It seems you can’t go a day without reading or hearing about social’s newest sweetheart – Pinterest. The social network has exploded in the last few months. To get a handle on the opportunities, we needed to sift through the buzz, explore the usage patterns and analyze the data. So we worked with HPCC Systems from LexisNexis® Risk Solutions – an open source, enterprise-proven Big Data analytics provider.
Pinterest: Is Social Media's Newest Sweetheart the Real Deal?
1. Is Social Media’s Newest Sweetheart the Real Deal?
@interpolate @teresacaro
DOWNLOAD THE PAPER
www.engauge.com/insights/pinterest
#pingauge #pingauge
7. AGENDA
• Definition: What is Pinterest?
• Trends: What is Driving the Pinterest Craze?
• The Data
• The Opportunity
• HPCC Systems for LexisNexis® Risk Solutions
#pingauge
12. CREATING DIY SUPER HEROS
A recent Engauge Power Panel study* showed that
90% of people said they use Pinterest to
get ideas. Some even said that it makes them “feel
like they can do anything.”
*External and internal polling group
@teresacaro #pingauge
15. LIFE IS FLUID & NEEDS TO BE SHARED
• Volume & availability of content
• Niche communities
• Growing comfort & interest in curating and sharing content
Planning Experiencing Recapping
@teresacaro #pingauge
16. CONTENT CREATION TO CURATION
“People stare into a fire hose of information every
day, and it’s having an impact. They’re actively seeking
ways to not only filter and organize what they find,
but also to less stressfully consume more content.”
@teresacaro #pingauge
18. METHODOLOGY
We obtained the data below by reviewing publicly accessible Pinterest
user profiles and cleaned, processed and analyzed the data using the
HPCC Systems Big Data platform.
The bulk of the data discussed in this paper was obtained from spidering through
Pinterest user profiles obtained from Google search results and scraping the relevant
HTML. We also used frequency lists of gender and forename provided by the U.S.
Census to help determine the gender of Pinterest users. The scraping method was
chosen because Pinterest did not have an API at the time analysis began. After this
project had started Pinterest briefly announced an API, then pulled all information
about it, but left certain parts of the API publicly accessible.
@engauge @interpolate
20. WHERE PINS COME FROM
The top ten sources of pins by number of pins with the percentage of
total pins, and percentage of the top 10.
@engauge @interpolate
21. SOURCES OF PINS BY PERCENT OF TOTAL
AND PERCENT OF TOP 10
@engauge @interpolate
22. CATEGORIES
* Top categories and percent of Top10 by category.
@engauge @interpolate
28. CONNECTIVE TISSUE
Your website is your hub. Social is the connective tissue.
@teresacaro #pingauge
29. YOU’RE ALREADY THERE
You don’t need a brand page to have a presence. People might
be pinning your stuff already.
Even if you don’t
have a page, you
might be there
anyway.
@interpolate Exchange your URL: http://pinterest.com/search/?q=coca cola #pingauge
31. FEMALE BUYING POWER
Women control 80% of household discretionary
spending (according to the U.S. Census Bureau).
Clearly, if women are important to your brand, then
you want to reach them here.
@teresacaro #pingauge
32. FEMALE BUYING POWER
Women control 80% of household discretionary
spending (according to the U.S. Census Bureau).
Clearly, if women are important to your brand, then
you want to reach them here.
Yet, women aren’t the only ones
@teresacaro #pingauge
33. FEMALE BUYING POWER
Women control 80% of household discretionary
spending (according to the U.S. Census Bureau).
Clearly, if women are important to your brand, then
you want to reach them here.
Yet, women aren’t the only ones
on Pinterest...
@teresacaro #pingauge
34. “BOARD OF MAN”
Started on a
lark, it now has
211,000
followers.
@interpolate #pingauge #pingauge
36. WHAT THEY’RE DOING RIGHT
• Humanizing their brand
• Promoting yogurt consumption through recipes
• Allowing users to self-select content
• Limiting the number of pinboards and
pins
@teresacaro #pingauge
42. FINAL WORD
• Know your consumers.
Where they want to engage and why
• Fully optimize your website and its content.
Your pictures are revenue drivers
• Leverage social to connect it all together.
