Monday, 4 June 2018

Twitch - Otomo May Report 2018

Salutations and welcome to my 5th Twitch report of the year.

I had the absolutely brilliant idea of chronicling my growth on Twitch every month. You can check out what I did in April over here. But before we dive into what happened in May, let's give a short recap of what happened.

A brief history

Near the very end of April I was informed that my giveaways on a certain subreddit were wearing out their welcome. This is due to the fact that I would post a link to my twitch in these threads as a form of marketing. It's all very scandalous and you can read it here.

While the subreddit was fine with this when I started in February, a couple of users considered my near daily game giveaways to be spam. They reported this to the mods and the mods said that advertising of any form were bannable offenses.

This was quite a hit for my stream, I rely on my giveaways to supply 20-30% of my viewers a month and losing that was very unfortunate. And then there was another problem I had to deal with...

I do a daily steam giftcard giveaway on twitch, a $4 giftcard. I've been doing it for the last 2 months and it's worked out nicely. It's been a great way to give back a little to the people who support and watch me daily. 

Unfortunately, I had to stop doing daily giftcards in May. This is due to paying off some university debt. Compounding my lack of marketing with one less fewer reason to visit my stream added up to make May not as great as it could have been.

Still I had other strengths to rely on, like my charm and professional gaming skills. So let's see how much damage those did.

1- Average viewers.

During May I lost about 22% of my viewership compared to April. This is my first time losing viewers since I started documenting my progress. Despite the reduced average viewers, I think May was actually quite solid overall. 

I did only a handful of giveaways in May, so I expected some losses and they were fine. I was still at the same level of viewership as I was 2 months ago, I had a decent average everyday and it was a lot of fun.

All in all, it's hard to see it as much of a loss despite the dip in numbers.

2- Live Views

From the graph you can see that May had almost the same amount of total visitors as April. This was a nice surprise although there are a couple of important factors to remember. The first is that I had a very large (200 people) host, which helped a lot. There's more info on that host later.

The other factor is....

3- Streaming hours

I streamed quite a bit more in May than ever before, I've never streamed this much in a month, Almost 90 hours.

The extra hours of streaming helped make up for my lack of other activities. It also brings to mind how some streamers would stream for 6-8 hours a day, everyday. Those kind of hours are insane but do help a lot.

Still, if I had a choice between streaming for more hours or streaming less but with some marketing, I'd definitely pick marketing. Very long streams are quite tiring and it's far better to work smart than to work hard when possible. 

Games I streamed

There's a silver lining to every cloud, and in May I was able to get more accurate data on the value of some games. By cutting down on marketing, we can see each game's appeal to viewers. I will list the games below and how many followers they each have on Twitch.

It's very clear that even with my very solid regulars, some games are simply much better to stream than others. The picture below shows the fluctuation each game has.

Reminder though that every streamer and every channel is different. If you intend to stream these games don't expect the exact same results.

Battle Brothers (13k): I talked about this game in April and my thoughts on it haven't changed. It's a fantastic game with tons of replay value. I averaged 24 viewers and 42 browsing views during my 4 days of Battle Brothers, this is due to playing it a lot in April, which meant I had a sizable following for that game.

Slay The Spire (165k): I had really high hopes for STS, I would regularly be in the top 5 streamers for that game and yet... It really didn't deliver in terms of viewership. Averaging only 16 viewers and 40 browsing views for 7 days of STS. Neither my current audience nor passers-by considered STS a good game for me to stream.

RimWorld (163k): The biggest surprise of the month came from RimWorld. I hadn't heard much of the game before playing it and had little expectations when it won the poll. But boy did it deliver. Averaging 21 viewers and an insane 63 browsing views in just 3 days, RimWorld is an excellent streaming game.

Sadly I didn't quite enjoy my time with it and have put it on hold to stream in the future.

Bloodstained: Curse of the Moon (3k): Now before you think that I suddenly turned into a good streamer while playing this game, I must clarify something. I had a huge host on that day, around 200 viewers came from Sheepiihd. This skews the numbers heavily and distorts the data I have of the game. Still, despite having an average of 38 viewers and being the #1 streamer for some time while I streamed this game, I only managed 7 browsing views. So it might not be the best game to speedrun if you're aiming to grow a channel. 

Also, major thanks to my friend Terror for this game, it was a fantastic experience.

Axiom Verge (4k): Another game that I received as a gift and enjoyed immensely. Despite being a very good game, Axiom Verge isn't a good game for new streamers to try. It has a tiny community and though I was the only person streaming it for 7 days at a decent average viewers of 21, I only got 13 browsing views.

But it's a very fun platformer and I wanna thank my friend 514 for this game.


May felt like the first month this year that was "different". I had to refocus my efforts into new yet familiar things and approach streaming in a different way. It felt like a more back to roots approach for me, the same way I streamed last year where I didn't try to reach out to new viewers and focused more on streaming for streaming's sake.

