5 emotional stages every data scientist goes through on a project

(and how to survive them)

I wish someone had taught me how to manage my emotions as a data scientist.

Not the code.

Not the math.

But the internal rollercoaster that comes with every new challenge.

Because almost every project I’ve worked on—whether personal or professional—has followed the same emotional pattern. I used to think the low points meant I wasn’t good enough.

Now I know they’re just part of the process.

This is what I wish I could tell every early-career data scientist: you’re not failing—you’re just in the middle of the emotional cycle of change.

Here’s what that looks like (and how I’ve learned to ride the wave instead of fighting it).

Let’s walk through each stage.

---

Stage 1: Uninformed optimism—"This is going to be awesome!"

I open a new notebook. I feel energized. Motivated. Ready to build something cool.

“This won’t take long,” I tell myself.

I picture clean outputs. A working model. Maybe even a recruiter who’s impressed with what I built.

I feel unstoppable—until I actually start working.

---

Stage 2: Informed pessimism—"Oh... this is harder than I thought."

Suddenly, the cracks appear:

• The data is missing half the columns I need.

• My model accuracy is worse than random.

• The goal keeps changing, or worse, isn’t clear at all.

I start second-guessing myself. I try three different approaches in one afternoon and none of them work.

This is usually where the negative self-talk starts to creep in:

“Why is this taking me so long?”

“Other people would have finished this already.”

---

Stage 3: The Valley of Despair—"Maybe I'm just not good enough."

This is the hardest part.

I feel stuck.

Overwhelmed.

Like I’m wasting my time.

There have been moments where I’ve closed my laptop mid-day and thought, “Maybe I’m not cut out for this.”

But every time I’ve hit this stage—and pushed through—I’ve come out stronger. More creative. More capable.

That’s the trick: realizing this isn’t the end.

It’s just the middle.

---

Stage 4: Informed optimism—"Wait, this might actually work."

Eventually, something clicks.

I find a better question to ask.

Or someone gives me feedback that changes everything.

Or I stop trying to be perfect and just simplify the problem.

Suddenly, I’m back in motion. I’m building again. And this time, it’s grounded in experience—not just enthusiasm.

---

Stage 5: Success and fulfillment—"I made it through."

This is the moment I publish the post, finish the dashboard, or deliver the results.

It’s not perfect.

But it’s done.

And it’s real.

I feel proud—not just because it worked, but because I didn’t quit when it didn’t.

Funny enough, that pride doesn’t last long… because the next project starts soon. And the cycle begins again.

---

If you’re in the messy middle—read this twice

Every data scientist I know (even the really senior ones) go through this cycle. They just move through it faster—or have more support when they hit the dip.

So here’s my advice if you're in the middle of a project, and it’s not going the way you hoped:

• Don’t quit during Stage 3. That’s where the magic is.

• Reach out to someone who's been there. (You're not supposed to figure this out alone.)

• Remind yourself: it’s okay to struggle. It’s not okay to disappear on yourself.

This is what real learning feels like.

Until next time,

Penelope