Inspiration

While walking past the streets we are overloaded with information. Nice posters of various concerts, parties and cultural events often capture our attention, yet we miss to remember all of them. We often took a picture of the poster and then totally forget about it.

On the other hand, the effectiveness of printed ads could be hardly measured. There is no real interaction with passersby except for several seconds of their attention if poster ad managed to attract them by visuals or content, despite their busy walk and thoughts.

Solution

An event discovery app that enables audience direct action (setting a reminder for the event, registration and purchasing tickets) while it measures the impact of offline (printed) ads.

How does it work?

App enables passersby to snap a poster or a banner, crop the photo to reduce all the background noise and leave only the name of the event. After a cropped photo is being uploaded to the app, user gets a list of results - various event links (official web page, Facebook event, Eventbrite posting etc) and selects one of them, upon which he can set a reminder for, directly purchase a ticket and discover similar events. That way he ensures he doesn't miss a great event.

Snapit also enables users to discover events nearby they might be interested in.

The app communicates with a user is in form of a chatbot called Snapson. Snapson returns results and reminds user of an event.

Value proposition

  • For user: Never miss a great event in your city again
  • For event organizers: track and measure engagement with the potential audience and get insights on the user profile interacting with your printed ads
  • For ticket selling platforms: new channel to reach customers that seen printed ads and also measure their engagement

How we built it

To build the app we used the Node.js platform in order to combine javascript to asynchronous requests from the Facebook Graph API. We extracted the event details by implementing the Optical Character Recognition (OCR) Function of the Google Cloud Platform in order to create a legible stream of characters that would feed the source code.

Challenges we ran into

  • when the cropped image of text is being uploaded to OCR by Google Cloud we still have to optimize the cleaning up the noise around the text that could in cases of abstract typography that partially or fully prevents text recognition.
  • assigning tags/keywords to events that are being suggested to a user, to ensure app suggests him relevant events

Business Model

Our customers and partners are companies that benefit from our app and pay us per click that direct our users from app to the companies’ page. Our monetization method will be affiliate marketing. When SnapIt bot returns the results in form of the event link, it shows to user a 'Buy a Ticket' option, that directs the user to the event registration and ticket reservation platforms such as Eventbrite, Ticket Master, London Theatre Direct, and others.

What makes SnapIt unique is the form of a customer journey “from picture of an offline ad to event booking”

Payment models: CPA and CPL/CPC. CPA - Cost Per Acquisition (fee payable upon completed sale) CPL - Cost Per Lead, CPC - Cost Per Click (commission paid per click/lead)

Market potential

4M tickets are purchased per week via Eventbrite alone. We can scale to various verticals - from educational events (workshops and classes) and sport events to business/industry related events (such as conferences). Basically any kind of event that is promoted via offline media.

What we learned

A lot about chatbots and image recognition.

Built With

Share this project:

Updates