Measurecamp 6 Workshop: Data Integration

Data Integration13th March 2015 09:30 – 12:30

Sean Burton

Overview

• A bit about me…

• Who’s in the room…

• Some background…

• Getting started…

• Exercise!

• Bringing it all together…

• Next steps…

• The wrap up.

0191 704 2045 | analyt.co.uk | [email protected] | @analytdata

Intros…

Sean Burton – [email protected] | @sean_d_burton & @analytdata | analyt.co.uk

I'm passionate about improving customer experience and business value by using

a blend of data, technology and psychology.

About me:

• Formerly the Director of Measurement at Seren Design Ltd.

• A 15 year career covering: eLearning, Content Management Systems,

Interaction Design, Product Management, Web Analytics, and Data

Visualisation.

• Extensive experience with FTSE 100 companies across financial,

telecommunication, gaming, and retail sectors.


Who’s in the room…


In the room…

• 16 people


Experience…

Technical Ability…

Some of the upfront questions

• How to you join data from different sources? What happens if the data sources aren’t linked?

• What happens when your customer journeys cover & cross multiple domains?

• How do you can you improve marketing performance/customer journeys using data?

• How do you get VoC data into Google Analytics?


Some background…


…the web is littered with abandoned baskets

‘Headline’ Conversion Rate – offline benchmarks

0

20

40

60

80

100

'High-ticket'

speciality

General

speciality

Footwear Clothing Department

store

Mixed-

goods

Grocery

Source: RCT Analytics Ltd

Online


Interesting finding…

“By improving user experience, a typical business can improve

sales by 64%.”

Hewson Group, 2003


The Key Business Processes

VisitAware Browse Engage Transact RepeatConsider

Acquisition Marketing

Site and Content Development

Retention Marketing


The challenge…

You

Affiliates

Survey

Data

Email data

Ad-serving

data

Transactions

Customer

data

Web

analytics

PPC data

Panel data

ISP data

Analyst

data

GRP data

Offline

sales data

Promotion

data

Performance

data


Familiar?


Everything should be made as simple as possible but

not simpler. “

”

Good KPIs are “Übermetrics”…

Good KPI

Strategic measures of

success

ActionableEasy to

understand

Based onvalid data


Key questions

customer journey profile

What do they buy?

Who visits and buys from the site?

How are they using the site?

Why do they visit the site?


Core data relating to CX

• Web Analytics – the what• Detailed traffic measurement tool showing volume of traffic across your digital

real-estate.

• Customer Relationship Management (CRM) – the who• A database with details of who your customers are, what they have bought,

and their communication preferences.

• Voice of Customer (VoC) / Surveys – the why• Responses to online or market research questionnaires. Typically run by

external agencies and hosted via an external tool.


Survey approach

site experience

visit intention

outcome versus intention

entry survey exit survey

expectations delivery against expectation

visit profile

visitor profile


What people say vs what they actually do...

Over a third of people say that they intend to buy something

Conversion rates are in the normal range

People say that they are going to visit 2.5 categories on average

On average they visit 1

Say Do

Nearly half of people say that they are likely or very likely to use the search function

Less than 20% of visits have a search in them


Customer Experience

• Do you know who your customers are?

• Do you listen to them?

• Do you know what they think of you?

• Are changes to your web site driven by people within the business rather than by the people that actually pay them – the customer?

Ultimately, it’s best to try and avoid…


Most people don’t like bad experiences

CO

NV

ERSIO

NS

ACQUISITION STRATEGIES

Acquisition & Conversion

USA

BILITY


“I got there in the end but it was a

struggle. Probably best try

www.acme.com”

…not what I expected

I’ll try another site

This is a nightmare!

Grrr, where’s the phone?

What a load of b*ll**ks!!

Optimising Processes – survey first

• Ask about reasons for visiting, i.e. task intent.

• Trigger survey at the end of key processes, e.g. purchase or registration.

• Capture trigger URLs and web analytics visitor cookie values.

• Segment survey data based on visitors failing to complete their task and then review their comments

• Use web analytics data to confirm traffic volumes for trigger URLs – this will allow you to scale the issue


Optimising Processes – analytics first

• Use web analytic data to find process steps with significant drop-out

• Segment out visitors that start process, but fail to complete.

