Personalization in product recommender systems in industries outside of real estate will soon impact how consumers choose—or will want to choose—real estate professionals on brokerage sites. The basic concept: How would Amazon.com recommend a real estate professional? To answer this there are two basic sides to consider: customer behavior within a system (and increasingly outside of the system; see what RETargeter is doing) and attributes and behavior of the real estate professional.
At a very basic level, recommender systems track and log consumer behavior and then match appropriate products and services based on this behavior. The key is that these products and services have particular attributes that “match” the behavior of the consumer. For example, assume Consumer A purchased five historical novels over the past five months, a recommender system likely would recommend another historical novel as a next purchase. So how could this impact real estate professionals?
First, assume a brokerage has a system that logs consumer behavior (login times, locations searched, favorite properties, map searches generated, etc). Second, assume a brokerage has segmented its agent base by basic factors (such as top neighborhoods serviced by the agents, top 10 zip codes serviced by the agents, lifestyle attributes, designation, luxury expert, waterfront expert, client service satisfaction ratings, MLS performance, etc). Next, the real estate professional recommender system could work similarly as to how a book recommender system works. And I know that some listing aggregators already offer this type of service, but these services on generally pay-to-play. What I am suggesting is that brokerages need to do something similar with their system and offer it free-of-charge to their agents.
For example, lets assume Consumer B registers and saves a luxury property overlooking a lake, the system could automatically “recommend” agents who work the zip code of luxury property AND are luxury agents AND are waterfront specialists. Next, let’s assume Consumer B clicks the profiles of each of the recommended agents, he or she will then see overall performance ratings, specific testimonials, and specific customer satisfaction ratings. The benefit to the consumer is that they’re presented with the “best” professional based on their interest, which supports customers-for-life marketing best practices. The benefit to the real estate professional is that they’re in front of the consumer faster and in context to the search process. This type of a process promotes a personalized experience which is key factor in capturing consumer mindshare. And, indeed, there is research that supports this proposition.
[F]ocusing just on the ability of IT to support strategy and processes bears the risk of not utilizing the full potential of innovative technologies[.]
It’s clear there is a critical interdependence between both marketing and IT departments. As marketing seeks to “engage,” “relate with,” and “delight” customers in the continuous battle for share-of-mind and share-of-heart, relegating IT to the sidelines as bench support is not a good strategy. Rather, incorporating IT vision is a critical component in setting strategy. The complexity of consumer interactions with a firm’s brand, demands increasingly sophisticated infrastructure and data management tools to ensure that a firm can meet the needs of these consumers.
Similarly, firms ought to align financial management goals within this marketing-IT milieu. Financial concerns, in this context, center around setting proper marketing metrics to measure ROI and lifetime value of a customer. The paper points out that
[A] number of ﬁnancial concepts (e.g. capital asset pricing model, portfolio theory, and real option approaches) have recently been transferred to customer portfolios…Such “marketing metrics”, based on these approaches and thus taking a future-oriented, long-term, cashﬂow oriented, and risk adjusted perspective, allow for an identiﬁcation and measurement of the economic value contribution and the ROI of marketing[.]
To enable such penetrative insights, firms need to leverage data mining tools to create timely (i.e., near real-time) metrics to be shared across business to ensure uniform adherence to meeting clients’ expectations.
I thought this kiosk flyer below, which I found pinned to a wall somewhere in Estes Park, Colorado, is a novel use of QR.
What I like is how the purveyor has conveniently arrayed the subject matter. As a tourist, I especially liked the choices put before me and the prospect of interacting in an interesting way with the “place” the QR took me. And it is on this latter point where I was let down. The “Arts” QR simply took me to a website, listing a series of events with more links to click. Understandably, having this list of links is definitely a convenience and I would not have visited the site but for the QR. However, I cannot help but think that an opportunity was missed that could have “rewarded” or “delighted” or “surprised” me with some experiential marketing. For example, the QR could have landed me on a page inviting me to play a video that has interviews with local artists and events organizers…let me feel their passion, let me understand their love for their community…grab my engagement by letting me know the impact of my participation. Then I’d be much more likely to click the links related to each event. Further, since these arts events are seasonal, this is a process one can repeat. In real estate one could use QR in a similar manner by showcasing a homeowner interview, interviews with shop owners, or a narrated neighborhood tour. How do you/would you use QR?
