Re-imagining the agent-seller relationship with proptech
Proptech has been experiencing strong growth since around 2014 (think Zillow, Purplebricks or Rightmove) but a focus on buyers has left sellers behind.
When Helmsmen approached us they had a simple idea: When buyers look for their ideal property, they have a must-have list. What if sellers could apply the same logic when choosing an estate agent?
Few offerings, digital or otherwise, analyse agents based on their past performance, customer reviews and market experience. This means that sellers have no means of identifying the agent with the best track record tailored to their property and personal needs.
As a result, most property sellers are forced to choose an estate agency based recommendations, rather than facts.
It is still very early stage for us, but we have now matched over £100M worth of property to the right estate agent
Two-week sprints were used to conceive and iteratively develop a platform that could adapt quickly as our client fleshed out his ideas.
We began by analysing the types of data sellers had an interest in before building a corresponding set of filters: How quickly does an individual agent sell properties? What areas of specialty do they have? What percentage of the asking price do they achieve? And so on.
Once we had the filters nailed down, we ideated and decided on solutions for underlying search areas, third-party integrations, databse solutions and overnight jobs, and built out the rest of the pages with content and an set of user account areas.
Automated deployments were made to the test environment to allow the client to follow the journey and give regular feedback as we built.
The marketplace was launched with some early-adopter agents, and sellers began to be matched to agents almost immediately.
Functionality was sufficient for build-out to focus on delivering UX improvements rather than technology upgrades, and the choice to integrate with agency software providers front-loaded the project for data-driven product development.
We chose to build a Minimum Viable Product (MVP) to enable our client to start matching sellers to agents in a semi-automated fashion. Once the marketplace model was validated, we would begin product development based on user feedback and evolving business needs.
We focused heavily on the geospatial requirements we felt our client would require over the first two years of business, incorporating point, radius, distance, rectangle, true polygon, intersection, cluster and subtraction searches.
Integrating the platform with our client’s choice of agency software provider allowed us to deliver details of properties each agent had on the market, or had sold, on a daily basis, as well as to provide scope for strong data analytics and search results for each individual agent.
Automating the on-boarding process for a more streamlined approach