This entry is about advertising in Facebook. Like my readers know I have created a word game called Word Slinger for Android. It’s published in Play Store and even though it was more like a learning project rather than a serious attempt to create something magical, I would like to see if other people enjoy the game and probably learn a bit more what could be improved.
So for that purpose I need players and getting players is very difficult for an indie developer. This is something you will encounter always when you search how to get visibility and/or players. There are some sites that do reviews but some are utterly expensive and the free ones get massive amount of requests for reviews you need to be very lucky to get reviewed.
My brother and his colleagues created also a game and they utilized Facebook ads. Having heard about the positive experiences (combined with the earlier knowledge, thanks Milan) I decided to give a go.
I wanted to keep my expenses very low just to learn the mechanics and see how it goes. Creating an ad in FB is relatively easy. I say relatively because there are some things that can go wrong. It’s easy to select your target audience by selecting the country. This selection has the biggest influence on the suggested bid. Suggested bid is the amount of money you are willing to spend for certain action your targeting. To simplify this: If you want to get downloads for your app with fixed amount when any user clicks the ad then the max bid is what you are ready to pay for the click. It’s more complex than this but for simplicity I leave it here.
There’s a catch too. If you set the bid too low you are competing against the other advertisers and your ad is never shown if the bid is too low. I have run so few campaigns so far that I have just used the suggested minimum bid as my max bid. Occasionally this is enough but sometimes you need to raise the max bid.
There are other criteria to be selected like age, sex, interests and so on which all effect on the amount target audience but also to the bid. I have decided to target those FB users that I feel are matching my possible player base. This would be something like people of certain age, interest to word games and so on.
You also need to select the ad images. The images are FB cover photo size (1200 x 627). If you are running a longer lasting campaign you probably should select several images so that images are rotated. This doesn’t prevent the same ad to be shown twice to same person but it reduces the chances. If frequency goes a lot more above 1.00 then you can think of adding another image.
The real challenge in my opinion comes with the image approval process. The images should comply with FB ad photo criteria which state that images should not contain more than 20% text. This is, well, interesting to say the least. When you create the ad, there’s an automated process that approves your images (at least what I know of). So even though your image might have more text it will get approved. The problems might come when your ad goes live and starts to deliver.
When the ad goes live it can get disapproved all the sudden, in the middle of campaign. I don’t know if this is done by a person or another process but this has happened to me 4-5 times. While FB support is very friendly and willing to solve the issue this can be problematic because of time zone difference. You might have to wait up to 20 hours for the reply and you can basically say goodbye for a planned weekend campaign if you have targeted far east like I did.
My product is a word game and it contains a lot of words. I thought I would have to create some sort of ad without words in it but according to FB you can use screenshots from your product. This is good to know and luckily I have managed to create ads with the screenshots. There’s still one remaining issue. When ever I create a new ad with the same images they always get auto approved and when they start to deliver they get disapproved. The only thing I currently can do is to contact FB and wait for another day to get them manually approved.
It would certainly be very interesting to hear any similar experiences or solution to this problem. Please, share your thoughts and ideas with me!