How to engage your audience (aka Instagram is kicking my butt right now). - Part IV (the final one!)

Over the past few weeks, I've been sharing my progress on experimenting with quite a lot of different things on my Instagram account - after I noticed a massive drop in engagement, presumably because of a change in Instagram's algorithm.

If you're new to this series, you can read the first three parts here before diving into Part IV - the final part!

A quick note: As this is my last post in this mini-series, I would love to hear your thoughts in the comments below - what did you like, what didn't you like? And: Would you like me to do more experiments - and on what? 

Now, during the last two weeks - since my last progress report - I tried to get back to a normal Instagram routine because I figured running the experiment forever and ever would a) get boring, b) not get more conclusive and c) not be manageable anymore! After all, there are only so many hours in a day, and I desperately wanted to find a way to back to an Instagram routine that worked for me AND made my posts visible.

Before we dive into what I did, though, and talk about what I've learned from this experiment, let's take a quick look at the numbers!


Much like in the previous weeks, my follower growth rate fluctuated around 1% per week for the past two weeks. I went from 3,534 followers on June 6 to 3,633 today (June 19) - absolutely in line with everything I've seen on my account in the past few weeks and months.

Follower Growth from June 6 to June 19

Follower Growth from June 6 to June 19

Like History:

Moving on to something a bit more exciting! Like history - the average number of likes I got on all my posts in a day - fluctuated a bit during the course of the experiment.

We started off with a benchmark of 500 from February, and an initial average number of likes per day of 350 at the beginning of the experiment. In the first week, this increased to 400 before decreasing to slightly above 300 in the following two weeks.

Today, I'm happy to report that we're back to old strengths - almost 400 likes per day on average! If we took out today (where I only posted once until now) we would even be above 400. So something seems to be working better these past two weeks than the weeks before!

Like History from June 6 to June 19

Like History from June 6 to June 19

Comment History:

The comment history is slightly less exciting. During the course of the experiment, the average number of comments per day has always been around 20 - and the past two weeks were no different as you can see below. 

We even see the same pattern: One or two days per week generate a lot of comments (usually on a day where I post something controversial ahem...) while the rest of the weeks stays around the same.

Comment History from June 6 to June 19

Comment History from June 6 to June 19

Love, Talk and Engagement Rate:

You can also see these developments in like and comment history in the Love Rate and the Talk Rate:

While the Love Rate is up by more than 10%, the Talk Rate decreased by almost 10%. Now, if you look at overall Engagement, this is up by almost 10% - which is a good thing because it means that the increase in Likes we saw is not only driven by the follower growth we've seen, but by people actually liking more of my posts. 

What did I do differently?

If you remember the last progress report on the Instagram experiment, I was a bit bummed. The tactics I tried before that post didn't work at all - quite to the contrary, they actually seemed to decrease how visible my posts were.

So at the end of the last post I announced that I was going back to what seemed to have worked in the first week of the experiment - and that really did the trick!

Here's a quick overview of what I did on my Instagram feed in the last two weeks:

  • Posting three times a day. This seems to be the main driver for increased likes - which totally makes sense, if you think about it. But! The Engagement Rate shows that it's not only a maths things (the more you post the more likes you get), but that people actually engage more with my content. I think it's possible that Instagram shows your posts more often if you actually post more. I'm sure there's a threshold to this - posting 15 times a day won't get you more visibility most likely - but for now, three posts a day seems to be a good spot for me so I'll keep it that way!

  • Taking good photos. I'm happy to say that I haven't used a single stock photo since I started my experiment. I'm trying to get better at photography because good pictures clearly matter (my crappy collage I shared last week clearly didn't get a lot of likes...) while trying to not be too monotonous in how they look like. Still working on this, but enjoying the process!

  • Writing more like me. I've seriously enjoyed this SO much. Getting back to using my own voice on Instagram and not being stuck on "sales-y" copy writing has proven invaluable. I like to think that this also drives engagement, but alas, you can't measure that, so I just choose to believe that it does. Plus, it makes me happy, so there's that. 

  • Switching up & using less hashtags. This is probably the one thing that made the biggest difference (in addition to posting more) compared to the two weeks before: I went back to using hashtags on every photo, but I switched them up. I chose to go with 6 to 12 hashtags per picture, depending on how I felt like, and always try to incorporate a few more prominent ones (like #knittersofinstagram) and a few smaller ones (like pattern or KAL specific hashtags). 

  • A note on video: While I wanted to experiment more with Live Video, I sadly didn't get around to doing it. I had planned on doing a Live Q&A for my new Creative Boss BETA Instagram Mini-Class to promote the BETA sign ups, but as they filled up super quickly, I killed that. Still, I do see an uptick in engagement on the days where I share a lot of Stories, so I think there must be something to Stories having to do with how visible your posts are. That being said - and maybe that's just confirmation bias from my side - I think it makes a difference if you show your face and talk to the Stories audience or if you just post pictures on Stories. Not totally sure about that though - would love to hear your thoughts on that!

What I learned from the experiment:

  • Switching things up is fun! As is doing these kinds of experiments and reporting back on them. I think I might do that again! Is there anything specifically you'd want me to experiment on? Please do comment below!

  • It's important to distinguish between follower growth and engagement. While engagement - both in forms of likes and comments - seemed to be impacted a lot by my experiments, my weekly follower growth rate didn't change at all over the course of the experiment. I think this is super interesting! I had always assumed they were correlated, but they don't seem to be. I need to do some more thinking about why that might be the case.

  • Being authentic is important. I had somehow lost my voice (and my joy in IG) before the experiment. Doing this forced me to be more like myself and less focused on what people might want to read and this is something that was so valuable. In this particular growth stage of my company, it's so tempting to do something because of the money, and not because it feels right, so taking my Instagram back to its roots and just being myself was not only great for the feed, but it was great for me personally.

  • Take good photos. Quality nerd wins over lazy ass. Clearly. Forcing myself to take pictures and not use stock photos anymore has resulted in more engagement (this is clear to me) and me needing to step up my photography game. My feed is not yet there where I want it to be in terms of visual consistency, but I'm still learning, so that's okay.

  • Hashtags matter - but only so much. If I had to give a recommendation regarding hashtags, I'd say this: Use them on your photos, but not too many - tops 12 - and switch them up regularly. 

I hope you enjoyed this mini-series on my Instagram experiment! I would love to hear your thoughts in the comments below - what did you like, what didn't you like? And: Would you like me to do more experiments - and on what?