Our First Twitter Ads Test (and a weird Twitter Bug)

HAPPY NEW YEAR! It’s 2019 and our new year’s resolution is to double the number of customers we have (We’ve a family trip to Disneyland Tokyo planned). This means looking into alternative sources of traffic and leads, no more can I rely on “witty bantz” from Twitter or going to BrightonSEO and shouting at random people that they’ve not bought my product (this has worked in the past but it doesn’t scale) No, not this year, this year is the year we do some PPC properly. Facebook, Linkedin, Twitter, Instagram, Google, Bing, PornHub… If there are ads available, I’m gonna give it a go. So we started with Twitter, everyones favourite way to avoid ‘proper work’ and invest in a social profile. I’ve come up with a bit of a campaign (that I’m slightly proud of) that explains clearly some of the issues we solve with LW
https://twitter.com/littlewardenapp/status/1081211253572407297
I’ve created a series of about 20 of these all with different scenarios and photos (all properly licensed of course!) So when you click to see a Tweets stats, you are given the option of “Promoting a Tweet” to a wider audience.
Once you decide to promote using this method the only options given are location and budget.
So either Twitter;
  • Will show it to EVERYONE in that location the tweet (not good)
  • They will use some Machine Learning fun to work out the best people to show it in that location (maybe good)
I knocked the money spent down to the smallest amount £50 and the campaign was on the way.
As you can see, at the start of the campaign, the organic stats for the tweet were ‘okay’ considering it was Christmas week and a Friday, not our best tweet but not bad. I updated twitter throughout the day with the spend and impressions.
https://twitter.com/littlewardenapp/status/1082271786807541760
When I saw we had two likes, I was interested in the people that this campaign was targetting to our like stuff so I popped over to the tweet and saw…. no new likes. Twitter as part of the campaign was taking credit for likes we ALREADY had.
Twitter Likes Before Campaign Started
Twitter Likes After Campaign Started
You can see our blue organic likes have turned into yellow promoted likes, this is massively worrying as without a valid explanation it means I can’t trust any of the stats that I get from Twitter. I asked a few people who I know use Twitter Ads but none of them had seen anything like this before, so we kept the campaign going but kept an eye on the stats. The final stats of the campaign and what we actually got for our £50 budget.
  • Impressions – 22,854
  • Media Engagements – 164
  • Detailed Expands – 45
  • Profile Clicks – 4
  • Link Clicks – 3
  • Likes – 0 (Allegedly 3)
  • Retweets – 0
  • Replies – 0
  • Followers – 0
  • Conversions – HAHAHHAHA
So not the best use of £50 out of the company Kitty but it was worth a try, before I put anymore money into Twitter Ads though, I wanted an answer on what happened with the likes, so I contacted Twitter Ads support.
Well I tried, but you wouldn’t believe it.. the issue box only allows 280 characters! So I tried again with a much more ‘direct’ message. I received an instant auto response with a rather worrying line embedded in the middle of it.
They want me to reply with more info… but they didn’t let me give any info in the first step which feels like a massive waste of my time and an annoying way of them filtering out queries. And even if I do that, I might not even get a response! Thanks for your money but we are too busy with other customers. I replied to the email and we will see if we get a response. Next up is Facebook. UPDATE: Twitter have gotten back to my detailed question (with images and dates) with this response that basically treats me like an idiot.
Not helpful at all, I have replied to this email with a less than happy tone. SECOND UPDATE  That’s it, as far as Twitter is concerned they don’t need to answer my question. As far as I’m concerned this isn’t good enough
If you’ve enjoyed this post, why not signup to our mailing list or even better BUY THE BLOODY PRODUCT.