The document discusses recommendations for improving recommendations on the Zillow real estate website. It recommends:
1) Adding an onboarding wizard to collect user preferences and encourage ratings of properties.
2) Displaying a progress bar to motivate users to provide more ratings to improve recommendations.
3) Allowing users to provide 5-point ratings of properties to collect explicit preference data.
4) Adding a "Recommended" tab displaying properties predicted to match a user's preferences based on combined collaborative and content-based filtering models.
5) Only sending property emails when high-probability matches are found to improve conversion rates.
This document discusses how to get more local business from your web presence through Google Places and Local Mobile Fusion. It explains that Google Places is important for local businesses and provides information like business name, address, photos and reviews. Local Mobile Fusion helps claim and optimize a Google Places page to rank higher on search engines by adding photos, videos, citations and reviews. It also includes setting up listings on top local directories to generate more citations and visibility.
Winning data science competitions, presented by Owen ZhangVivian S. Zhang
<featured> Meetup event hosted by NYC Open Data Meetup, NYC Data Science Academy. Speaker: Owen Zhang, Event Info: http://www.meetup.com/NYC-Open-Data/events/219370251/
What are segments, and why should you care about them? This presentation explains and then walks you through using Google Analytics' new Advanced Segments feature.
- The document discusses how to use predictive analytics and A/B testing to increase user conversion and revenue for a gaming startup.
- By analyzing user data using decision trees, they found paying users had different experiences than non-paying users, spending more virtual currency on items.
- They implemented an A/B test where new users in group B were forced to spend coins on virtual items initially to mimic paying users' experiences.
- The success of this change would be measured by whether group B bought more virtual items and spent more coins, indicating aligning user experiences increased conversion.
Analytics, Search, Social Media, and Optimization: Why Has Marketing Gotten S...Kate O'Neill
From search and social media to analytics and optimization, marketing has really gotten geeky. It's nearly impossible to keep up, so what should business owners know about online marketing in order to make good decisions about their web presence? This presentation is both a broad overview of key web marketing disciplines as well as a quick dive into some of the concepts and vocabulary behind them.
Presented on Wednesday, August 18th to the Women Business Owners Special Interest Group of the Nashville Area Chamber of Commerce.
In this slideshare, we break down Robinhood's fantastic mobile user onboarding experience. It's by far our favorite one we've seen so far, so let's jump in.
The CHR Group: Search&Display Digital Advertising (patented technology)Jonathan Zaback
The CHR Group has licensed three US patents to make display/banner ads interactive. These patents enable consumers to interact with the ad without leaving the page they are on. Huge ROI for clients and gives you a new product to sell in your media buying department.
In this slideshare, we break down Robinhood's fantastic mobile user onboarding experience. It's by far our favorite one we've seen so far, so let's jump in.
This document discusses how to get more local business from your web presence through Google Places and Local Mobile Fusion. It explains that Google Places is important for local businesses and provides information like business name, address, photos and reviews. Local Mobile Fusion helps claim and optimize a Google Places page to rank higher on search engines by adding photos, videos, citations and reviews. It also includes setting up listings on top local directories to generate more citations and visibility.
Winning data science competitions, presented by Owen ZhangVivian S. Zhang
<featured> Meetup event hosted by NYC Open Data Meetup, NYC Data Science Academy. Speaker: Owen Zhang, Event Info: http://www.meetup.com/NYC-Open-Data/events/219370251/
What are segments, and why should you care about them? This presentation explains and then walks you through using Google Analytics' new Advanced Segments feature.
- The document discusses how to use predictive analytics and A/B testing to increase user conversion and revenue for a gaming startup.
- By analyzing user data using decision trees, they found paying users had different experiences than non-paying users, spending more virtual currency on items.
- They implemented an A/B test where new users in group B were forced to spend coins on virtual items initially to mimic paying users' experiences.
