Can AI Go Door Knocking?
The answer may surprise you (and this is not a post about robots).
Lead generation. Prospecting. Hunting and Farming. Whatever term you use, you know that a critical part of the real estate agent job is filling the pipeline.
The channels and methods available for doing so are expanding everyday - as are the shortfalls of each. The digital world has accelerated the frequency of communication and as a result the signal to noise ratio of getting through to a prospective client is quite low.
Which leads me back to good, old-fashioned door knocking. Mapping out a route (easy); wearing comfortable shoes (hopefully); managing the anxiety of what lies behind the door when it actually opens (challenging). While analog, door knocking is still a tried and true method for many agents as they work to build a local market presence and seek out off-market opportunities, albeit time-consuming and not without a physical and mental toll.
When the industry talks about the rise of AI and all the ways in which it can/might/should change the daily lives of agents, lead generation is a clear space in which the combination of a super powerful AI language model and lots of consumer data can unlock powerful insights that can generate opportunities. So much so, that the expediency associated with leveraging AI to find new business can overshadow the need to pursue other, less-efficient methods (e.g. hitting the pavement).
But how should agents think about their business development mix? Is AI going to really change everything about how agents work? Can AI knock on doors?
In short, no. Of course not. Yes, I’m sure the technical capacity for a robot to knock on doors and leave fliers is possible. Is that a good idea? I think we know the answer.
The right question to ask, however, is a different one:
Is Data Crunching the new Door Knocking?
This question reflects one of the larger shifts I believe we’ll see with how agents spend their time as a result of advances in AI (and with how consumers want to be approached). Let’s get into it.
Tech-Driven Lead Generation
The name of the real estate lead generation game for the last couple of decades has been CRM. CRMs are everywhere. Agents do not struggle with not enough CRM options to choose from. Quite the opposite - they are overwhelmed by all of the CRM choices available. And despite all of the benefits of leveraging a CRM to the max, adoption rates of CRMs by agents remain shockingly low (like in the 5-10% range). Why? A lot of reasons, but my guess is it’s just too much work to maintain and update a CRM to get any real value out of it. So what ends up happening is agents rely on their phone, their inbox, and maybe a spreadsheet or two to manage their pipeline and run their business.
And there’s nothing wrong with that. Technology sometimes can be a solution looking for a problem, and being an agent is such a unique job that everyone has their own method for making it work. But AI is quickly changing the dynamics here in a few ways:
Data entry: AI assistants are able to run complex requests and speak to multiple platforms at the same time. A use case here would be directing an AI Assistant to take certain contacts from your inbox, summarizing your recent conversations and meetings with those people, and inputting that information into your CRM in a structured way.
Data mining: By leveraging third-party datasets in conjunction with first-party data that an agent may already have, agents can enrich the profiles of their contacts quickly. Examples could be taking a contact’s name and email address and augmenting it with their social profiles, job history, and other data that is publicly available.
Outreach: This is the trickiest part of the equation. Of course, AI can draft messages and send them - either one by one, or programmatically. Depending on the nature of the outreach, this might be a good approach. But the more qualified and segmented your target audience, the better approach is to use AI to help craft highly personalized messaging (quickly) using all of the datasets outlined above.
Reduce, Reuse, Recycle
We’ve all heard the adage “garbage in, garbage out” when it comes to maintaining a CRM. The quality of your data input largely determines the quality of the insights you glean from it.
The same is true with building AI. The quality of the training dataset is mission-critical to the quality of the output. The difference is, data that historically might be viewed as useless (or perhaps, unusable) can now be activated in valuable ways.
Take Reddit or Quora as examples. These sites are goldmines of random, threaded conversations about a number of topics. Easy to drop into a spreadsheet and organize? Absolutely not. Valuable to train AI models on how people communicate, iterate, and derive conclusions on specific knowledge topics? 100%.
The takeaway for agents is clear - all of your random interactions, messages, emails, texts etc. with your clients could, in the aggregate, become a really valuable source of information for an AI Assistant that can identify insights and patterns within them that you may not ever have been able to just by inputting some data into a basic CRM form.
So yesterday’s data garbage might actually be the very thing that, if reshaped appropriately, ends up unlocking serious value today. It might mean that while you’re out knocking on doors, AI can help you knock down some walls.
The question is - are you ready to give it a shot?
Thanks for reading this week’s edition of the Sidekick newsletter. More to come - and please share with any friends and colleagues that might be interested!
Michael Martin
Co-Founder/Co-CEO
Try Sidekick