Software Architect / Microsoft MVP (AI) and Technical Author

Ads API, AdTech, C#, Cognitive Services, Prototyping, Social Media, Twitter

Twitter, Promote 2018 and Just-in-Time Audiences

I’ve kept in contact with the team at Twitter over the past couple of years through projects I’ve worked on as part of the #Promote initiative.  Last year’s submission saw me extend the solution I’d built to surface commercial intent.  You can find out more about that here.

I was invited to their HQ in NYC off the back of that submission and whilst I didn’t win, the experience was great!  After a few chats with some of the team there, access to internal alpha APIs was granted which exposed rich consumer and user data.  This included a wide range of data-points from levels of homeownership, down to the types of TV shows users were tuning into.

I integrated this API (at the time known as ‘AX’) into my existing solution and was quickly able to use these consumer insights to present even more actionable data through web dashboards.  About six months later, this internal Twitter API endpoint was rebranded as Audience Intelligence and released to the public.

#Promote 2018

I received a nudge from some of the team at Twitter that #Promote, an initiative that recognises some of the most innovative marketing solutions would be running again this year.  Dates were initially sketchy but in the past few weeks have been confirmed.  The cut-off date is August 17th which doesn’t leave much time.  I’ve decided to submit another entry though!

JiT Audiences (Just-in-Time)

My current solution, using configurations of your choice, has a feature that generates dynamic audiences (segmented Twitter user accounts) when they are expressing commercial intent. It’s able to do this by leveraging ML and NLP.  Microsoft Cognitive Services saved me heaps of development time from that perspective.

It can also derive actionable insights such as the sentiment of tweets, popular locations, and top keywords.  This processing is done in real-time.  You can then publish these audiences on Twitter to serve them creative if you wish.

The plan is to introduce messaging to the advertising targeting palette by fusing JiT Audiences with some of the recently introduced messaging endpoints Twitter have released and engage with audiences directly.  This can be done using a combination of:

  • contextual messaging
  • quick replies
  • mentions

The key part to all of this is that it’s happening in near real-time.

It will allow digital marketers to target the right person, at the right time, with the right message and at the same time, help improve user engagement and increase awareness.

Twitter Investor Relations released a series of tweets are part of their the Q2 Earnings report for 2018 a few days ago.  One of the key revenue priorities in that report is to ensure that advertisers can reach audiences when on Twitter when they are most receptive:

I think JiT Audiences fits nicely with this priority!

Next steps

There is a lot to be done and not a lot of time.  Fortunately, I know my way around the Twitter API and have a reliable process I use to build, test and integrate new endpoints with my existing solution.  I have a few weekends to get this out…

Get the latest content and code from the blog posts!
I respect your privacy. No spam. Ever.

Leave a Reply