY it (G-suite + Slack + AI is a powerful combo if setup right).
Suggested Cadence
The frequency of your “formal” performance review cadence should be inversely proportional to how much time it takes to complete. The more frequently you’re asking people to participate, the more lightweight it should be.
- On one extreme, you can go “new age” and scrap formal reviews all-together. We’re actually not opposed to this if you’re replacing these “formal” reviews with frequent, structured 1:1 convos where performance is a standing agenda item.
- On the other end of things, you can adopt a “traditional” approach, and have an annual (or bi-annual) formal review cycle.
We know this isn’t the most helpful answer, but
Our recommendation: We know this isn’t the most helpful answer, but the cadence that works for you will depend a lot on your org, what it values, and how you’re structuring your overall performance philosophy.
Regardless of cadence, hold your people accountable to deadlines or the process loses value.
Chart & Carry pro-tip: In our experience, the #1 indicator that your performance process isn’t agile enough is if folks are unable or unwilling to stick to the deadlines associated with them. If that’s the case, time to streamline it.
Things We Recommend Skipping For Now
Tying scores to pay/promotions
Feel comforted knowing that most smaller smaller orgs (heck, even the larger ones) are winging it when it comes to compensation frameworks.. That’s generally okay, especially at the early stages.
What you don’t want to do is tie vague, biased, or inconsistent performance scores to raises and promotions directly. Performance reviews can (and often should) be one of many inputs into comp and promo decisions. Directly tying performance scores to compensation also sets the (unrealistic) expectation that external factors - like funding, financial performance of the org, etc - don’t matter.
360 feedback
It’s really hard to do 360 degree feedback well, especially at a company’s early stage. We typically recommend you revisit this later once you’ve established a culture of feedback.
Normalization curves
Simply not needed (or recommended) until you’re much larger. Even then, these get messy.
Want help building a review process your team won’t groan about? That’s what we do.