- The success of this change would be measured by whether group B bought more virtual items and spent more coins, indicating aligning user experiences increased conversion.
Analytics, Search, Social Media, and Optimization: Why Has Marketing Gotten S...Kate O'Neill
From search and social media to analytics and optimization, marketing has really gotten geeky. It's nearly impossible to keep up, so what should business owners know about online marketing in order to make good decisions about their web presence? This presentation is both a broad overview of key web marketing disciplines as well as a quick dive into some of the concepts and vocabulary behind them.
Presented on Wednesday, August 18th to the Women Business Owners Special Interest Group of the Nashville Area Chamber of Commerce.
In this slideshare, we break down Robinhood's fantastic mobile user onboarding experience. It's by far our favorite one we've seen so far, so let's jump in.
The CHR Group: Search&Display Digital Advertising (patented technology)Jonathan Zaback
The CHR Group has licensed three US patents to make display/banner ads interactive. These patents enable consumers to interact with the ad without leaving the page they are on. Huge ROI for clients and gives you a new product to sell in your media buying department.
In this slideshare, we break down Robinhood's fantastic mobile user onboarding experience. It's by far our favorite one we've seen so far, so let's jump in.
Digital Performance Analyst
The document discusses how analytics can be used to influence changes using a case study of analyzing an IPO service. It identifies key drop-off points in the service's conversion funnel and explores using multivariate testing to determine how to improve the areas causing users to abandon the process, such as by testing different page layouts, wording, or designs. While multivariate testing provides the best data-backed approach, simple analyses of metrics like bounce rates, time on page, and conversion funnels can still provide insights into how to guide changes when more sophisticated testing isn't possible.
This landing page for a book on starting an online business has several strengths: it uses numerous social proof elements like media logos and a testimonial from a bestselling author to build trust; the headline combines social proof, specificity and curiosity; and the subheadline frames the customer as a "hero" who can achieve what others have. However, there are also weaknesses such as an overly busy design with too many fonts, sizes, and colors that make the page feel hectic; and too many social proof elements that may be overkill. The page could be improved primarily by simplifying the design and look/feel to make it easier to process and more trustworthy.
Presentation on Lean Analytics at MicroConf 2013. Understanding what metrics are the most value, when, for your type of business.
* What makes a good metric?
* Types of metrics (qualitative vs. quantitative, vanity vs. actionable, etc.)
* Lean Analytics framework
Shared a number of case studies: Airbnb, Buffer, ClearFit, OffceDrop and others.
Idean works very closely with startups through carefully selected incubators and accelerators.
This experience led Idean's head of design to conduct a presentation and a large group workshop on October 17, 2013 for 8 startups at the Blackbox incubator.
Kindly find the presentation attached. It has some screens of different research methodologies. Enjoy!
This document introduces the concepts of lean analytics and the lean analytics framework. It begins with an introduction to lean startup methodology and emphasizes the importance of testing hypotheses through minimum viable products and the build-measure-learn loop. It then discusses different types of metrics and introduces the lean analytics framework, which focuses on empathy, stickiness, virality, revenue, and scale. For each stage, it provides examples of relevant metrics for different business models. The framework is intended to help companies adopt a lean approach to analytics and product development.
5 Steps to Creating Data-backed Personas for User Experience (UX) DesignAngela Obias
I've become a persona skeptic and it's because I've seen many an "imaginary" persona in my life.
I respect the integrity of personas, and I just really wanted to share, in my own little way, how anyone can apply personas to a web design project, using the actual data-based process.
Notes inside! Practical advice for measuring, analyzing, and reporting your n...Sarah M Worthy
This document provides an overview of a presentation on measuring, analyzing, and reporting nonprofit digital data. The presentation covers defining key performance indicators (KPIs) to track an organization's goals and strategies. It also discusses connecting digital marketing tactics to results using analytics tools like Google Analytics. The presenter provides tips on consistently measuring the right metrics over time and avoiding "vanity metrics" that don't provide useful insights.
