Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

A DataSci Proposal for [ ] in [ ]

Accompanying presentation for a proposal to [].

Mark Wall

on 19 August 2014

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of A DataSci Proposal for [ ] in [ ]

Glocal Social Network Analytics
Glocally Optimize
Acquire Cross & Upsell Retain
There is Elastic tension between social networks and real places
Local Measure & the Datification of Social Place
A Social Network
is made of:
A set of
: Individuals, objects, concepts usually over a time period
A set of
(interactions) between nodes.
Marketing models
Are either:
based on a social network conversation OR
location based targeting
No one has mastered
targeting... yet!
Valuable Information from a Network
of a node is crucial
Central nodes propogate more information

of it's links also
Is it well connected to nodes and places?

of it's neighborhood
Does it have churners? High value people?

It's role in the

Is it a
influencer, leader, networker, bridge?
Every Customer Interaction
A social network is an imagined community.
They are
dynamic ... and can be
tied to places
A social network is a
of nodes and links
Directed - Elanor visited Stacey
Undirected - Elanor and Stacey talked
Weighted - Elanor visited Stacey for a week
It can be directed or undirected
It can be very fluid over space and also time
Value of Role in the
Social Network
Increasing brand loyalty using place awareness and attachment
Which places do you feel most attached to?
Where you were born? Went to Uni? Work? Live?
A Location based network is made of
places have different uses and meanings at different times of day
Directed - F1 will be in Melbourne in May
Undirected - Qantas flies between Melbourne and Perth
Weighted - Qantas flights are discounted from Melbourne to Perth in May
network is a graph of nodes and links interacting locally on a global scale
A data science proposal for [Removed], [Removed]
A data science proposal for [Removed], [Removed], © Mark Wall, 2013

A set of
- places, landmarks, events, businesses: fixed in space

A set of
(interactions) between nodes.
because customers interact with everyone, behaviour can be
e.g. Influencer influences their friends...
or does not: some "viruses" spread, others do not
"Glocal" Methodology
Objectives of
Glocal SNA
are both descriptive and
Descriptive objectives
what happened
Improve knowledge of customer behaviour in socio-spacetime
Detect phenomenon propogation (product adoption, churn)
Detect interest group communities & roles ('leader', 'influencer', 'bridge' nodes.. )
Predictive objectives
what can happen
Improve predictive models (churn, cross-sell, fraud, customer aquisition, net promoter score ... ) by integrating
© Mark Wall, 2013. All Rights Reserved.
analysis or
based on
network evolution

over space and/or time
Building Data Science Into The Product
After you've talked to customers they..
talk to each other
and talk to
in their social networks
in their
Global and local business needs to
accurately measure this social - local
These are
They span:
2 communities
2 places
A combination

are access paths to new customers, cross sells, improved loyalty
Other interesting roles can be identified:
Local Leader
Social influencer

Glocal Communities and Roles
is a set of nodes that can be detected from a dataset
refers to the dynamic evolution of a community as it interacts globally and locally
Source: Granovetter, 1973: The Strength of Weak Ties
Step 1: Detect the community using entity resolution, location analysis and 'SNA' algorithms.
Step 2: Track and predict interactivity over time & space.
Step 3: Capitalise on new information as it arises.
Local leader
Social influencer

Glocal-Social graph Interaction
New opportunities for innovation
Graph based recommendation engines
Local + Social Graph interactive models
Social Graph
Glocal Graph
More fluid, dynamic
Boost Aquisition
Acquire new clusters of customers
Predict "honeypots" over space time as networks evolve and interact
Identify "Off-Net" potential customers with high intensity connections
Space is static
can be fluid e.g. a bar can be a restaurant 8am-3pm
Next best activity
: e.g. "Looks like you're not in Sydney that week? How about this event in Melbourne instead?"
Localise brand
: Launch new product and services (Up sell and Cross sell through better local market segmentation
Understand Satisfactions
and Gripes: Sentiment and Localised mapping
of Interest Groups-
commonalities of culture,
nationality, language.. any
other group to target
over space-time
There are local reasons why things go viral. How can we model them?
Next steps...
Whiteboard Session

Confirm options Outlined in Proposal document
Plan a session to outline the proposed technology stack
Architecture transition
Options for working together are described in more detail in the accompanying word doc.
Options are suggestions...
For your consideration
This Prezi has been modified from the original for public hosting.
Full transcript