• Cross-reference visitor IDs with survey data to see if any visitors had left feedback

• Use feedback to help isolate the problem


The Peak End Rule

• Proposed by Barbara Fredrickson and Daniel Kahneman

• Theory states that people remember the most intensely felt moments of an experience and the feelings at end

• Understanding how your customers are feeling at the end of a process is very powerful


http://en.wikipedia.org/wiki/Barbara_Fredrickson

http://en.wikipedia.org/wiki/Daniel_Kahneman

Peak End Rule

Duration

Exp

erie

nce


Getting started…


Getting started

• Basic• Simply show VoC alongside web analytic reports

and basic demographics from CRM

• The mid-point (correlate)• Overlay key VoC metrics, e.g. recommendation,

with key web analytics data, e.g. orders

• The goal (associate)• Fully integrated customer feedback covering a

range of data sources


Integrating CRM & Analytics

• Many Analytics providers can now do this ‘out of the box’, e.g. Adobe’s (Omniture) Genesis platform

• Common approaches are to:• Capture a user ID when a user logs into the site based on their CRM record,

which can then be used to link their clickstream data with their CRM record

• Selectively write out details of the user’s CRM record to the web site, allowing these data to be captured in the web analytics tool as custom fields, e.g. Gender, Age, job role, etc.

• Make sure you’re completely transparent with your users as to how you store and use their personal data – do not breach their trust!


Integrating VoC & Analytics

• Many VoC providers now offer Google Analytics integration ‘out of the box’ and some support other vendors as well

• I’ve personally worked with eDigitalResearch, OpinionLabs, SurveyMonkey, SurveyGizmo, and Cvent

• Many of the above, simply require you to give them your tracking details and they can then pass back survey response data into your web analytics.

• Alternatively, you can pass a user ID over to the survey, and so long as the survey provider allows you to export raw response level data you can then integrate with your web analytics at a later point.


Example Survey capture

Web site feedbackWe’d love to here your thoughts on how we can

improve our site

Yes | No

Site Survey

1. Which of the following best describes your reasons for visiting the site today:

• [Drop-down of key tasks]

2. Where you able to complete this?• [Drop-down: Yes, Partially, No]

3. How easy did you find it?• [Single select on scale 0-10]

4. Would you recommend this site to your family and friends?

• [Single select on scale 0-10]

5. Any other comments?• [Text box]


Example data capture – Google Analytics

<script>

var _gaq = _gaq || [];

_gaq.push(['_setAccount', 'UA-XXXXX-X']);

_gaq.push(['_setDomainName', 'analyt.co.uk']);

_gaq.push(['_setAllowLinker', true]);

_gaq.push(['_setCustomVar',1, ’Task Intent', ’Buy fireplace', 2]);

_gaq.push(['_setCustomVar',2, 'Task Completion', ’yes', 2]);

_gaq.push(['_setCustomVar',3, 'Recommendation', 9, 2]);

_gaq.push(['_setCustomVar',4, 'Easy Score', 7, 2]);

_gaq.push(['_setCustomVar',5, ’Comments', ’great site', 1]);

_gaq.push(['_trackPageview']);

(function() {

var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;

ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';

var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);

})();

</script>

Initial Google Analytics variables

• setAccount = defined the profile to send the data to

• setDomainName = sets the scope of the cookie

• setAllowLinker = allows GA to work across multiple

domains

Sets Google Analytics custom variables

• Using all 5 ‘free’ custom variables (Premium has 50)

• Task Intent, Completion, Recommendation, and

Easy Score all set to whole visit

• Other possible items of interest: Gender, page

ratings, etc.


Example…


Exercise…


Exercise (30 mins)

• In groups of 4… (10 mins)

• Choose a site you all know

• What does ‘good’ look like for them?

• What might they want to measure?

• What would an ideal blend of CRM + VoC + Analytics?

• How would you present it back to the business? To an executive audience/

You’ll then present back to the group for discussion (20 mins)


Bring it all together…


Bring it all together

• Use your web analytics tool as a hub for much of your analysis

• You CRM tool can act as the hub for acting on the data, for example pushing updated email lists to your email provider

• Create a measurement framework about what ‘good’ looks like for your site. Capture these events as success goals and associate these successes with the users CRM record. Equally, make sure that you capture pain points so you can proactively resolve them.