Landing Page Optimization, What is Landing Page Optimization, Examples, 5-second Testing, Fogg Behavior, AIDA Model, Marketing Models, LPO Checklist, Limbic System, Types of Landing Page Optimization, Five Second Test
This document summarizes and recommends 7 internet marketing products. It begins by discussing the author's experience of wasting money on ineffective products before finding 5 paid products and 2 free products that were truly valuable. The top recommended paid product is Market Samurai, described as the best keyword tool that also helps with content creation, pay-per-click campaigns, and more. The second is Get Response, an email marketing platform emphasized as important for building an email list to generate ongoing revenue. The summary concludes after briefly introducing the 5 paid and 2 free product recommendations.
This document contains terms and conditions for a book about making money online. It notes that while efforts were made to verify the accuracy of the content, errors may exist. It also states that the book is not intended as a source of legal, business, or financial advice. The document includes a table of contents that lists 8 chapters that provide steps and strategies for making money online through affiliate marketing and building a website or blog.
This document provides tips and strategies for businesses to optimize their local Google listings. It discusses the importance of having an accurate listing with complete information, linking to the listing from websites and other online profiles, leveraging additional third party listing sites to generate backlinks, getting customer reviews and submissions, and following Google's guidelines to avoid penalties. The goal is to generate trusted sources of information about the business online to improve search engine rankings and visibility on Google searches.
The document discusses metrics and monetization strategies for software startups. It recommends focusing on metrics that improve customer retention and lifetime value over the long run, such as increasing retention rates from 80% to 90%, rather than one-time gains. High leverage opportunities provide ongoing revenue through evangelism and happy customers who continue using the product. The document also provides tips for optimizing different parts of the customer funnel from acquisition to purchase.
You discover how to find that niche, choose in affiliate program,free training,make a website or blog, and marketing methods. Also to get training on generate more sales in a day.
how to earns using Facebook passive incomeKumar Kishor
After creating a suitable profile the continuous step of ensuring the line of communication is kept open at all times is important. Positioning the business platform and staying in touch with the viewers will give the site the exposure it needs.
Unfortunate
This document provides tips for getting an SEO analytics program beyond the "stone age" by focusing on opportunity, competition, and attribution. It recommends using analytics tools to find opportunities by analyzing popular pages, keywords, and competition. Proper attribution is important to understand what is truly driving conversions and avoid costly mistakes. Log files can provide a better picture of attribution by showing all customer visits and referrals before a conversion.
First things first, I want to make a statement here and some of you may
agree some of you may disagree.
I want to say that standard “text only”sales letters still work. They still work
today but because people are just seeing them over and over and over
again. A lot of people have got used to it. Yes, they still converse. The only
real way of knowing whether or not they're going to convert for you is by
testing it.
Web metrics-cell-carrier-buzz-on-the-web-12063Lextant
The document discusses various tools for analyzing online conversations and user-generated content regarding cell phone carriers like Verizon, Sprint, T-Mobile, and Cingular. It explores blogs, photos tagged on Flickr, videos on sites like YouTube, and social news sites like Digg. The tools can provide insight into topics discussed, tone of conversations, how content is organized and tagged, and what generates the most online buzz. However, the data does not provide demographic information about users or their locations.
Web Metrics - Cell Carrier Buzz on the WebDan Rockwell
This is a presentation i gave to team members where I work about how to harness potential web tools to tap into how brands resonate and live in the consumers web world...
You’ve spent countless hours searching for the right business. You finally stumbled across “Doing YouTube without making videos” and you know this is finally the answer you’ve been looking for.
You love the idea that it’s very lucrative, doesn’t cost any money to get started, doesn’t take much time to work on, it doesn’t require any technical skills or experience, and you don’t have to be on camera or use your voice.
