Most of the past few years working on Mango Display has been around trying to build a product people want. I have always loved this reference by Brian Chesky of Airbnb – and wanted to focus on getting to at least a 100 people who loved our product.
We found them one at a time. Some by search and some by outreach on posts and reddit. They helped shape the product and direction that it is today. We now have many multiples of that 100. Mostly grown by organic word of mouth by our customers.
In our current world of subscription fatigue, to see a review with this statement below is the best compliment we could ever receive 💚
“One of the few things worth the subscription price in our household.”
Mango Display customer review on Amazon
Scaling to keep up:
Our customer growth increased massively last year and our service level and uptime took a hit towards the latter part of last year, as our services could not handle the increased load. This was unacceptable to our customers and us. So this year’s (2024) Quarter 1 plan was a focus on stability and reliability.
We executed and completed the following:
- Portal bug fixes and standardized auto reconnection of live web socket updates (many displays run 24×7)
- AWS EC2 instance upgrades, load balancer optimization and autoscaling in place
- MySQL 8.0 database upgrade done, using blue/green deployment zones for the first time!
- Github Code branch clean up and standardization done.
- Auto deployment of code via Jenkins pipelines from Github done.
- AWS Cloud Front compression and caching configured for the webApp.
- Many, many major and edge case bugs 🐛🪳🐜 have also been squished.
We also managed to release two huge features that users have been asking for – ‘Templates’ and ‘My Files’. These have made a huge difference in simplifying the user experience, improving customer retention and lowering support questions.
Marketing execution:
Along with the plans for some more big feature releases on our roadmap, I now plan focus more on paid marketing in the next 3 quarters of 2024.
We have done some paid marketing, but it still is less than 10% of where our new customers come from. I believe our product feature set and quality is now at a stage (finally!) where adding paid acquisition to fuel increased growth will be an interesting stage.
Which brings me to Customer Acquisition Cost (CAC), Life Time Value-gross margins (LTV), Return on ad spend (ROAS) and all the other fun acronyms and metrics of getting the marketing process optimized to get our product to the right customers and keep the growth pipeline flowing.
We currently have paid ads running on both Meta and Google platforms. And both are performing pretty well, returning a LTV:CAC ratio of 5:1. This is a pretty good number (Put a dollar in and get five dollars out (over the LTV timespan) . Except that running it on a small scale does not guarantee that scaling ad-spend budgets will scale at the same ratio. In fact in our research and our own tests, the ratio deteriorates as both Meta and Google platforms deploy all the capital allocated for that period without forcing the optimization for conversions.
There is also the issue of true attribution. A users sees the ad Facebook, checks it out but does not buy yet. Then remembers the name and later uses google search to get to Mango Display and converts to a paid customer, and both Facebook Paid ads and Google organic search attribute that conversion to their platforms 😅
So it’s time work with some more AI agents to create, improve and track the creatives, hooks and target demographics and set a feedback loop with some detailed analytics to put the marketing machine into overdrive on both the organic and paid side of equation. I have some ideas 💡 and will work through them and maybe update with the results at the end of year.
What I realized, though, is that as much fun as product ideation and development is, B2C marketing – it turns out – is really fascinating too!
Dave's Dharma 