• Publish key demographic details from your CRM or VoC responses back to your site so that these can be tracked in your analytics. Ideally, do this via a non-PII ID stored in a cookie, to allow multi-visit analysis.


Customer data integration


The Benefits

Increase Revenues

More transactions

Greater transaction values

Higher advertising revenues

Stronger brand equity

Decrease Costs

Less support costs

Higher customer retention

Greater channel efficiency

Lower development costs


ROI

Return

25.5m

2.75%

£56.70

£38.98m

£5.23m

Scenario 1 Scenario 2

% increase

5%

15%

10%

% increase

2%

10%

5%

Return

26.25m

2.87%

£59.40

£44.75m

£11m

Baseline

Visits 25m

Conversion Rate 2.5%

Order Value £54.00

Revenue £33.75m

Uplift


A few examples…

• Telco client• Improved customer service channel by integrating VoC & Web Analytics into

CRM

• Gaming client• Used VoC & Web Analytics to highlight key pain during launch of a new betting

platform

• Banking client• Used VoC tracker to highlight key drivers of satisfaction across wide range of

product categories


Case Study…

For every hour Apple spent calling their Detractors they generated $1,000 in additional revenue

http://www.forbes.com/sites/stevedenning/2011/08/26/another-myth-bites-the-dust-how-apple-listens-to-its-customers/


The next step…RFML Analysis


RFML Analysis

RecencyThe elapsed time since the last purchase made by a customer

Metric: Days since last visit

Frequency

The average time between purchases for a customer

Metric: Average no. of days between purchase

MonetaryThe total value of purchases made by a customer

Metric: Total order value or Average order value

LatencyThe time between two events, such as first visit and purchase

Metric: Days between first visit and purchase


RFML – Visitor Scoring Recency Score:

0: Visitor that have not visited the site within six months

1: Visitors that have visited the site within the last six months

2: Visitors that have visited within the last month

3: Visitors that have visited within the last week

Frequency Score:

0: Visitors that visit the site less than once every six months

1: Visitors that visit the site more than once every six months

2: Visitors that visit the site more than once a month

3: Visitors that visit the site more than once a week

Monetary Score:

0: Visitors that have never purchased

1: Visitors that have purchased at least £100



Latency Score:

0: Visitors that have never purchased

1: Visitors that purchased within six months of their first visit

2: Visitors that purchased within one month of their first visit

3: Visitors that purchased within one week of their first visit


Visitor Bandings (sum of scores):

0-1: Time wasters

1-5: Casual glances

5-9: Engaged Users

9-12: Stars

Time wasters

58%

Casual Glancers

23%

Engaged Users10%

Stars9%

Case Study…

Group ARecency: Visited within last week

Frequency: Visits 6 or more times a month

Monetary: Average order value of £15

Latency: Less than a week

Group BRecency: Within the last 6 months

Frequency: Less than once a month

Monetary: Average order value of £350

Latency: More than 6 months

http://images.google.com/imgres?imgurl=http://racked.com/2007_03_Staples.jpg&imgrefurl=http://racked.com/archives/2007/03/29/staples_to_fulfill_greenpointe.php&h=294&w=590&sz=43&hl=en&start=1&sig2=S3yZ3UedkzzGvyqnPJsY-A&tbnid=uELiU276rGqZgM:&tbnh=67&tbnw=135&ei=Pf3oR-zbKZXWgQKxzMHeCA&prev=/images?q=staples&gbv=2&hl=en

http://images.google.com/imgres?imgurl=http://racked.com/2007_03_Staples.jpg&imgrefurl=http://racked.com/archives/2007/03/29/staples_to_fulfill_greenpointe.php&h=294&w=590&sz=43&hl=en&start=1&sig2=S3yZ3UedkzzGvyqnPJsY-A&tbnid=uELiU276rGqZgM:&tbnh=67&tbnw=135&ei=Pf3oR-zbKZXWgQKxzMHeCA&prev=/images?q=staples&gbv=2&hl=en

