AdQuick is marketplace to buy outdoor ads (billboards, transit, etc). Customers include Lyft, Instacart & more. Rails/React stack.
At AdQuick, we're building the Amazon of Advertising. Amazon started with books, and we're starting with billboards. But ultimately we want every single marketer in the world to use AdQuick to do marketing & advertising, across many different verticals, and we intend to be the most customer-centric marketing tech company on the planet.
Here's some info about us:
- We've raised $1.1M, led by Garry Tan & Alexis Ohanian (Reddit co-founder).
- We're ex-Instacart, based in Venice, LA and growing very quickly.
- We have repeat buys from Lyft, Instacart, Peloton & more.
Why Our Customers Love Us
Our customers are often spending $200k just on one purchase. For them to build targeted, measurable advertising campaigns, their Chief Marketing Officer has a few options:
- Hire a team of people to coordinate across ad spend across multiple channels. Billboards, buses, Facebook, Google, Radio, TV, Reddit and the list goes on.
- Go to an ad agency which charges 10-15%.
- NEW: Use AdQuick to do all of this better, faster, and cheaper, for just 3.9%.
Right now we just handle the outdoor portion, but you see where this is going :) .
Our Technical Challenges
- We have a complex frontend user interface. Users should be able to view 100,000+ markers and transit lines on a map without their browser crawling to a halt. We need to support complex and interesting map visualizations – layering political voting data, Census data, Foursquare data and more.
- We're a search company. AdQuick conducts a search every time the user pans the map. Searching by geography, demographics, AdQuick score and a myriad of other attributes from disparate data sources is computationally expensive. It will be a challenge to scale while keeping search response times and server costs low.
- We're a data science company. We integrate with Google Analytics & AdWords and other data sources to measure impact of physical advertising on online behavior (CTR in a region). ROI data from prior campaigns can help create more effective future campaigns. As our data warehouse grows, crunching this data will be increasingly challenging and more powerful. We'll use machine learning to create better campaigns overtime.
- Our backend systems have to deal with a myriad of inventory CSV, PDF formats, requiring robust and fast file validations, normalization and ingestion systems.
- We scraped Instagram and used image recognition to see if people shared photos of billboards. Turns out, a lot of people Instagram'd Drake's board. Innovative ROI tools are a core part of our offering.
If you want to learn more, here's some reading you might find interesting:
How we used image recognition to measure the online reach of the billboard we sold to Drake (yes, the rap superstar).
Our $1.1M seed round funding announcement on TechCrunch
The slow journey to AdQuick launch