Open Source Contributions to Postgres: The Basics POSETTE 2024ElizabethGarrettChri
Postgres is the most advanced open-source database in the world and it's supported by a community, not a single company. So how does this work? How does code actually get into Postgres? I recently had a patch submitted and committed and I want to share what I learned in that process. I’ll give you an overview of Postgres versions and how the underlying project codebase functions. I’ll also show you the process for submitting a patch and getting that tested and committed.
Digital Performance Analyst
The document discusses how analytics can be used to influence changes using a case study of analyzing an IPO service. It identifies key drop-off points in the service's conversion funnel and explores using multivariate testing to determine how to improve the areas causing users to abandon the process, such as by testing different page layouts, wording, or designs. While multivariate testing provides the best data-backed approach, simple analyses of metrics like bounce rates, time on page, and conversion funnels can still provide insights into how to guide changes when more sophisticated testing isn't possible.
This landing page for a book on starting an online business has several strengths: it uses numerous social proof elements like media logos and a testimonial from a bestselling author to build trust; the headline combines social proof, specificity and curiosity; and the subheadline frames the customer as a "hero" who can achieve what others have. However, there are also weaknesses such as an overly busy design with too many fonts, sizes, and colors that make the page feel hectic; and too many social proof elements that may be overkill. The page could be improved primarily by simplifying the design and look/feel to make it easier to process and more trustworthy.
Presentation on Lean Analytics at MicroConf 2013. Understanding what metrics are the most value, when, for your type of business.
* What makes a good metric?
* Types of metrics (qualitative vs. quantitative, vanity vs. actionable, etc.)
* Lean Analytics framework
Shared a number of case studies: Airbnb, Buffer, ClearFit, OffceDrop and others.
Idean works very closely with startups through carefully selected incubators and accelerators.
This experience led Idean's head of design to conduct a presentation and a large group workshop on October 17, 2013 for 8 startups at the Blackbox incubator.
Kindly find the presentation attached. It has some screens of different research methodologies. Enjoy!
This document introduces the concepts of lean analytics and the lean analytics framework. It begins with an introduction to lean startup methodology and emphasizes the importance of testing hypotheses through minimum viable products and the build-measure-learn loop. It then discusses different types of metrics and introduces the lean analytics framework, which focuses on empathy, stickiness, virality, revenue, and scale. For each stage, it provides examples of relevant metrics for different business models. The framework is intended to help companies adopt a lean approach to analytics and product development.
5 Steps to Creating Data-backed Personas for User Experience (UX) DesignAngela Obias
I've become a persona skeptic and it's because I've seen many an "imaginary" persona in my life.
I respect the integrity of personas, and I just really wanted to share, in my own little way, how anyone can apply personas to a web design project, using the actual data-based process.
Notes inside! Practical advice for measuring, analyzing, and reporting your n...Sarah M Worthy
This document provides an overview of a presentation on measuring, analyzing, and reporting nonprofit digital data. The presentation covers defining key performance indicators (KPIs) to track an organization's goals and strategies. It also discusses connecting digital marketing tactics to results using analytics tools like Google Analytics. The presenter provides tips on consistently measuring the right metrics over time and avoiding "vanity metrics" that don't provide useful insights.
Landing Page Optimization, What is Landing Page Optimization, Examples, 5-second Testing, Fogg Behavior, AIDA Model, Marketing Models, LPO Checklist, Limbic System, Types of Landing Page Optimization, Five Second Test
This document summarizes and recommends 7 internet marketing products. It begins by discussing the author's experience of wasting money on ineffective products before finding 5 paid products and 2 free products that were truly valuable. The top recommended paid product is Market Samurai, described as the best keyword tool that also helps with content creation, pay-per-click campaigns, and more. The second is Get Response, an email marketing platform emphasized as important for building an email list to generate ongoing revenue. The summary concludes after briefly introducing the 5 paid and 2 free product recommendations.