Case Study…

http://images.google.co.uk/imgres?imgurl=http://blog.kir.com/archives/images/dell_logo020307.jpg&imgrefurl=http://nailo19.wordpress.com/2008/02/29/dell-takes-on-eee-pc/&h=480&w=640&sz=33&hl=en&start=12&sig2=xsy99nIQiIZ1mMOuVRNNug&um=1&tbnid=vDCkrslBpqF2fM:&tbnh=103&tbnw=137&ei=S-boR67FHpaigALWyKXeCA&prev=/images?q=dell&um=1&hl=en&rls=SNYJ,SNYJ:2005-31,SNYJ:en&sa=N

http://images.google.co.uk/imgres?imgurl=http://blog.kir.com/archives/images/dell_logo020307.jpg&imgrefurl=http://nailo19.wordpress.com/2008/02/29/dell-takes-on-eee-pc/&h=480&w=640&sz=33&hl=en&start=12&sig2=xsy99nIQiIZ1mMOuVRNNug&um=1&tbnid=vDCkrslBpqF2fM:&tbnh=103&tbnw=137&ei=S-boR67FHpaigALWyKXeCA&prev=/images?q=dell&um=1&hl=en&rls=SNYJ,SNYJ:2005-31,SNYJ:en&sa=N

Experiences spread quickly nowadays

Sale

Lift

Profitability

Loyalty

1

2

3

4

5 Advocacy

Integrated customer feedback

Automated technical

issues email sent to

operational teams

Customers with low

CSAT scores

contacted by

Customer Services

Customer invited to

community

Measure Act Advocate

Customer

encouraged to use

Social Media

Web Analytics

and/or channel data

Operational

Monitoring

Customer

Experience Tracker

Measure CSAT + Operational Performance

Analyse data to establish scale of impact and scope of recommended actionsContact Customer to explain

and/or resolve issue

Encourage customers to

join community

Monthly CX

Dashboard

Quarterly Workshop

Free Text Mining

/ Verbatim Analysis

Daily/Weekly Rapid

Reporting

UX Review / lab

research / EEG

Operational teams

fix issues based on

cost/benefit

Statistical analysis &

modelling

Segmentation &

Process

Optimisation

Training &

customised reports

to improve use of

data

Operational teams

investigate technical

issues

Data sources Techniques Outputs / Deliverables Operational activities Advocacy

The wrap up.

The wrap up

Accept your baby may be ugly!

Be prepared to ‘watch’ and ‘listen’

Ensure ongoing measurement

The wrap up (cont)

Create remarkable moments

Act quickly and respond to change

Act quickly and respond to change

Understand

MeasureFix & Act

Advocate

A few helpful links…

• Passing Web Analytics Data to CRM (Adam Greco)

• Blog: http://adam.webanalyticsdemystified.com/2010/05/10/crm-integration-passing-web-analytics-data-to-crm/

• YouTube: http://online-behavior.com/emetrics/integrating-web-analytics-and-crm-739

• Great blog detailing how to tie Google Analytics and SalesForce together

• http://www.e-nor.com/blog/google-analytics/integrate-salesforce-and-google-analytics

• A great blog by the wonderful Avinash with tips on choosing a survey providers

• http://www.kaushik.net/avinash/eight-tips-for-choosing-a-online-survey-provider/

• And of course vendor support pages, for example Google Analytics:

• https://developers.google.com/analytics/devguides/collection/gajs/gaTrackingSite

• Data Import(Dimension widening) in Google Analytics

• https://support.google.com/analytics/answer/3191589

http://adam.webanalyticsdemystified.com/2010/05/10/crm-integration-passing-web-analytics-data-to-crm/YouTube

http://online-behavior.com/emetrics/integrating-web-analytics-and-crm-739

http://www.e-nor.com/blog/google-analytics/integrate-salesforce-and-google-analytics

http://www.kaushik.net/avinash/eight-tips-for-choosing-a-online-survey-provider/

https://developers.google.com/analytics/devguides/collection/gajs/gaTrackingSite

https://support.google.com/analytics/answer/3191589

[email protected]

@analytdatatwitter

webanalyt.co.uk

http://www.surveygizmo.com/s3/1800143/MeasureCamp-V-Training

Training Feedback

http://www.surveygizmo.com/s3/1800143/MeasureCamp-V-Training

Measurecamp 6 Workshop: Data Integration

Data & Analytics

voc data

customer experience

data visualisation

blend of data

analytdata18core data

data sources arent

product management

cxweb analytics