60% More Engagement With Hyper-Personalization for a US Proptech
60%
More engagement
Engagement growth measured after the PoC launched.
10%
More conversions
Conversion growth measured over the same period.
6
Weeks to a PoC
From kickoff to a working proof of concept.
The client
Newzip runs real estate as a service: mortgage lenders can pull property services into their own offer, centralising the transaction, cutting costs and lifting lead engagement and conversion.
Home buyers usually deal with loan officers and agents separately, which makes for a disjointed experience. Newzip pairs lenders with top agents nationwide so clients get one concierge service instead.
The challengeWhat the teams actually needed
Newzip's platform already served a broad audience well. The open question was whether hyper-personalization could lift satisfaction and engagement further, and the answer was needed in weeks.
Challenge 01One answer for every buyer
Recommendations were built for a general case, while each client's financial situation called for its own.
Challenge 02 Local insight at the right level
Buyers needed insights about the specific places they were considering.
Challenge 03Little bandwidth to experiment
As a startup, Newzip had limited time for AI exploration. The PoC had to fit the budget and a four-to-six-week timeline.
What we did
An AI proof of concept, scoped and shipped in weeks
The build was a prototype web application in three modules: a database for customer data, operator views that generate personalised content, and a landing page that shows it.
Technical requirements pinned down first
The proof of concept was scoped for precise technical requirements before any model work began.
Research, then the right model
Research led to model selection, then content prototyping with GenAI, with the system developed, tested and connected to customer data.
An API and a prototype site
An API interface opened the way for integration, a prototype website was linked to the model, and support continued through integration with Newzip’s own platform.
Sources and data, sorted en route
The team worked through the reliability of documentation sources, the difficulty of extracting and interpreting data, and client-specific issues that surfaced during implementation.
Client testimonial
It's difficult as a startup to divide your time between evolving your existing product portfolio and experimenting with new technologies. We decided to partner with Netguru to help us beef up our AI exploration and are excited about the results so far.
More like thisOther places we scaled product teams for growth.
What’s next