An integrated effort drives powerful results
#pingauge
@engauge
44. INTRODUCTION
http://hpccsystems.com
LexisNexis® Risk Solutions
More than 15 years of Big Data experience
Provides information solutions to enterprise customers
Generates $1.4 billion in revenue (2010)
HPCC Systems
Launched in June 2011
Open source, and enterprise-proven Big Data analytics platform
Helps enterprises manage Big Data at every step in the Complete Big Data
Value Chain
Risk Solutions
40
45. HPCC and Machine Learning
http://hpccsystems.com
HPCC is a massive parallel-processing computing
platform
Truly Parallel Machine Learning Algorithm Execution
E
S
P
Risk Solutions
46. THE COMPLETE BIG DATA VALUE CHAIN
http://hpccsystems.com
Discovery &
Collection Ingestion Integration Analysis Delivery
Cleansing
Collection – collecting structured, unstructured and semi-structured data
Ingestion – consuming vast amounts of data including extraction,
transforming and loading
Discovery & Cleansing - clean up, formatting and statistical analysis of the
data
Integration – linking, indexing and data fusion
Analysis – statistics and machine learning
Delivery – querying, visualization, and redundancy, enterprise-class
availability
Risk Solutions
42
47. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
Risk Solutions
43
48. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
You surely can’t glance over every record;
Risk Solutions
43
49. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
You surely can’t glance over every record;
And it may not even have records…
Risk Solutions
43
50. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
You surely can’t glance over every record;
And it may not even have records…
What if you wanted to learn from it?
Risk Solutions
43
51. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
You surely can’t glance over every record;
And it may not even have records…
What if you wanted to learn from it?
Understand trends
Risk Solutions
43
52. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
You surely can’t glance over every record;
And it may not even have records…
What if you wanted to learn from it?
Understand trends
Classify into categories
Risk Solutions
43
53. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
You surely can’t glance over every record;
And it may not even have records…
What if you wanted to learn from it?
Understand trends
Classify into categories
Detect similarities
Risk Solutions
43
54. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
You surely can’t glance over every record;
And it may not even have records…
What if you wanted to learn from it?
Understand trends
Classify into categories
Detect similarities
Predict the future based on the past… (No,
not like Nostradamus!)
Risk Solutions
43
55. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
You surely can’t glance over every record;
And it may not even have records…
What if you wanted to learn from it?
Understand trends
Classify into categories
Detect similarities
Predict the future based on the past… (No,
not like Nostradamus!)
Machine learning is quickly establishing as an
emerging discipline.
But there are challenges with ML in big data:
Thousands of features
Billions of records
Risk Solutions
43
56. MACHINE LEARNING IN BIG DATA
http://hpccsystems.com
How do you extract value from big data?
You surely can’t glance over every record;
And it may not even have records…
What if you wanted to learn from it?
Understand trends
Classify into categories
Detect similarities
Predict the future based on the past… (No,
not like Nostradamus!)
Machine learning is quickly establishing as an
emerging discipline.
But there are challenges with ML in big data:
Thousands of features
Billions of records
The largest machine that you can get,
may not be large enough…
Get the picture?
Risk Solutions
43
57. ECL-ML: HPCC SYSTEMS MACHINE LEARNING
http://hpccsystems.com
A fully distributed and extensible set of Machine Learning techniques for
Big Data
State of the art algorithms in each of the Machine Learning domains,
including supervised and unsupervised learning:
Correlation
Classifiers
Clustering
Statistics
Document manipulation
N-gram extraction
Histogram computation
Natural Language Processing
Distributed and parallel underlying linear algebra library
Risk Solutions
44
58. TAKE AWAYS
http://hpccsystems.com
A fully parallel set of Machine Learning algorithms on Big Data
gives you full insight
Outliers matter, especially when those outliers are the exact
reason for the discovery effort (for example, in anomaly
detection)
Dimensionality reduction can conduce to information loss:
why risk losing valuable information when you can have it all?
Leveraging a fully parallel machine learning solution on Big
Data will help you find fraud, bring products to market faster,
and become more competitive
Organizations that continue to do machine learning on
summarized data risk losing ground to their competitors
Risk Solutions
45
60. THANK YOU.
Engauge is a full-service marketing agency for the
digital and social age. We help grow our clients'
businesses by leveraging creativity and technology to
connect brands and consumers through the most
relevant content and channels.
WWW.ENGAUGE.COM
ATLANTA, COLUMBUS, ORLANDO, PITTSBURGH
#pingauge