COLORBOND Nation
New COLORBOND steel roofs use ~20% recycled steel and sourcing it through third parties is costly.
But what if we were able to leverage new technologies to locate that steel more efficiently and at a larger scale?
We set out to map all steel roofs in Australia.
To do this we first scan the Google Maps Platform using Recognition AI to identify all ‘roofs’, then implement a training model to isolate steel roofs. We leverage the same training model to identify older steel roofs based on year-on-year Google Maps imagery – which allows us to focus only on aging roofs. We then use BlueScope-owned data to exclude COLORBOND steel roofs still within warranty. Then we cross-reference properties with ownership and rental data to determine interest and ability to afford.
What we’re left with is a rich data set of warm leads for a targeted acquisition campaign.
We would then target prospects most likely to need a new steel roof with a DM pack that directs them to a hyper-personalised microsite and ‘Buy Back’ offer. The microsite pulls in the customers address, a streetview image of their property and an estimate on the amount of steel sitting on their roof. All instantly generated by AI.
The program could easily be extended to steel fences and garage doors with the addition of Google Street View data. And not to forget the substantial industrial warehouse market that blankets the outer suburbs.
By targeting new roof prospects with a hyper-personalised experience and an enticing ‘buy back’ offer all generated instantly by ai we’re closer than ever to creating a nation of COLORBOND.
WINNER: CXL Wavemaker 2023
Best use of Generative AI to solve a business problem