I tried speedrunning my new Instagram to 10K followers

A month ago, I set a goal of speedrunning my new @kallawaymarketing Instagram account to 10K followers as fast as possible.

I’ve scaled other accounts to 1M+ followers and 1B+ views, but I wanted to get deep in the 2026 trenches and prove that my skills and systems were still sharp.

Caveat: I “started” this speedrun at ~4K followers, but the account was completely dormant (hadn’t posted for several months).

The content system I ran is the exact same system I teach to everyone that I work with.

And it’s actually super straight forward…

Step 1: Run content research on Sandcastles to find proven winning topics in your niche and build your first test batch of 10 videos

Just like with paid ads, you want to run a test batch of 10 organic videos to get data back on what works best before deciding your content strategy.

This first step is the most important part of the process.

Almost everyone fails without doing this because they are just blind guessing every time they make a video.

To pick your first 10 video topics, study what's working in the niche, build hypotheses around formats and topics, pick a few that interest you and then find a unique contrarian insight that 99% of people would be shocked to hear but could quickly believe to be true.

Across the batch of 10 videos, I usually like to pick 3 buckets of topics and make 3-4 videos each. Each video has to have a unique contrarian take.

Sandcastles is the fastest way to find winning topics for any creator or batch of creators.

Step 2: Make and post the batch of 10 as fast as possible

Once you have these 10 validated topics and contrarian takes…film, edit, and post this first batch as fast as possible.

I posted 1x per day, 10 days in a row, but you can go 2x per day across 5 days if you want to get data back even faster.

Assume nothing will work. Set expectations at zero. The goal is to get data back. Ship fast to learn faster.

Step 3: Look at analytics from the batch of 10 (as a group) to identify significant outlier performers

Studying one video in isolation is not super helpful unless you really know what you’re doing. It’s much easier to spot patterns when you have multiple videos to compare.

I also use Sandcastles for this analysis because it’s amazing at surfacing non-obvious insights.

I pull up my channel, I sort by Outlier Score or Views, and then I study the performance of the 10 videos in a cohort.

In most batches, you'll typically see at least 1 outlier (performing 5-10x better than the average with respect to views, conversions or followers gained). If you see none, that’s okay, you’re probably still improving your baseline content skills, so reset and go to batch 2.

In the first batch of 10 for my marketing channel, I had 3 clear outliers.

Step 4: Extract the patterns that worked for the outliers

Watch the outliers back to back. Then, watch the losers.

Figure out what you did in the outliers that you didn’t do in the others.

I can spot this stuff this automatically because my brain is hypertuned for content, but know that it takes time to develop this skill.

Sandcastles is actually a huge helper on this piece as well.

You can just deep analyze all 10 of your videos (to get the full transcript, hook breakdown, topic, format, storytelling structure, etc.), export the data into a CSV (or use the Sandcastles MCP), drop it into Claude, and ask it to spot the patterns for you.

For my first batch, I noticed that the spoken hooks and visual hooks in the 3 winners were similar to each other, but very different from the other 7 in the batch.

This was the missing piece.

I make content all day and even I make mistakes in my own videos sometimes.

This batch analysis exercise is super valuable to help spot the mistakes and correct quickly.

Step 5: Roll those learnings forward to the next batches

This is what I’m in the process of doing now.

My system is crushing on my own channel exactly as expected.

It has a 100% hit rate for everyone I’ve ever taught it to that has followed through for more than 4 batches.

I call it the 10x Batch System.

This is what I've built my entire systematized content strategy around, and what I teach to people in SFA.

Funny enough, the hardest part isn’t the system, it’s knowing how to look at the batch and spot the patterns in the outliers that you should carry forward.

21 days ago, my account was at 4K followers. Now it’s at 15K.

I could have started at 0 instead of 4K. This playbook would have still worked the same, it just might have taken a little bit longer to get the wheel spinning to 10K. Probably 20-30 days instead of 12.

In my next post, I’ll explain why I started this 2nd account in the first place, and why more creators & businesses should consider doing the same in 2026.

See you then. Keep going 🫡

P.S. My goal is to help 1M business owners make $1M from content ($1T in market impact). If you’ve hit this goal, would love to hear from you to understand what I should cover next.