Who’s Really Connected in Dutch Hip-Hop?

This one took way longer than I expected.

Me and ChatGPT were vibe-coding our way through Dutch hip-hop collaborations when we hit a wall. I was chasing a slick interactive edge chord diagram, but it just wouldn’t land the way I pictured it.

Eventually, I had to admit the insights were already there. The data was solid. Maybe it doesn’t have to be 110 percent pretty to be worth sharing. 😉

Rewinding for a sec.

Back in 2018, I wrote one of my first music blog posts on Medium about Dutch hip-hop. Wild to look back now, both at the tools I used and the way I thought about the scene. At the time, Dutch hip-hop was dominating Spotify in the Netherlands. That hasn’t changed completely, but the genre has evolved. It’s blending with pop, crossing into other spaces, and feeling more fragmented and fluid than before.

I wanted to reconnect. So I pulled together a dataset to sketch the shape of the scene today. Who’s releasing? Who’s collaborating? And how are those connections shaped by popularity?

🔍 Starting from the ground up

I began with a scraped list of Dutch hip-hop tracks released in 2024, pulled from a Glenn McDonald playlist aggregator. The goal was to avoid editorial bias, no playlist push, no algorithmic boost, just a raw view of who’s putting music out.

From there, I grabbed each artist’s discography from the past two years. For every main and featured artist, I assigned a Spotify popularity segment (where available):

🟡 Very Low
🟠 Low
🔵 Mid-Low
🟢 Mid
🔴 High
🟣 Very High

Then I looked at how often artists from one tier collaborated with others.

🔍 Who is working with who?

The chord diagram below shows how artists across different popularity tiers collaborate. Each arc represents a connection. The thicker the arc, the more collaborations between those groups.

Here’s what stood out:

🎯 Mid-tier artists = the scene’s backbone

They connect upward and downward, bridging the scene and driving the most collabs per artist.

💎 Top-tier artists are selective

Very High popularity artists mostly collab with peers — and barely touch the lower tiers. Only two collaborations crossed from “Very High” to “Very Low.”

🔁 Emerging artists stick together

Low and Very Low artists mostly collaborate horizontally. These tight webs often stay within their own echo chambers — local scenes, subgenres, maybe shared friend groups.

📈 Who leads and who features?

Next, I looked at the share of releases where each artist appeared as a featured act rather than the main artist. The horizontal axis shows the percentage of their releases where they were featured. The vertical axis shows the total number of songs they released across the two-year period.

Key patterns:

🟡 Very Low: Self-released and starting out

216 artists | Median 16.7% featured

The largest segment in the dataset. Most artists here are early in their careers, focused on releasing music under their own names. Feature rates are low, but not negligible — there’s a quiet network of peer-to-peer collaboration happening below the radar.

🟠 Low: Still solo-heavy, but more mixed

177 artists | Median 14.3% featured

Despite being relatively large, this group shows the lowest median feature rate. Most artists are still steering their own projects. That said, a few highly collaborative outliers lift the average — signs of artists testing collab strategies to grow reach.

🔵 Mid-Low: Quietly collaborative

152 artists | Median 17.8% featured

This group starts leaning more into collaboration. While still primarily solo-driven, feature rates tick up and the density of mid-volume collaborators grows. Artists here seem to be working their way out of the echo chamber — often leading, but not exclusively.

🟢 Mid: The engine of the scene

111 artists | Median 18.2% featured

With the second-highest population and consistently high release volume, this group is the backbone of Dutch hip-hop. While most artists lead their own tracks, even a ~18% feature rate across this group translates into a lot of collaboration. They’re the network builders.

🔴 High: Collaboration as strategy

105 artists | Median 25.0% featured

This is where strategy really splits. Some artists maintain full solo control, while others float above 40–50% feature rates. With a higher feature % and sizable cluster, this group contributes to a large share of total scene collaborations.

🟣 Very High: Selective, but not isolated

35 artists | Median 22.7% featured

Despite being the smallest segment, artists here collaborate more than expected. With a median feature rate over 22%, the top tier isn’t just commanding attention — they’re also showing up on others’ tracks. These collaborations are fewer in number, but likely higher in visibility and cross-audience impact.

🎥 The one that almost broke me

There was one last visual I really wanted to include. A full tangled web of every artist and every connection across popularity tiers. Zoomable, bundled, slick.

And honestly?
It looked really cool.

Here’s a preview:

But I got stuck. Not because the insight was missing. It was there. But I kept tweaking little things. Node order, edge curves, label clutter, all of it. A classic perfectionism spiral.

The truth is, the first two visuals already told the story. But I was too deep in the weeds, chasing a perfect version of something that was already working.

Lesson learned again.
Sometimes done is better than perfect. Especially when the insight is already right in front of you.

✌️ One last thing...

If I had more time (or, you know, label budget), I’d dig into how collaborations actually impact streams, do featured spots lead to measurable bumps? Maybe next time.

🔜 Next up: Popprijs trajectories

I pulled all artists from the Popprijs (a major music talent award in the Netherlands) over the past few years to see what happens after they’re in the spotlight.

Specifically, I’m tracking changes in:

• 🎧 Spotify monthly listeners

• 📸 Instagram following

• 📺 YouTube views

The big question:
Does Popprijs actually move the needle?

👨‍💻 A little code detour for the nerds

Getting back into the swing of things, I was mildly (read: deeply) annoyed to find out that Spotify has been quietly sunsetting some of its API endpoints — including the one for “Fans also like.”

Not great.
I’ve always believed that if you're building public-facing features on the backs of artists and creators, those data points should be open by default.

That said... there’s always a workaround. 😏

So, for those of you who want to pull “Fans also like” data anyway — here’s a little scraper that still gets the job done.

That's a wrap for this week! Let me know what you think by voting in the poll below. Catch you next time! 👋