marketsoft and marketing cube data quality to cc-v3
TRANSCRIPT
Customer-centricity starts
with Data Quality
Customers and data
Introduction and what to expect:
Data quality 101
Data quality frameworks
Making the most of 1st, 2nd, and 3rd party data
Practical experiences
Introductionand what to expect…
Introduction
Big DataOmni-channel
Real-time
Seamless Integration
Low Hanging Fruit
Transparency
What not to expect!
Introduction
Data
Quality101
“Degree of excellence exhibited
by the data in relation to the
portrayal of the actual scenario”Wikipedia
Data Quality…
“The state or desired state at
which data assets realise their
maximum potential”Marketsoft
Data Quality…
Data Quality…
“Data is the new oil” Clive Humby
The good
Data Quality…
Best Practice
Data Quality Best Practices
boosts revenue by 66%
Goes a Long Way
If the median Fortune 1000 company were to
increase the usability of its data by 10%
company revenue would be expected to
increase by $2.01 Billion dollars.
Data Quality…
The bad
Dirty Data Barrier to Entry
$600B
The cost of bad or ‘dirty” data
exceeds $600 Billion dollars
for US businesses annually. of survey respondents
cite Data Quality as a
46%
for adopting BI/Analytics
products.
BARRIER
Data quality has impacts across this entire journey
Data Quality…
Strategic Journey
Segmentation
Lifecycle
Timing/frequency
Platform(s)
Device
Treatment
Message
Offer
Quality
Data
QualityFramework
Traditional approaches
1. Do nothing (despite the risk)
2. Enterprise data quality
3. Ad-hoc using common tools (Excel)
4. Outsource
Framework…
Data Quality Maturity
Framework…
Valu
e
Difficulty
Ad-hoc
Reactive
Defined
Proactive
Predictive
Awareness
1st Party Data
Marketing
• Affinities
• Behaviors
• Location
• Identity
Business
• Propensities
• Experiences
2nd Party Data
Partner
• Affinities
3rd Party Data
External
• Demographics
• Missing data
Data Quality implications…
Consideration
1st Party Data
Marketing
• Triggers
• Segmentation
• Value
Business
• Suggested
Treatments
2nd Party Data
Partner
• Personalisation
3rd Party Data
External
• Event triggers
Purchase
1st Party Data
Marketing
• Experience
• Cross-sell/upsell
Business
• ROI
2nd Party Data
Partner
• Optimisation
3rd Party Data
External
• Credit
• Risk
Advocacy
1st Party Data
Marketing
• LTV
• Next Best Offer
Business
• Customer
experience
• Support
2nd Party Data
Partner
• Influence
3rd Party Data
External
• Influence
• Nurturing
1st Party Data
Marketing
• Analytics
• Marketing
automation
• CRM
Business
• Sales
• Customer
service
2nd Party Data
Partner
• Partner CRM
• Partner
Analytics
3rd Party Data
External
• <various>
Customer JourneyData Points
Framework…
Discovery,
Benchmark and
Audit audit and benchmark
to assess current state
against desired state
Matchingfuzzy and analytical
matching
methodology
Monitoringreporting,
dashboards,
and KPIs
Batch & Real-time
deployment
Standardisation
data point and
field level
Map your data assets – data type, ownership, governance
• Owned marketing data
• Customer/CRM data – name and address, phone, email,
activities – affinity groups, conversion path
• Analytics – model scores, customer journey
• Marketing automation – behavior, engagement
• Owned business data
• Sales data – transactions - LTV, Risk class
• Customer service – sentiment, trends, leading indicators
• 3rd Party
• Data appends
• Census, ABS, macro data
Framework…
Discovery, Benchmark and Audit
Define a customer journey – arbitrarily or analytically
Map data assets to customer journey
Cluster by size and importance to customer experience
Audit and Benchmarking
1
2
3
4
5
Outcomes – Discovery, Benchmark and Audit for each data point…
1 Consistency
2 Completeness
3 Accuracy
4 Precision
5 Missing or unknown
Types of Data Quality Considerations…
Framework…
Discovery,
Benchmark and
Audit audit and benchmark
to assess current state
against desired state
Matchingfuzzy and analytical
matching
methodology
Monitoringreporting,
dashboards,
and KPIs
Batch & Real-time
deployment
Standardisation
data point and
field level
Reporting, Dashboard, and KPIs
In order to move beyond symptom treatment to a proactive approach,
a consistent monitoring layer is critical:
Types of Data Quality Considerations…
General data point changes
over time
Validation changes to benchmark
Data coverage of key fields
to benchmark
Key shifts in data consistency
Qualitative outcomes
PracticalExamples
Reactive
Case #1
Scenario: Digital marketing database used for automated email marketing
Data: Single database, digital focus, data quality never assessed
Practical examples
Data Quality Outcomes
1. Inconsistency: Identified and cleansed 10 fields with data variable
inconsistencies due to manual data capture
2. Incompleteness: Filled data gaps in key fields including title/gender,
address elements, and phone numbers
3. Accuracy: Identified 7% of hard-bounce email addresses; identified
15% of people who had changed companies based on public social
sources
4. Precision: 7% Duplication; also unrealised potential to segment at a
family, address, or true company level
5. Missing or unknown: 0.5% deceased customers; geo-demographic
profile information used to augment segmentation model
Practical examples
Ad-hoc
Case #2
Scenario: B2B sales focused organisation with new large, “thin”
prospect database, yet limited sales capacity to action
Data: Several new B2B prospect databases,
existing customers and pipeline
Practical examples
Data Quality Outcomes
1. Inconsistency: Identified key inconsistencies across the new
prospects, including level of company information provided
2. Incompleteness: Gaps in company details in known operating markets
3. Accuracy: Identified issues with key variables such as Employee Size
and applied model to improve accuracy by 43%
4. Precision: 80% Duplication; also unrealised potential to segment at a
true company level and to show parent to child company relationships
5. Missing or unknown: Appended and modeled key variables such
including employee size, annual revenue, operating markets, HQ
market, revenue velocity
Practical examples
KeyTakeaways
1 Define your objectives
2 Understand your data assets
3 Tie everything back to a customer experience
4 Establish a framework
5 Find a trusted data partner
Takeaways…
Thank you
“The state or desired state at which data assets
realise their maximum potential”
Marketsoft