It was liberating to just stream and not worry about anything else. I had a lot of awesome streams in May and some streams that I was able to sit back and think "What could I have done better?". I was also pleasantly surprised to know that if I ever stuck to streaming just 1 game, like Rimworld or Battle Brothers, I could probably make it as a streamer.

Still, I have goals I aim to achieve this year, and I won't reach it if I don't try a little harder. So I'm looking to make June my best month for streaming ever.

So stay tuned and check this blog month when I have the june update. Till then, thanks for reading and let me know what you thought of this month's recap.

Favorite Clip 

Trying to survive while panicking in Bloodstained was incredibly fun. 

Friday, 4 May 2018

Twitch - Otomo April Report 2018

Hello hello and welcome to my monthly Twitch report.

This is my fourth issue ever, I only started chronicling my Twitch journey since the start of 2018. Check out what I did in January, February and March. Now without further ado let's look at what I did in April and how much it helped.

Thursday, 12 April 2018

Twitch - March 2018 Report

Hi everyone and welcome once again to my monthly twitch report.

Last post I talked about the things I tried in February, which included giveaways and trying to reach new people on subreddits.

In March I continued doing the same along with a few new experiments.
March Experiments.

1- Dressing up

One of the things I've noticed about some big streamers is their decision to completely forgo their real life personality and trade it for a persona that they use while streaming.

It's a lot easier to say and do things you normally won't if you convince yourself that you're portraying a character, this opens up a lot of room for experimentation and appeal to multiple audiences.

I tried dressing up, donning on a couple of different outfits while playing the game Super Seducer. I thought it was a good time to debut these personas because Super Seducer is the type of game that relies on the streamer to be entertaining

Conservative Arab costume. Just missing the beard! 
I was going for suave but creepy. Similar to the Super Seducer

The idea of doing these sort of gimmicks to reach new people is worth trying. It gives you flexibility and shows your viewers you are open to ideas. First time viewers will also be impressed with the effort and would be more likely to come again and see what crazy thing you're gonna wear ...or not wear ;) 

In both February and March the amount of viewers who converted to followers was almost exactly the same (both at 5%). Bear in mind that the graph above also takes in account regular viewers. Twitch doesn't have a stat that shows ONLY new viewers but all viewers who come to a stream, so keep the above information in mind.

Overall the sample size is small (I only dressed up for 2 days out of 23) and the general volatility of Twitch means that you must do the same test many times to get conclusive results. Both days of dressing up gave me above average viewers at 23 and 27 compared to my daily 20. Great results but the sample size was small and more testing is required.

Conclusion: If you are playing something with slow gameplay or replaying a favorite game, try out a new personality. Spice things up a little and see how your audience reacts.

2- Momentum 

This isn't quite a scientific section but I think anyone whose streamed had those feelings of "wow this is awesome" whenever the stream is going great. This feeling permeates through the streamer and the viewers and is the difference between an awesome stream and an Ok stream.

This isn't something you can turn on or control, but what matters is when you feel the stream is growing and getting better, try to stream longer and more often to keep the momentum going. 

I should point out that I streamed a lot more in March than I did in previous months. I streamed around 40 hours in January and February, but increased that to 55 hours in March. I felt the stream was picking up steam and I had more energy due to being excited!.

This feeling of excitement wasn't just in my imagination, but the chat as well was more active. I had almost double the messages per stream in March as I did in February, and more than 3 times than January.

Conclusion: Try to be more flexible, if you stream 2 hours and a lot of newcomers show up in the last 10 minutes, go for another half an hour. If chat is being extra active and discussions are flying, then keep it going and don't shut down the stream. If more people are showing up to each stream then stream more days if possible. 

3- New chatbot and viewer polls.

I've been using Nightbot since I found out about it and it's been an extent. Nightbot allows you to make your own commands, which can feature links and some interactivity. I felt like there were more things I wanted from a chatbot and so I decided to try out Streamlabs chatbot.

The 2 benefits of Streamlabs that I was most interested in were the currency option and making your own sound commands. Since  my experiments in February I noticed great growth and met lots of wonderful people. However I did also meet people who seemed less interested in my content and more in my giveaways over anything else.

In an attempt to curb the problem, I added a currency system. Basically you get 1 coin for every minute you're in the stream and to join the giveaway you need 15 coins, you can also join multiple times up to a limit.

Adding a currency system opens up new options as well, viewers can spend coins to run special commands and vote on polls. This allows regulars to have more say on how the channel proceeds compared to first time viewers who may not come again. (Coins can be spent to vote for the next game to play or what action to perform in Telltale type games).