This document contains terms and conditions for a book about making money online. It notes that while efforts were made to verify the accuracy of the content, errors may exist. It also states that the book is not intended as a source of legal, business, or financial advice. The document includes a table of contents that lists 8 chapters that provide steps and strategies for making money online through affiliate marketing and building a website or blog.
This document provides tips and strategies for businesses to optimize their local Google listings. It discusses the importance of having an accurate listing with complete information, linking to the listing from websites and other online profiles, leveraging additional third party listing sites to generate backlinks, getting customer reviews and submissions, and following Google's guidelines to avoid penalties. The goal is to generate trusted sources of information about the business online to improve search engine rankings and visibility on Google searches.
The document discusses metrics and monetization strategies for software startups. It recommends focusing on metrics that improve customer retention and lifetime value over the long run, such as increasing retention rates from 80% to 90%, rather than one-time gains. High leverage opportunities provide ongoing revenue through evangelism and happy customers who continue using the product. The document also provides tips for optimizing different parts of the customer funnel from acquisition to purchase.
You discover how to find that niche, choose in affiliate program,free training,make a website or blog, and marketing methods. Also to get training on generate more sales in a day.
how to earns using Facebook passive incomeKumar Kishor
After creating a suitable profile the continuous step of ensuring the line of communication is kept open at all times is important. Positioning the business platform and staying in touch with the viewers will give the site the exposure it needs.
Unfortunate
This document provides tips for getting an SEO analytics program beyond the "stone age" by focusing on opportunity, competition, and attribution. It recommends using analytics tools to find opportunities by analyzing popular pages, keywords, and competition. Proper attribution is important to understand what is truly driving conversions and avoid costly mistakes. Log files can provide a better picture of attribution by showing all customer visits and referrals before a conversion.
First things first, I want to make a statement here and some of you may
agree some of you may disagree.
I want to say that standard “text only”sales letters still work. They still work
today but because people are just seeing them over and over and over
again. A lot of people have got used to it. Yes, they still converse. The only
real way of knowing whether or not they're going to convert for you is by
testing it.
Web metrics-cell-carrier-buzz-on-the-web-12063Lextant
The document discusses various tools for analyzing online conversations and user-generated content regarding cell phone carriers like Verizon, Sprint, T-Mobile, and Cingular. It explores blogs, photos tagged on Flickr, videos on sites like YouTube, and social news sites like Digg. The tools can provide insight into topics discussed, tone of conversations, how content is organized and tagged, and what generates the most online buzz. However, the data does not provide demographic information about users or their locations.
Web Metrics - Cell Carrier Buzz on the WebDan Rockwell
This is a presentation i gave to team members where I work about how to harness potential web tools to tap into how brands resonate and live in the consumers web world...
You’ve spent countless hours searching for the right business. You finally stumbled across “Doing YouTube without making videos” and you know this is finally the answer you’ve been looking for.
You love the idea that it’s very lucrative, doesn’t cost any money to get started, doesn’t take much time to work on, it doesn’t require any technical skills or experience, and you don’t have to be on camera or use your voice.
Open Source Contributions to Postgres: The Basics POSETTE 2024ElizabethGarrettChri
Postgres is the most advanced open-source database in the world and it's supported by a community, not a single company. So how does this work? How does code actually get into Postgres? I recently had a patch submitted and committed and I want to share what I learned in that process. I’ll give you an overview of Postgres versions and how the underlying project codebase functions. I’ll also show you the process for submitting a patch and getting that tested and committed.
Global Situational Awareness of A.I. and where its headedvikram sood
You can see the future first in San Francisco.
Over the past year, the talk of the town has shifted from $10 billion compute clusters to $100 billion clusters to trillion-dollar clusters. Every six months another zero is added to the boardroom plans. Behind the scenes, there’s a fierce scramble to secure every power contract still available for the rest of the decade, every voltage transformer that can possibly be procured. American big business is gearing up to pour trillions of dollars into a long-unseen mobilization of American industrial might. By the end of the decade, American electricity production will have grown tens of percent; from the shale fields of Pennsylvania to the solar farms of Nevada, hundreds of millions of GPUs will hum.
