What you need to know about Book Trailers. [Part 2 of 2]

image

A little while ago, I created a trailer for my book, The Scissors and the Sword. Many authors create book trailers or pay companies/individuals to create them, and it was one of the last promotional methods that I had yet to explore. I’ve since been asked about how I made it. Additionally, most authors don’t provide figures for views/ads, so I thought I’d do that too.

[Part 1 can be found here]

This part focuses on my personal experience with the book trailer I created using the methods described in the prior post, including expenses and figures. Read on for detailed info!

 

The Trailer

If you’ve read my previous post, or have followed my blog for any length of time, you’ve probably already seen this – but for those of you that haven’t, here’s the book trailer:

Having seen that, let’s have a look at cost.

 

Trailer Expenses

I was fortunate in that I already possessed a computer with iMovie, and I was able to download Audacity (a free, open source audio editor) to help master the sound – so in terms of software, there were zero expenses.

image

Within the approx 80 seconds that the trailer runs, there were the following assets to account for (all prices are USD, and I’ve rounded a little to make the figures a bit easier to process):

  • The piano sound for the logo – $10 (AudioJungle)
  • The main music track – $10 (AudioJungle)
  • The voice-over – $10 (Fiverr)
  • Collected video clips – $200 (VideoHive), averaging at $9 each

The first version I made used many clips from Adobe Stock; these were amazing quality but I had to replace them as they pushed the cost well above $500, and $200 was supposed to be my budget. To economise, one of the longer clips (of London) was a video I recorded myself, on a train journey. I did try others, but unfortunately this was the only one I was prepared to use.

This gives a total cost of $230 / £180 / 200Euro.

 

Once it was up-and-running…

image

As soon as the trailer was on YouTube and had processed, I waited until that evening and launched it. I posted to my social media channels (mainly Twitter & Tumblr) with whom I have a collective follower count of around 2,500; not an enormous number but it was enough to get some initial views. I also posted it on Reddit, in ShamelessPlug, BuyMyBook and I also started a discussion thread in SelfPublishedWriters as to whether book trailers were of any use.

Soon I had views coming in, but it didn’t all go brilliantly. I’m a big gamer and I frequent the PSVita Reddit, so I tried hosting a giveaway of PSN cards there to help with promotion; this was deleted by a mod, which I guess was fair as it was off-topic.

Additionally, I paid 5$ to someone on Fiverr (who will remain nameless) who was going to set up some backlinks and such to improve the video’s SEO… But all it did, from what I can see via YouTube’s analytics, was get me a few hundred views from some far-Eastern nations such as Vietnam (via bots, I assume).

image

Fortunately all of the suspect fraudulent views came in over a short period of time, from mainly non-English-speaking countries – that makes them quite easy to filter out.

10 days in

The video has now been up for 10 days, and has had approximately 2,000 views. Filtering out the sudden influx of views from Vietnam etc., I estimate the video has been seen around 1,000 times. The video also has no dislikes and 6 likes.

image

In terms of genuine traffic, the views were dominated by those who came via Reddit, who contributed over 69 views all on their own. There-after it’s a mix of sites.

image

This is a closer look at the initial spike. This is just from me telling everyone I could about the video via various channels.

Once the initial rush slowed down, I turned to YouTube ads.

Google/YouTube Ads

With a video on YouTube, it’s very easy to use Google’s interface to turn that video into a YouTube ad; one of those ones which comes on just before a YouTube video of a similar topic. I set up my video with two campaigns…

  • Regions – United States and India*
  • In-Stream ads and discovery ads
  • Marketing to all ages, people who identified as “avid readers”

*I picked India because it has the highest number of English readers of any Amazon region, even if it’s a second language for nearly all of those readers. As views only cost pennies, I thought I’d give it a try.

image

The first peak contains the “natural” views, with the bulk of the “bot” views removed. The other peaks are from YouTube advertising. The slither over the top consists of views from other sources.

Having already spent a fair amount of money on the video, I thought I’d just put a small amount into this, so collectively I’ve only spent £16. However, this was enough to get over 500 people to view the advert.

Note: YouTube only counts an ad “view” when someone watches the video for over 30 seconds. Any less than this is just considered an ‘impression’. For those who are interested in such things, my video has an approx 20% ‘view rate’, which means 1 in 5 people keep watching it once they’ve seen the start. This is actually quite a good ratio; apparently anything over 5% is considered good.

As you can see…

image

I invested about 30$/£16, which yielded around 550 views – which was around 6c/3p per view.

image

This is a closer look at the USA campaign (the two campaigns are quite similar in terms of their figures, which was interesting – India was slightly cheaper per-view but overall the difference was marginal).

The most interesting figure here was the “view rate”. The upper row of in-display ads are the ones that show up adjacent to a video on YouTube – ‘suggested’ videos, whereas the lower row were the “in-stream” videos that appear as usual before a YouTube video (the ones you can skip after 5 seconds). As you can see, after a few days and thousands of “impressions”, the suggested ads only yield 1% of views whereas the in-stream ads yield about 20%. This is a much bigger difference than I was expecting.

Unfortunately, though, I can’t see what percentage of people actually clicked on the video to go look at the product on Amazon. This seems a vital bit of information; I’m not sure why this isn’t provided.

 

Now for a bit of maths…

image

My Kindle version costs £1.99/3$, and I get 70% of that from a sale after Amazon takes their cut (I’m purposely ignoring the paperback for now, for simplicity’s sake).

As we’ve observed, £1.30 buys me around 50 views of the video (though I don’t know how many click-throughs to Amazon that yields). This means I would need to get 1 sale per 50 views, or 0.5% conversion rate.

I can tell you straight away that I didn’t even approach this.

 

How did this affect sales?

Unfortunately, this is the bad part.

image

1 copy has sold.

However, given my typical sales patterns, this could have sold anyway, even if I had made no trailer, as I sell a few units every week anyway, with no advertising.

Conclusions

So, the obvious conclusion is that this wasn’t a good marketing method. I mean certainly, the video wasn’t perfect, and I wasn’t shopping it out to the biggest audience, but despite that, it was a fair wedge of cash for a reasonable ad that produced almost zero return-on-investment. The figures are so stark that it seems unlikely that changing some minor elements of the method would improve the results by several orders of magnitude.

I don’t feel too bad about it, because I believe you need to try these things in order to understand them – otherwise you never try.

Would I make another trailer, for a future book? I have mixed feelings about that.

Financially, hell no. I see zero reason to pursue it as a money-making endeavour.

On the other hand, though, ‘brand awareness’ is more difficult to measure. In the future, if I have a channel full of trailers for different books, that might lead to new cross-linking and metrics that I can’t really predict. I also enjoyed the process and I’m proud of what I produced, so it isn’t wasted time so much as exploring an option that didn’t yield the desired results.

Of course, the trailer is going to remain up on YouTube; who knows how many views it will garner in 6 months?

This is Lilith K. Duat’s trailer for their novella, Balance in Chaos. This was notable among the examples I’ve seen because Lilith’s trailer was practically free, made with a tool called Animoto. I might do something closer to this next time.

The main lesson here is not to try and “punch above weight” and produce a high-end trailer for an indie book project, until I have a bigger readership. I would probably do better making a cheaper, 30-second trailer built up from images and spending a larger amount of money on promotion than trying to make a lavish trailer.

Sadly not the happy ending I was hoping for! Still, let’s see how things go next time!

Follow me either here or on Twitter for more self-publishing info!

One thought on “What you need to know about Book Trailers. [Part 2 of 2]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s