Conclusion: Got a favorite quote from a game? did someone clip your scream during a horror game? whatever it is you can think of, make a command for it and let chat spam it. It gives streams a bit more personality and that always helps.

March Numbers!

Click me!

1- I didn't have as much fluctuations in March as I did in February. February had some days when I was streaming to 8 people and others where I was hitting 25 average. 

2- The number of people finding my stream from just browsing Twitch was 3 times higher in March compared to February. While there is no one reason for this, it probably boils down to a few factors.

Since my channel grew, people didn’t have to scroll down the page as far to see me. The closer to the top, the more people will see the channel!

3- Followers are the biggest source of my views, but not by much. I was a little shocked to see that they only account for 20% of my views, it's a little hard to imagine where the rest come from though. 30% of my views from Twitch are either people browsing around or followers, the other 70% is largely unaccounted for since Twitch doesn't seem to track them.

Now let's compare the last 3 months and see how things are going.

 Average Viewers 

While March had slightly less growth percentage wise than February (which saw 50% increase) March was an amazing month for me, getting an almost 40% increase in viewership.

When streaming pay attention to what time people are tuning in to watch you. If you stream 4 hours but nobody shows up in the first 2 or last 2 hours then consider moving your times around to take advantage of this.

I stream at a sweet spot, 11 PM (GMT). It's a good time for my US audience to tune in as it's after work/school, but before bedtime.

New Followers

While the percentage of people who followed in March was slightly less than February, it was double in volume. I only started trying out new things in the second half of February, so perhaps if I had started earlier then we would see a smaller gap between months.

Either way, a lot more people in March were interested in a return visit and that's great news.

 Total Unique Views 

March saw more than twice as many people find my stream as February. Most of my views came from what I tried to do in February but I also posted a link to my stream using Twitter and on my company homepage. 

Games I played

In March, one of the things I started to pay attention to was how many people are following a specific game. If a game had a ton of followers, would that reflect on my viewer list? The results I got suggested that wasn't the case. 

1- Resident Evil 4 (163k Followers): I started the month with Resident Evil 4 and managed to finish it. With a lot of people following the game, tons of mods and nostalgia, RE4 gave me a lot of value for my time.

I recommended this game for streaming in February and maintain that it is fantastic for growing your channel. I played RE4 from March 1st and 2nd. 

2- Final Fantasy Tactics (16k Followers): I spent most of March with FFT, a turn based strategy RPG with a strong emphasis on the job system. Tactics has many of the same pros as RE4, they are both fan favorite games in very well known series and both are quite long in the tooth, giving them nostalgia points.

FFT is quite long though, sitting at around 40 hours easily, this is both a good and a bad thing, it's good because it removes the headache of picking another game so soon after starting but it also means not reaching new communities for the duration of the play through.

Still, minor gripes aside, Tactics was very helpful for growing my channel and I recommend it.

3- Gorogoa (2k Followers): An absolutely gorgeous game that uses art as a way to solve puzzles. I was very impressed with it. And it's the most creative and original game I've played this year.

However there are several downsides to Gorogoa for streaming. Firstly it's a 2-3 hour puzzle game with no replay value, once you finish it (in one stream probably) you have no reason to revisit it. It's hard to reach new audience when you spend so little time with one game.

Secondly it's expensive for it's price, at $15 for 2 hours of entertainment.

Overall while the game itself is wonderful, I don't recommend it for new streamers. It's worth it if you have a consistent audience who can appreciate the art and the puzzles, otherwise stay away.

Cringe: The Game!

4- Super Seducer (19k Followers):
Firstly, don't judge me, I played this game for the plot okay? 

Super Seducer is one of the saddest games anyone can play, it is just one one episode of harassment after another. By the time you finish an episode you'll feel like you need to shower.....twice.

There have been tons of reviews for this game so I won't make another one. What I will say though is that I streamed this game almost the day it came out and it.....wasn't so bad to stream.

With a lot of cringe and bored acting, the ability to make choices and zero gameplay, SS is ironically a really good streaming game. You can make polls to discuss what is the best option to tackle your date (in some instances literally) and it's fun to see what is "right" and what is "wrong" according to the game.

SS gave me the most active chats the days I streamed it, with more than 1500 messages everytime. People like to talk about it.

For $15 and 10 hours of gameplay, there are worse options than Super Seducer.


March was fantastic for me, the best month of the year so far. While I didn't try quite as many new things as I would have liked, just continuing what worked and adding a few variety streams really helped. 

Streaming to a bigger audience is tough though, there are more conversations going on and messages are easy to miss. I sometimes miss the first message a newcomer says and might alienate them but it's a learning process and I am adapting.

In April I intend to upgrade my PC to take advantage of more quality options, add a Green screen and test adding images to marketing posts.

If you enjoyed reading this or want me to discuss certain topics indepth, please leave a comment with your thoughts. Any and all feedback is welcome.

