TL;DR> If you are a band, musician or business guy/girl that dreams of a future where there’s a need to have revenue from music then you should absolutely take a closer look at AudioKite. However, if in your dreams revenue is not a factor, then leave AudioKite in the back of your mind, but don’t waste time and/or money on it. And even if this service will give you good results, there’s still the real world that will have to decide. So use it wisely, or be wise and maybe just don’t use it ;)! Read more on why I believe this is true below…
What is AudioKite?
AudioKite promises to give you “Market Research for Music Creators”. In short, it says it’ll survey your music to thousands of active consumers so you can learn about the strengths and weaknesses of your music, the attention span, what genres it fits into, with what brands you can assiociate it etc.. etc.. All this is brought to you in a single page report that you scroll, hove and click on – in a very friendly and nice UI environment. You can find all of this in more detail on their website audiokite.com – they even break down what the audience (demographics) is you’re sending your music to.
Is that really true?
To be fair, I heard about this platform through Brian Hazard of Color Theory, who wrote an article about his experience with AudioKite on his blog “Passive Promotion“, which dates from 2014 ~ roughly two years before I tried it out. It sounded amazing in that article, so I gave it a shot. However, things aren’t quite as they seem. As you’ll read below, I do have some remarks and points to make to state that it isn’t a tool for everyone out there. So before you even start I’d consider you to read up about the service and decide whether it’s something you feel you need to invest in with your music, or not.
My experience With audiokite
I signed up, chose their “Commercial Potential Report” with 50 listeners three times, coming at a price of $29.99/report. I uploaded three tracks of which I felt were varied enough to give a clear view per report of the range/spectrum of the audience at AudioKite and their reports. I went for a smothh, vocal heavy, electronic triphop like song called “Never Look Back ft. Ellia Bisker”. Next I chose for “Not What I Expected ft. Fallon Nieves” which is a more upbeat dark electro track, still vocal heavy but also relying on a lot of synth elements – no conventional radio material. To close, I picked “Wild Ride ft. Miss FD“, a typical big beat industrial-electro tune. Complete with distorted vocals, energetic beats and a big focus on the musical/electronic elements. Here are the links to the songs and the accompanying reports I got from AudioKite which I will discuss in this blogpost:
- “Never Look Back ft. Ellia Bikser” [ open report | stream track ]
Scoring a 7.0 on average, 96th percentile*, +0.6% from the AudiKite average.
- “Not What I Expected ft. Fallon Nieves” [ open report | stream track ]
Scoring a 6.5 on average, 71st percentile*, +0.1% from the AudiKite average.
- “Wild Ride ft. Miss FD” [ open report | stream track ]
Scoring a 6.2 on average, 49th percentile*, -0.2% from the AudiKite average.
If you’re seeing a trend in the scoring versus the genre and moods of the tracks, then that’s to be expected, at least for me it made a lot of sense. It’s what I predicted in terms of overall enthusiasm for one song against another – which proves to me at least that the AudioKite audience is quite predictrable – even though I tested it only once with three songs. It did match my expectations for this run.
*percentile: Generally speaking for AudioKite, the higher towards 100, the better. 96 would mean that for every 100 songs reviewed only 4 would’ve gotten a better review. This way you kind of know where you stand towards other songs uploaded on the service – which means 7.0 is a pretty high score. [ read more ].
The first thing that I noticed when I uploaded my “electronic” music is that I couldn’t fit it into a good category. I wasn’t able to drill down further into the genres of electronic music, as they were so limited. I also feared that when I’d put in “indie” people would expect rock music, or when I’d put in jazz/blues people would expect just that and nothing electronic at all. So tracks 2 & 3 (Not What I Expected & Wild Ride) were filed under “Electronic” and I took the risk with song 1 (Never Look Back) to file it under Jazz/Blues. I felt this was a weak point and that I were to be potentially judged by the wrong audience as I could not specify any further. On the other hand I though that this shouldn’t matter as people do browse around and can’t choose what plays on the radio either.
Thick elephant skin
Before uploading and going all in, take into consideration how well you can deal with criticism. If you get demotivated, down, depressed or generally unhappy with negativity and personal opinions/taste of individuals, then I’d advise you to let someone else do it for you – so they can filter out the rough parts – or just simply don’t do it all. However, I feel this is actually a good element in the AudioKite system. These unfiltered messages are the humanity, the voice of the audience that listened to your track. Raw and unfiltered, as no one would ever tell you when you let them hear your music. It’s honest, it’s sometimes brutal, but sometimes so sweet as well. I really liked this part of AudioKite – even though it doesn’t really add substantial value, it’s a human touch that adds an extra dimension to an otherwise numeric filled statistical report…
science & surveys
When doing a proper survey, 50 people per song (which I chose) is really not enough to get a good scientific representable measurement to work with. But if you have to crank up that number to get a good scientific sample size (taking into consideration the population, margin of error and confidence levels), you’d have to invest quite a bit of money. For example, if I start with 500 people per song I would’ve ended up with $249.99/song – quite expensive for just one track when you’re an indie band… And do consider that when you want to get serious with these statistics, 500 is probably the minimum you want if you want to be taken seriously when using the report to pitch to say promotors, radio, labels, etc.. You’d have to ask yourself the question then, what do I want to use this report for, and really weigh out the pros & cons.