Right now, “The Big Shift” is steamrolling us. We are essentially addressing the 55+ audience, and wondering why our businesses are eroding? We must begin investing in the 18-44 year old audience, if we want to remain relevant in 2020, while optimizing profitability from catalog mailings to the 55+ audience.
Very poignant observation, and very applicable concepts to the real estate industry.
Is Google building a stealth social network? Well-reasoned argument that Google is doing this, and that Google’s +1 initiative is part of a series of tactics Google has recently deployed to continue playing its ground game in the social sphere.
Rich media + display ads + social = advertising perfect CTR-engagement metric storm? “Rich media” (aka multimedia) has been around for some time. Similarly, rich media has had periodic bursts of hype and utilization for over a decade (anyone who was in the email marketing space around 1998/1999 will recall the covey of rich media vendors present at the variety of “internet conferences” that occurred during the same time period). Well, it seems rich media is back again (like a poltergeist?) and advertisers seem excited (according to the article). What’s interesting to me is whether there’s an opportunity for enhanced engagement via a rich media ad conduit that will support social CRM initiatives.
1) Strategically embrace Web 2.0 and facilitate consumer control over certain elements of an overall marketing plan
Supporters’ blogs and You Tube postings were also brought inside the campaign through the website, where the online team could help consolidate the energy and contacts generated by them.
2) Test, measure, refine, roll-out; keep what works, ditch what tanks; no “sacred cows”
[The new media team], meanwhile, was constantly testing different versions of its call-to-action pages, including requests for donations and voter registration. Did more people respond if it included video or text? Should the sign-up prompts be on the right column or in the center? Should they have a “learn more” button or direct sign-up? Once they discovered the most effective version, they replaced all the others with it. Among their lessons: Video can sometimes be a distraction rather than a help.
Once users input a location they want to learn more about on Trulia, Placecast will access that data and apply it as a key component along with common demographic data points like psychographic information to provide more targeted ads.
This process makes sense especially at the zip code level (see previous posts on zip code optimization) because demo/psychographic differences exist between zip codes–even contiguous zip codes. Accordingly, if I’m looking in a zip code that trends more affluent, Trulia can now serve ads that appeal to an affluent consumer (Jaguar advertisement). Alternatively, if I’m searching in a zip code that trends more middle of the road, Trulia can now serve an ad that appeals to a bargain shopper (Toyota Corolla advertisement).
For real estate, I’d like to see a twist on this process: somehow also deduce from where a consumer searches so as to better deploy advertising resources with respect to select properties. For instance, let’s assume you’re a firm situated in a Utah ski resort community, and that you know based on previous dealings with out-of-market buyers that your to primary “feeder” markets are Chicago and Orlando, and that these primary markets are generally interested in purchasing luxury-oriented rental income properties.
It’d be a great service to be able choose which of your top properties to “enhance” that exist in a specific zip code and display the “enhanced” versions of these properties only when a consumer from either Chicago or Orlando conducts a search in the targeted zip code. Employing a scheme like this, one makes an ad buy based on a “known” marketing attribute (i.e., based on personal experience) along with hyper-targeting, which should translate into higher quality clicks to the “enhanced” properties and, thus, increase the potential ROI on those ad buys.
As always I am grateful to Owyang to lend his insight and foresight. Here’s another excellent missive on the “Intelligent Web”. In summary, he posits that machines will begin extrapolating relationships and driving recommendations for connections from the juxtapositions and nexus between “our behaviors, context, and preferences”. Sounds a bit like the semantic web. Spinning through the comments on this post brought me to the Innovation Insight blog where Guy Hagen explores MIT research related to “reality mining”, which you can find more about on the MIT Web site. And this research paper out of UC DAVIS demonstrates how the MIT Reality Mining data set was utilized in tracking behaviour via mobile phones.
