analytics crawl before walk
Post on 16-Apr-2017
64 Views
Preview:
TRANSCRIPT
CRAWL BEFORE WALK: ANALYTICS BIG PICTURE
MANGESH CHAUDHARI
111
LAYERS OF ANALYTICS
The Essentials
Advanced Level/Nice To Have
Make Things Talk To Each Other
Predict The Future & Compete
1
2
3
4
THE LANDSCAPEMarketing Experiences• Mobile Marketing• Display & Native Ads• Video Marketing & Ads• Search & Social Ads• Communities & Reviews• Email Marketing• Influencer Marketing• Social Media Marketing• Sales Enablement
• Events and Webinars• SEO• Customer Experience/VOC• Loyalty/Referral/Gamification• Personalization & Chat• Testing & Optimization• Interactive Content• Content Marketing• Creative & Design
Marketing Operations• Audience & Market Data• Channel/Local Marketing• Asset & Resource Management• Call Analytics• Team & Project Management• Vendor/Data Analysis• Performance & Attribution• Visualization• Web & Mobile Analytics• BI, CI & Data Science
Middleware* Data Management * Tag Management * Identity Management * Cloud Integration * APIs
Backbone Platforms* Platform * CRM * Campaign & Lead Management * Experience Management/Web Content * E-Commerce
Infrastructure* Database & Big Data * Cloud/IaaS/PaaS *Mobile App Dev & Marketing *Web Dev * Rest of the Internet!!
THE LANDSCAPEMarketing Experiences Marketing Operations
Middleware
Backbone Platforms
Infrastructure
18 10
555
43
COMMON PITFALLS• Translation of business objectives at each level into metrics• Expecting analytics as one time exercise• Fear of experimenting• “I’m not technical” OR “I need insights to make decision”• Impatience - “We want everything now!”
• Budget restrictions OR desire for free• Select vendor with longest list of functionality• Not developing skills set within the organization
My Data IS
Than your Data
Bigger!
My Business IS
Faster!Than You Business
Clear CommunicationChannels From Management
Less Organized Businesses Running On Gut Feeling
Adapted from: Analyzing the Analyzers: An Introspective Survey of Data Scientists and Their Work by Harlan Harris, Sean Murphy, Marck Vaisman
Coding/Computing
Statistics
DomainKnowledge
Big Data
OperationalMetrics
TraditionalResearch
MachineLearning
CASE STUDY: NEWS MEDIA
TRADITIONAL ORGANIZATION STRUCTUREAdvertising Sales
/Business DevelopmentEditorial
Sing
apor
e
Asia
Wor
ld
Politi
cs
Busin
ess
Tech
nolo
gy
Life
styl
e
Opi
nion
Spor
ts …
Auto
s
PrintingPress Distribution Subscribers
Research
DIGITAL NEWS ORGANIZATION
Technology Team(Development+AdOps)
Social Search Desktop Mobile Video
EditorialSinga
pore
Asia W
orld
Politi
cs
Busin
ess
Tech
nolo
gy
Life
styl
e
Opin
ion
Spor
ts
Ad Sales/Business Development
MATRIX OF METRICS
Management How much revenue will it make in 6 months?
Ad Sales
Editorial
Technology
Social/Search
Desktop
Mobile
Video
Audiences Captured &Revenue
LEVEL 1 LEVEL 2 LEVEL 3 LEVEL 4
Revenue
Visitors and Page Views
Page Views/Traffic Load
Referral Source andShare
Visitors, Visits, Page Views
Downloads, Visits,Page Views
Engagement MetricsBy Content Type
Views and Content Viewed
Engagement MetricsBy Content Type
Referral Source, ShareType, Keywords
Heat maps, AB Tests
Views by ContentBy Content Type
Revenue by Content Type & Platform
Revenue, GrowthNew Audiences
Rate Card, Throttling,Audience Segments
Ability To Look AtRevenue in Real-time
What Are Best PlacesTo Place Put Ads
What Format OfContent By Channel
Effectiveness Of Design& Performance
When and Where ToUse Maximum Traffic
When To Post?What To Post?
How To Drive MaxEngagement?
What Content TypeWorks Best For Each
Channel?
Engaging Phrases & Timing of Content
Continuous DesignImprovement
Sentiment Analysis, Timing and Competition
Required Traffic To MeetRevenue Goals?
Feedback To Ad-OpsFor Throttling?
Recommendation Engine
Thank You!
top related