Contradicting comments & metrics
The audience are allowed to place comments, as raw and fun as they are, they’re also pretty contradictory a some times… For example in the track “Not What I Expected” some really like the vocals, whilst some dislike them – same for the instrumental bit. There’s some screenshots of quite contradicting reactions below, after reading those I’m sure you’ll get my point.
Again, as I said before in this blogpost: even though they are contradictory and quite useless in the general stats, I quite like them because they are so raw and honest. It’s a strong point of the system that they actually let people voice their opinion without filtering it. It goes to show that everything is quite relative and one person can adore the song while another one simply hates it, and others just don’t care enough to even write a comment!
It was difficult for me to decide whether “Brand Associations” and “Movie and Televion Associations” were relevant to me, or anyone else. At least in my view and for my music, it was irrelevant how “Not What I Expected” releated to brands like “Apple”, “Gucci” and “Victoria Secret”. I don’t think this report will convince companies, advertisers or tv & movie to license my music?
The “Song Classification” was a bit weird to me as well. In here they evaluate “In what physical environment the song could be played”.. The track “Not What I Expected” rated high on a college party & skateboard park, as well as in an art gallery.. At least my song’s title has relevance to the results they show me here.
The evaluation for music outlets go from “None” (red – not shown here), “Digital Streaming” (yellow), “Terrestrial Radio” (blue), “Satellite Radio” (green), “Digital Radio” (lilac). If thie pie chart resembles the chart of every other song uploaded on AudioKite it wouldn’t blow my mind completely. As everyone knows digital streaming is very popular nowadays, ofcourse it will score high. I don’t see the point of including this metric, apart from filler material. It’s irrelevant to me at least.
I did like the metric “Regardless of how this song clas classified, what genre is the most appropriate”. All three songs I uploaded were ultimately classified as “electronic” and scoring high(er) on “pop” than I expected.
The fan profile seems like a good idea, and I generally like to see this reflected. It’s a sum of the other metrics shown to you, but broken down into a profile, for a listener that would either really like or really dislike your music.
I already knew this
Apart from that, a lot of things being asked I already knew. This might be the biggest let down for me in fact. I thought I’d learn something – and while maybe some people would say I did, I don’t have a feeling I tested enough people (see sample size / etc..) to really get a good indication.
Now this all feels quite generic and a bit predictable. I know what kind of music I make, and per track I can quite honestly predict who will like it and who won’t.
So, when tested, I already knew that I’m not producing radio-ready mixes with intros that are too long for radio play. Songs that are quite agressive and sometimes not suitable for national FM airplay. I’m also not aiming to have a “hit” song.. So maybe I chose the wrong report from AudiKite. it’s what I have them as feedback as well, I’m waiting for their answer on this – as I feel maybe I interpreted the intent of the report/measurement wrong. I could’ve learned more if I chose another report, or targetted it specifically to a genre that matched more with the audience that got to listen to it.
DO i run a business or do I make music?
In the end you have to conside whether your dream really is to make music for a living, or whether you live to make music. In the latter case, AudioKite can be fun, but is an unnecessary cost.
They all hate my band name. That doesn’t come as a suprise, I know this already, it’s the worst bandname .. “XYZ” would do better. But, I’m stuck with it, and it’s already out there with my other music and projects. It’s difficult to remember and pronounce, unappealing, but it’s out there, whether they like it or not. I can’t just change it now. yet every single report hammers that the band name is bad.. Without taking these things into consideration.
My end conclusion is a like my TL;DR, only a bit lengthier. If you wish to run your music like a business, then go have fun with AudioKite. Even though it’s limitation and expensivene nature VS scientific relvance, it does provide you some tools to analyse your music – and afterwards present those reports to the people you want to convince.
However, if you’re a band or musician that writes music without conventions, if you have no expactions on commercial succes and don’t expect to make a living off your music alone. Then I’d advise you to wait it out and not to check in. As the relevance and genre specific needs are not tight enough, and the expenses you have to make for a quality report (read: scientific relevant) are too high in my view. You’re also probably a bit too unconventional for national airplay and probably you don’t want to be compared to Rihanna, Selena Gomez or Lana Del Rey (not that I don’t like some of their music). Take your time and money to the studio, to the mixing and mastering, touring, instruments, .. and if then you have spare time & money, maybe look at AudioKite, if you really want to – but it’s not vital.
(PS: The website itself is very easy to navigate and work with – the reports are nicely made and shareable. The support team is highly accesible and responded to me within a working day. Quite happy with that ;))