Favorite clip of March

I had a funny bug occur with RE4 that I enjoyed, check it out here. 

Friday, 30 March 2018

Twitch - February Report.

Hi everyone and welcome to my second post of what will hopefully be a monthly series on Twitch.

In my previous post I wrote about how I felt at the start of the year and how January wasn't a great month for me streaming wise. I'm happy to say that February went far better for me due to a great combination of new experiments and a solid selection of games.

Thursday, 29 March 2018

Twitch - January Report

Hi guys!

I decided that this year I want to make a monthly blog post about streaming, it will focus mainly on what I am doing every month, the things that worked and what didn't along with future plans. In the posts you'll also see the same monthly statistics that I see to give you guys more information.

But before all that, let's begin with a bit of background

Background about the streamer

Since I am writing this as a sort of reference/guide for myself it's important to share where I was at the start of the year, so that we can establish a sort of safe baseline to see how growth occurs with certain games and experiments. It's no use to say " I averaged 15 viewers in January, massive growth" when we don't know what I was getting before.

Firstly, I am a Twitch Partner, I write this not to brag but to add some legitimacy to this post. A lot of "how to guides" are written by people who streamed for a week and have 2 average viewers.

Secondly, despite being a Twitch Partner, I am a small streamer, I averaged 11 viewers on November and 8 in December with about 40 new follows in those 2 months. I have a handful of really dedicated friends who are regulars on my channel, they account for most of my viewers last year and I am intensely grateful to them.

Now you know what my streaming situation is, Partner, handful of regulars, minuscule growth. So let's talk about every month of 2018 in detail shall we?


(Cuphead, Okami, Evil Within)

When this year started I felt a lot more driven to invest more into Twitch, I felt like I had slowed down a bit after becoming partner. Many people aim to be a Twitch partner and when I achieved that (sorta by accident by the way) I just started it in a bit. That definitely took a toll on my creativity because in my head I was thinking " I'm a partner, why the hell do I need to put in more effort?" I had thought that by simply being partner eventually all the billions of viewers and millions of dollars would come.

I was wrong.

Turns out that being a successful streamer doesn't end at being a Twitch partner, it's simply a rung on the ladder you have to climb to get to the top, and sometimes being partner is a very low rung.

Unfortunately I can't say I did much new in January, I wanted to but I just didn't know what to do or how to approach anything, so in the end I just did more of the same and that resulted in one of the worst Twitch months for me.

The only gimmick I tried in January was to add a "word of the day" to my title. Shortened to WOTD, everyday I would have a rarely used word to my title, words such as Nefarious, Sporadic and such. WOTD didn't generate much buzz but a couple of viewers mentioned that it helped them learn new english words (shockingly not everyone is a native english speaker) and a friend reminded me to change the WOTD when I forgot. Overall it was a positive thing to add but it didn't really change much.

But let's move on to more factual information.

Let's talk numbers!

A little Information should be added here, firstly total unique views is a bit misleading since I believe it is bugged to not count some viewers, I could be very wrong here though. Either way Socialblade says that for January I had 887 views, slightly higher but not a major deal (for January).

The games I played had a pretty obvious role in my views (though the differences were not extreme). I started with Cuphead for the start of the year and the average viewer was at 10-12, I then switched to Okami for around a week and got around 7-9 viewers, I then swapped out for Evil Within and saw healthy growth of 10-13 viewers per stream.

This leads me back to my original assumption that for a variety streamer, it's almost always best to stream NOT popular games. Cuphead was a major hit when it came out due to it's art and difficulty, it brought many big streamers who raged and caused highlights on youtube.

Growing a channel using Cuphead even 2 months after it's release was not great, there were still plenty of streamers out there playing it in January.

Okami was an even worse experience, it was released in the middle of December and playing it in January just meant I was 1 out of 200 streamers playing it. I also didn't really enjoy the game nor do anything interesting. Okami definitely hasn't aged as well as some games, with somewhat sloppy controls and no real challenge. It's still an amazingly artistic game but sadly it never appealed to me and that showed on stream.

Evil Within was the last game that I streamed in January and it easily was the best. A horror game with a good mix of action and crafting along with a solid story meant that there was a lot to like as both a streamer and a viewer. I also had much less competition since the game came out late 2014, meaning it's been done by all big streamers and very few people still streamed it.

Overall January was a bad start to the year, I finished Cuphead but didn't benefit much from it (viewerwise) I quit Okami a week in due to not enjoying it or growing from it. I did pick up a little in the end due to Evil Within but I as a streamer hadn't done anything unique or special for that growth to occur.

Still, while January was bad, I did try out new things with great success in February.

Favorite Highlight

Here's my favorite moment from streaming in January, beating the final boss in Cuphead after about an hour of grinding (Casino boss was much harder).