Imagine an iPhone application overlayed on a real estate firm’s listing data set, where the iPhone reports back over time thousands of user’s mobile browsing habits (i.e., driving around looking at homes for sale or rent). Having such data would allow firms to target advertising, Web site promotions, and give predictive insight over their competitors with respect to fluctuating markets (e.g., patterns will emerge over time that will tell a firm which neighborhoods, etc, are capturing consumer interest, thus enabling a firm to deploy marketing and agent resources towards these locations ahead of their competition).
Assume you’re a brokerage firm with a wide distribution of properties over several zip codes. Aside from basic syndication to online aggregators, what’s another strategy to market your listings? One fee-based option that many aggregators offer is enhanced listings. Before you pay, however, ask them to prove their merit.
Assume you cover these two zip codes 28226 and 28104. According to Claritas, homes in these zip codes have very different consumer attributes (you’ll have to enter the zip codes yourself to get the results).
Armed with the demographic information, you should ask your online aggregator to give you a demographic break-down, at the zip+4 level, of it’s user audience on the search patterns and niche pages/sections of its site. For example, assume a consumer is searching an aggregator’s site in your coverage area (indicated by the consumer entering city name or 5-digit zip code as search criteria). Based on these entries, relevant properties are returned to the consumer. It’s at this moment aggregators give you an opportunity to have an enhanced listing display to this consumer.
Now it’s your turn to push back: ask for the historical demographic breakdown of the users who entered those search criteria: does the demographic base skew towards segment A (assume A is more likely to own an inexpensive American made car and have a household income below $50,000) or segment B (assume B is more likely to own an expensive foreign made car and belong to a country club)? Once you know, you will know which listings to enhance, while including appropriate imagery and content triggers that appeal to the lifestyle attributes of your targeted demographic segment.
For example, if the base skews towards segment B, perhaps you choose to only enhance listings that are 1) above $750,000, 2) close to a country club, and 3) have ample space for a boat.
Thus, you’re consolidating your advertising resources by focusing on high-gain marketing activities that give you a higher chance of getting a high quality click/lead. It’s a win for the advertiser too because they’re serving you better by giving you the opportunity to gain a high value click/lead (thus promoting retention of their services), while legitimately asking for a higher CPM or CPC for such.
What does The Filter have to do with real estate search? (NOTE: I went through the The Filter Q&A and have to say it was eerily prescient).
What if there was a site where a consumer would 1) define the location where they want to live (via natural language, drop down, or map search), 2) answer a simple set of “lifestyle-oriented” questions, the answers to which would bump up against Claritas’ Prizm database and 3) where a real estate broker would have performed a similar zip+4 coding of their listings? When the consumer presses the “Go” button, the answers to the lifestyle questions would peg a PRIZM code to them (via session cookie or registration ID) that would relate to the same PRIZM code tagged to the properties and deliver only those matched properties to the consumer.
The benefit to the consumer is they’ve cut through gobs of listings that may not fit their lifestyle and found the ones that do. The benefit to the broker is they’ve delivered a high value service to the client. If the broker then had live chat, IM, or showing appointment booking features on each listing, there’s a higher chance of getting a conversation started and higher quality inquiry on the listing.
Right result, right time, right for the consumer.
Arguably, nothing messes with a firm’s loyalty and/or CRM strategy more than a multitude of false consumer profiles polluting a CRM database. In seeking to elevate one’s marketing engagement index, it’s often helpful to understand the demographic profile of a consumer. But if such a consumer does not self-report this, or if such data is not inferred, then firms are at the mercy of the garbage.
The profiles users may contain fake information. We believe that our proposed algorithm can be used to identify and refine the profiles which contain bogus demographic information.
Essentially, this team analyzed web log files for search patterns and used an algorithm to predict gender or age. They claim a lift in accuracy of 30.4% on gender prediction and 50.3% on age prediction over traditional methodologies.
What makes this exciting is that, assuming futher testing bears out the team’s claims, companies like HitWise or WebTrends can incorporate this algorithm into its search pattern analsysis products. Firms can then use this core demographic information to craft more relevant landing pages, calls to actions, etc, on their websites.