1
The rich text element allows you to create and format headings, paragraphs, blockquotes, images, and video all in one place instead of having to add and format them individually. Just double-click and easily create content.
A rich text element can be used with static or dynamic content. For static content, just drop it into any page and begin editing. For dynamic content, add a rich text field to any collection and then connect a rich text element to that field in the settings panel. Voila!
Headings, paragraphs, blockquotes, figures, images, and figure captions can all be styled after a class is added to the rich text element using the "When inside of" nested selector system.
“Data is the new oil” turned out to be correct. But moving from a mountain of data to valuable insights wasn’t as straightforward as everyone thought. NineTwoThree and DataFlik made it happen for real estate data.
Let’s walk through how we helped DataFlik completely transform their business, from idea to predictive data that tells you exactly who will sell their house.
DataFlik is one of the country’s largest providers of wholesale residential real estate data. Their real estate lead generation platform is one of the most powerful ways for realtors to grow their business.
Still, real estate wholesalers have a tough time. Spending tens of thousands of dollars on fliers means nothing when they end up in mailboxes of people who aren’t ready to sell. Finding motivated sellers involves cold-calling, huge mailing lists, and tons of research.
That means a lot of data. When they approached NineTwoThree, they had a groundbreaking hypothesis: Using their data points, they could develop a ML model that could identify sellers before they hit the market.
Using this model, DataFlik’s customers could accurately predict which houses would sell, and target motivated sellers. They could better target their mailing lists, saving money and increasing sales.
The data was there. The only problem? They were a startup, and couldn’t afford a CTO. They didn’t just need a partner - they needed an entire outsourced engineering organization, with data science experts.
NineTwoThree was selected to help work together with DataFlik to develop their ML model and make their insight dreams a reality. With our experience in building cutting-edge machine learning solutions, we were selected to cut through the mountains of data and find the insights. We combined our extensive data science expertise with DataFlik’s incredible talent and data to find a solution.
And, spoiler, that solution is now a robust, staffed revenue stream for DataFlik.
Let’s take a closer look.
AI and ML solutions typically live and die on their data quality. NineTwoThree’s approach puts a huge emphasis on this critical step in building production AI apps. There’s not an AI model in the world that can make accurate predictions with terrible data.
Luckily, DataFlik invests heavily in their data sources. They have robust data on distress, demographics from the census, volume of deals, and more. This data needs to all be in one central place.
This allowed us to approach phase 1 - data analysis sessions.
“What about training the model? And creating data pipelines?”
DataFlik had to provide a large amount of data to create an effective machine learning model. We were looking at millions of properties across the United States, and each property has about 1,700 data points. The real estate data needed to achieve the desired goal is extremely expensive to purchase.
NineTwoThree views our partnerships as exactly that - a partnership. We’re just as invested in the success of the project as our clients. In this case, we had a great hypothesis and great data. But we wanted to make sure this was a ML solution that had promise, before committing to a large purchase and project.
We had to prove that our proposed model will achieve its goal of generating an effective motivated sellers mailing list - without developing the full model - to justify the purchase of such data.
To accomplish that, we worked alongside DataFlik’s market domain experts to identify high value data points to apply to our ML model. This establishes a firm foundation for our proof of concept.
Notice there’s no mention of generative AI here. GenAI has sort of become a catch-all term to describe any AI project, but classical machine learning techniques are often just as effective. They come with a huge bonus, too - being much cheaper.
NineTwoThree created an efficient, robust data pipeline that handled terabytes of data. This was non-negotiable - we needed clean, quality data constantly flowing into the model.
We even handled a 10TB data migration to a new provider - one of the real estate data providers offered us a better deal, with improved data quality, so we switched. A difficult task onto itself, and one we couldn’t have easily predicted in discovery. Accomplishing this with zero customer downtime was a huge achievement for the team.
As you might expect, processing all this data efficiently is another challenge. There’s no use in investing in a robust AI solution if it can’t give you insights in a timely fashion.
Most of this code was custom-written, because open-source models weren’t performant or applicable to DataFlik’s unique situation. We also used many different AWS services, almost 20 in total.
Our ML model predicted three things:
The results were incredible: we accurately predicted about 60-70% of home sales every month, and the model keeps improving.
One key insight we uncovered during the project was an operational improvement.
DataFlik finds the insights generated from our ML model and packages them up in a report to send to their customers. Generating these monthly reports was a painstaking, manual process. 10 percent of their customers requested customizations down to the street, year, subdivision…all by phone call.
These custom reports were generated and sent one by one.
We realized we could create two opportunities with one application:
We agreed this was a worthwhile investment, and added on a full-stack development team to assist in creating this application.
This was only possible because of NineTwoThree’s extensive experience building digital solutions - other agencies might just shrug and say that’s out of their scope.
It’s hard to overstate how much DataFlik’s company transformed in just over 2 years..
The project started with one research data scientist. Recall, DataFlik didn’t even have a CTO.
NineTwoThree spent over a year hiring, onboarding and training a full team of in-house experts:
We applied our years of expertise in hiring top 0.1% engineers, and ran that process for DataFlik. Now, their team is self-sufficient.
But, hiring is one thing; building a fundraise-worthy product is quite another.
NineTwoThree delivered. We didn’t just create a world-class product:
Thanks to our help and preparation, DataFlik took the entire project in-house.
Their team is scalable, credible, and ready for the next level of growth.
NineTwoThree’s expertise in UX design actually came in handy here.
The model we developed performed too well.
Each month, it was accurately predicting peaks and dips in the housing market - which caused problems for end customers.
They are used to predictable list volume - buying a set number of advertisements each month. They actually preferred stability and predictability over overall accuracy. We adjusted our model accordingly.
DataFlik was kind enough to leave us a public review on Clutch.
“Thanks to the algorithm that NineTwoThree Venture Studio has helped us construct, we accurately predict about 60%–70% of home sales. We’re one of the top five companies in the industry. We’re pretty happy with that, and it gets better every month. “
“NineTwoThree Venture Studio has been great. They’re pretty integrated into our daily staff meetings and work. Before them, we had a satellite office and a lot of overseas development. However, the language barriers were too rough, and communication was poor, so we swapped agencies.”
“The biggest difference marker is their knowledge. The team that supports and executes the implementation has vast knowledge and experience. Moreover, the founding staff is a treasure trove of insight, and everyone is incredibly strong at implementation.”
We’ve done dozens of AI/ML projects, and countless more in the app and product development space.
Reach out if you’re interested in transforming your business with AI-driven insights.
