8
4 Comments

Experiment with Shared Ads on Twitter

We ran a couple shared ad experiments on Twitter in July and I thought you all might be interested in our findings. A shared ad is just an advertisement that promotes more than one business at the same time.

Twitter Experiment #1 - Optimize for Clicks

4 members each paid $80 and we created a single promoted tweet with info about each of them. The tweet tagged all 4 businesses and linked to a joint landing page we made which served as a launching pad for visitors to click through to their individual sites. Here are our takeaways from the $320 campaign, which we set up to optimize for clicks:

  • $320 spent (4 members each paying $80)
  • 292,411 total impressions
  • 9,710 total engagements
  • 3,783 link clicks
  • 892 likes
  • 120 RTs
  • Each member gained ~40 followers.

https://cohoist.com/assets/img/samples/sample-tweet-02-and-analytics.png

I was one of the members that paid and I saw a 46% increase in sales in July over June for a 67.5% ROI. It's a short experiment so more comparisons would need to be made to draw long-term conclusions, but even half or a quarter of that ROI is still a respectable return on the ad spend.

We were really stoked about the 292k impressions. It's neat how combining funds like this helped each business get their brand in front of 4x more people than they would have been able to if they only bought $80 in ads for themselves. It's a powerful way to build brand awareness and increase name recognition.

Twitter Experiment #2 - Optimize for Followers

We self-funded a much smaller campaign simulating what would happen if we promoted 8 businesses in one tweet. This tweet was a pay-per-follower tweet and here's what we learned:

  • Twitter only charges for the 1st action a user takes! Most people followed more than one account so we were able to get a super low cost per follower
  • $21.75 spent
  • 19,635 impressions
  • 814 total engagements
  • 232 followers (Each tagged follower earned about 25 new followers)
  • 1.81% follow rate
  • $0.09 cost per follow

https://i.ibb.co/7NrqSdv/sample-tweet-02.png

The fact that Twitter only charges for the 1st action is the big takeaway from this experiment. This is a huge advantage for shared ads since engaged users take more than one action, but the group will only be charged once. That's important because it means each business is earning engagements while reaching 4x to 8x as many people as they would if they had to fund a solo campaign.

For every dollar you put in, you get $4 to $8 in advertising with shared advertising campaigns

Think about it. Our first campaign spread logos in front of 292k people. Yes, they had to share ad space but the second experiment proved engaged users take more than one action per post so these are people actively seeing your ad. It makes sense. You don't see people tagging competing businesses in a paid ad so it kind of slips around their ad blindness.

For the second campaign, we spent $21.75 to promote 8 accounts. That means each member would have paid $2.72 and earned 25 new followers. Imagine what would happen on a full-scale $640 campaign!

By the way, if you're curious how we design the shared landing pages we point our ads to, here's a sample screenshot:
https://i.ibb.co/T1HGWcz/shared-landing-page-ppc.png

What's next for Cohoist?

We're trying to decide if we're going to run a full-scale Twitter follower campaign next month or if we want to run shared ads on Reddit or Facebook. If you want your startup included in our next shared ad campaign, join us at cohoist.com and leave a comment below to let us know what kind of shared ads you want us to run next. We'll be sure to pass along what we learn here!

  1. 2

    I am up for it!

  2. 2

    Sounds promising 🌲Brand awareness campaigns on Twitter are not usually economic unless you have $$$. Similar to writing a guest post, featuring on someone else's podcast, or participating in a promotional software bundle, I can see marketers forming similar alliances to promote their brand and earn more followers through Cohoist.

    Here are my thoughts as a potential customer:

    • To participate and attain maximum value, a customer needs 2-3 similar brands to advertise alongside. Finding a common topic around which to promote each account while also allowing each account fair exposure is tricky and is reliant on network effects (more customers means more potential value to each customer)
    • While I think the sentiment of "for every dollar you put in, you get $4" is generally OK, every marketer understands that multiple calls to action will inevitably dilute the conversion %. I would imagine that the account listed first in the list will see greater engagement once you reach a point of statistical significance. As a prospective customer, this makes me apprehensive 😬
    • I was an early customer of CodeFund by @coderberry which allowed companies to advertise on GitHub repositories until GitHub effectively shut them down through policy. Do you see Twitter intervening as a risk?
    1. 1

      Thank you! You're absolutely right - it really is difficult to have a cost effective brand awareness campaign on Twitter. They're certainly a lot more expensive than throwing up some display ads or ads on other social media platforms, like Facebook.

      Your bullet points were spot on. Here are some thoughts on them:

      • This is definitely a challenge we figured we would have early on with Cohoist. Until we get a large enough audience, we're going to have growing pains trying to group businesses in a meaningful way so we can do effective audience targeting. Thankfully our members have been understanding. We've been able to talk to the businesses that don't have a good match and pause their payments until we can give them their money's worth.
      • We have the order of the cards on the joint landing page randomized periodically so the 1st business listed doesn't get an unfair advantage. We actually thought about having an upgrade option where people could pay a little extra to stay in the top 2 or 3 spots but we decided against it. It adds a layer of complexity we'd rather not take just yet. That's a really good point about diluting the conversion %. I can see people thinking Cohoist is better for building brand awareness (getting your name and logo out to more people) than it is for converting individual users. That's probably a fair characterization.
      • This is good history and a great thing for us to keep in mind. I had no idea. Thank you for letting us know! I don't see it as a big risk on platforms where you can tag multiple accounts (Twitter, Facebook), but I do see it as a risk with Google display ads. So far we haven't had any issues with them, but they do have some policies in their destination requirements regarding bridge pages that I can see causing some trouble for us down the road. Hopefully by diversifying across multiple ad platforms, we'll be able to survive being booted from a platform or two!
      1. 1

        This comment was deleted 4 years ago.

Trending on Indie Hackers
Passed $7k 💵 in a month with my boring directory of job boards 65 comments How I got 1,000+ sign-ups in less than a month with social media alone 30 comments Reaching $20k MRR using Cold Emails only 21 comments I made a list of sites to submit your Startup 13 comments Email marketing done right: 7 tips for designing campaigns that drive results 10 comments Guide: How to get your first 10 customers 9 comments