The AGI race has begun. We are building machines that can think and reason. By 2025/26, these machines will outpace college graduates. By the end of the decade, they will be smarter than you or I; we will have superintelligence, in the true sense of the word. Along the way, national security forces not seen in half a century will be un-leashed, and before long, The Project will be on. If we’re lucky, we’ll be in an all-out race with the CCP; if we’re unlucky, an all-out war.
Everyone is now talking about AI, but few have the faintest glimmer of what is about to hit them. Nvidia analysts still think 2024 might be close to the peak. Mainstream pundits are stuck on the wilful blindness of “it’s just predicting the next word”. They see only hype and business-as-usual; at most they entertain another internet-scale technological change.
Before long, the world will wake up. But right now, there are perhaps a few hundred people, most of them in San Francisco and the AI labs, that have situational awareness. Through whatever peculiar forces of fate, I have found myself amongst them. A few years ago, these people were derided as crazy—but they trusted the trendlines, which allowed them to correctly predict the AI advances of the past few years. Whether these people are also right about the next few years remains to be seen. But these are very smart people—the smartest people I have ever met—and they are the ones building this technology. Perhaps they will be an odd footnote in history, or perhaps they will go down in history like Szilard and Oppenheimer and Teller. If they are seeing the future even close to correctly, we are in for a wild ride.
Let me tell you what we see.
End-to-end pipeline agility - Berlin Buzzwords 2024Lars Albertsson
We describe how we achieve high change agility in data engineering by eliminating the fear of breaking downstream data pipelines through end-to-end pipeline testing, and by using schema metaprogramming to safely eliminate boilerplate involved in changes that affect whole pipelines.
A quick poll on agility in changing pipelines from end to end indicated a huge span in capabilities. For the question "How long time does it take for all downstream pipelines to be adapted to an upstream change," the median response was 6 months, but some respondents could do it in less than a day. When quantitative data engineering differences between the best and worst are measured, the span is often 100x-1000x, sometimes even more.
A long time ago, we suffered at Spotify from fear of changing pipelines due to not knowing what the impact might be downstream. We made plans for a technical solution to test pipelines end-to-end to mitigate that fear, but the effort failed for cultural reasons. We eventually solved this challenge, but in a different context. In this presentation we will describe how we test full pipelines effectively by manipulating workflow orchestration, which enables us to make changes in pipelines without fear of breaking downstream.
Making schema changes that affect many jobs also involves a lot of toil and boilerplate. Using schema-on-read mitigates some of it, but has drawbacks since it makes it more difficult to detect errors early. We will describe how we have rejected this tradeoff by applying schema metaprogramming, eliminating boilerplate but keeping the protection of static typing, thereby further improving agility to quickly modify data pipelines without fear.
Predictably Improve Your B2B Tech Company's Performance by Leveraging DataKiwi Creative
Harness the power of AI-backed reports, benchmarking and data analysis to predict trends and detect anomalies in your marketing efforts.
Peter Caputa, CEO at Databox, reveals how you can discover the strategies and tools to increase your growth rate (and margins!).
From metrics to track to data habits to pick up, enhance your reporting for powerful insights to improve your B2B tech company's marketing.
- - -
This is the webinar recording from the June 2024 HubSpot User Group (HUG) for B2B Technology USA.
Watch the video recording at https://youtu.be/5vjwGfPN9lw
Sign up for future HUG events at https://events.hubspot.com/b2b-technology-usa/
The Building Blocks of QuestDB, a Time Series Databasejavier ramirez
Talk Delivered at Valencia Codes Meetup 2024-06.
Traditionally, databases have treated timestamps just as another data type. However, when performing real-time analytics, timestamps should be first class citizens and we need rich time semantics to get the most out of our data. We also need to deal with ever growing datasets while keeping performant, which is as fun as it sounds.
It is no wonder time-series databases are now more popular than ever before. Join me in this session to learn about the internal architecture and building blocks of QuestDB, an open source time-series database designed for speed. We will also review a history of some of the changes we have gone over the past two years to deal with late and unordered data, non-blocking writes, read-replicas, or faster batch ingestion.
STATATHON: Unleashing the Power of Statistics in a 48-Hour Knowledge Extravag...sameer shah
"Join us for STATATHON, a dynamic 2-day event dedicated to exploring statistical knowledge and its real-world applications. From theory to practice, participants engage in intensive learning sessions, workshops, and challenges, fostering a deeper understanding of statistical methodologies and their significance in various fields."
4. This would be a great
opportunity to start off the
session with a
recommendation, or at least
an acknowledgement that
the user has returned
(The fold)
5. (The fold)
This would be a great
opportunity to start off the
session with a
recommendation, or at least
an acknowledgement that
the user has returned
This could be a place to
show off new
recommendations to kick
off the users’ new session,
but unfortunately it’s below
the fold
6. Currently Zillow is totally
search-focused, which may
be a reflection of historical
user behavior. Let’s continue
our last search and see
where how that goes…
10. But, how are the featured
listings determined? All the
top “Featured” listings have
special offers. Do
sellers/realtors pay for this
ranking? Is this in any way
actually helping me find
what I want and making my
experience better?
Switching to “Featured”
brings up better looking
photos, and at first glance
seems to be a better starting
point
12. Ok, Let’s take a look at this one…
There’s a TON of great info here
that could be used in a
personalized recommender.
Price, Zestimate, Special Offers,
description, facts and features.
Scrolling further down there’s 7
more sections of different data
about the property, all of which
is super valuable to build a
content-based profile of each
user.
13. Ok, Let’s take a look at this one…
There’s a TON of great info here
that could be used in a
personalized recommender.
Price, Zestimate, Special Offers,
description, facts and features.
Scrolling further down there’s 7
more sections of different data
about the property, all of which
is super valuable to build a
content-based profile of each
user.
14. We’ve got all the property
data we could ever hope
for. It’s an embarrassment
of riches. That’s why
Zillow is so great for
searching… but that’s not
enough to build a REALLY
effective recommender.
15. We need:
PREFERENCE DATA!
We’ve got all the property
data we could ever hope
for. It’s an embarrassment
of riches. That’s why
Zillow is so great for
searching… but that’s not
enough to build a REALLY
effective recommender.
16. We’ve got all the property
data we could ever hope
for. It’s an embarrassment
of riches. That’s why
Zillow is so great for
searching… but that’s not
enough to build a REALLY
effective recommender.
We need:
PREFERENCE DATA!
There’s only one mechanism
for users to explicitly tell
Zillow what they want – the
“Save” and “Hide” buttons
17. We’ve got all the property
data we could ever hope
for. It’s an embarrassment
of riches. That’s why
Zillow is so great for
searching… but that’s not
enough to build a REALLY
effective recommender.
We need:
PREFERENCE DATA!
There’s only one mechanism
for users to explicitly tell
Zillow what they want – the
“Save” and “Hide” buttons
These buttons aren’t
featured prominently, and at
no point has the UI
encouraged me to use them.
18. We’ve got all the property
data we could ever hope
for. It’s an embarrassment
of riches. That’s why
Zillow is so great for
searching… but that’s not
enough to build a REALLY
effective recommender.
We need:
PREFERENCE DATA!
There’s only one mechanism
for users to explicitly tell
Zillow what they want – the
“Save” and “Hide” buttons
These buttons aren’t
featured prominently, and at
no point has the UI
encouraged me to use them.
Furthermore, there’s no way
to differentiate between “I
LOVE THIS HOUSE” and
“Ehhh, I’ll look again later”
19. Aside from explicit preference
information, we also have several
mechanisms to measure implicit
user preference…
20. Aside from explicit preference
information, we also have several
mechanisms to measure implicit
user preference…
How many times have they
looked at this property’s details?
How much time have they spent
on this property? How many
times has this property shown
up in different searches they
conducted?
21. Aside from explicit preference
information, we also have several
mechanisms to measure implicit
user preference…
Did they look at all the pictures?
Did they look at them more than
once? Is this more than they
usually look?
How many times have they
looked at this property’s details?
How much time have they spent
on this property? How many
times has this property shown
up in different searches they
conducted?
22. Aside from explicit preference
information, we also have several
mechanisms to measure implicit
user preference…
Did they look at all the pictures?
Did they look at them more than
once? Is this more than they
usually look?
How many times have they
looked at this property’s details?
How much time have they spent
on this property? How many
times has this property shown
up in different searches they
conducted?
How far down did they scroll? Is
there a particular section they
always expand?
23. Aside from explicit preference
information, we also have several
mechanisms to measure implicit
user preference…
Did they look at all the pictures?
Did they look at them more than
once? Is this more than they
usually look?
How many times have they
looked at this property’s details?
How much time have they spent
on this property? How many
times has this property shown
up in different searches they
conducted?
How far down did they scroll? Is
there a particular section they
always expand?
And the most important of all –
Did they fill in their info and click
CONTACT AGENT?
24. Email is another type of
interaction that can be very
helpful for the user, as well as a
mechanism to obtain more
implicit preference data.
25. Email is another type of
interaction that can be very
helpful for the user, as well as a
mechanism to obtain more
implicit preference data.
A few days after my last search I
get an email showing me some
new properties that match my
search filter.
26. Here’s one other type of
interaction that can be very
helpful for the user, as well as a
mechanism to obtain more
implicit preference data.
A few days after my last search I
get an email showing me some
new properties that match my
search filter.
Any property I click helps to
refine my interests which is
great. On the other hand, if the
properties they show aren’t
good fits I’m likely to
unsubscribe and close this data
stream for this user.
27. Here’s one other type of
interaction that can be very
helpful for the user, as well as a
mechanism to obtain more
implicit preference data.
A few days after my last search I
get an email showing me some
new properties that match my
search filter.
Any property I click helps to
refine my interests which is
great. On the other hand, if the
properties they show aren’t
good fits I’m likely to
unsubscribe and close this data
stream for this user.
The fact that they’re showing me
rentals when my saved search
was properties for sale isn’t a
good indication that these are
personalized at all.
28. Recommender Gameplan
Step 1:
Add an optional wizard as part of the onboarding process. Think
Netflix:
- First they allow users to explicitly define which categories they
like and don’t like.
- Then, they actively encourage users to provide ratings which
are extremely useful for both content-based and collaborative
filtering recommenders.
29. Recommender Gameplan
Step 2:
Add a “Progress Bar” or some visual indication to the user of how
many properties they’ve rated, similar to the “Profile completed”
indicator on LinkedIn:
- Encourage users to rate lots of properties
- Explain the benefit of doing so clearly
- Research the number of ratings required for a reasonably
accurate preference model
30. Recommender Gameplan
Step 3:
Add a 5-point rating scale to each property:
- This is the most useful explicit preference data that can be
obtained, and is much better than just “save” or “hide”
31. Recommender Gameplan
Step 4:
Add a new “Recommended” tab to the map view:
- Display properties that match the current search criteria/map
view, ordered by predicted preference
- Create a content-based recommender for each property that
predicts interest in the property based on viewer characteristics.
- Create a collaborative-filtering/nearest neighbor model to find
similar users
- Combine both models for extra accuracy
32. Recommender Gameplan
Step 5:
Only send when high-probability properties are discovered:
- Stop sending new properties in daily emails that (partially?)
match the most recent search criteria
- Should increase the conversion rate and decrease the
unsubscribe rate