Data Meets Numberwang

We Love Data

Plouffey is our hero

BEHIND BRITISH PERFORMANCE

Doubled the size of the brand with less investment over time Fig 1 - We Have Doubled the Size of the Brand Whilst Investing Less in Marketing 2004-2007 400%

30%

Sales Value as a % of 2000

25% 300% 20% 250%

200%

15%

150% 10% 100% 5% 50%

0%

0% 2000

2001

2002 Sales as %2000

2003

2004

ATL as %sales

2005 Promotion as %sales

2006

2007

ATL & Promotions as a % of Sales

350%

Where data? Client Database

TNS / Neilsen/IRI

Tracking Bespoke Quant

Desktop Tools Weekly / monthly sales

Econometrics

Web Analytics Google Reports to city

Aren’t we all about method?

Brilliant planners zig zag

Qual

Quant

Qual

The Eon Example

Plough through regulator’s reports

Why is all this stuff so blokey?

Adult Panel Who does energy?

RWStreet Who does energy?

Touchpoints When bills paid?

The Family Energy People

The Holy Trinity Opinion

Qualitative Research

Data Interrogation

The future of planning at MediaCom Knowledge

Ethnography

Business Science

Strategic Planning + Activation Agility

Example 1: No stone unturned to sell our strategy to a sceptical client

A brand that’s about social sharing

Method to deliver

A hunch about social gaming

Games industry data to quantify

Guardian Guide survey to add weight

Combining TGI, Caviar.

Example 2: Skoda – the importance of WOM People talk more about Skoda (and not as positively as they do about the competition)

A Fusion of TGI/NCBS confirms this Owners/buyers

Prospects

Talked to a few family and friends*

i114

i113

Friends/family recommendation

i193

i169

I know a (small) amount*

i100

i114

Already had good knowledge

i119

i94

Owners are more likely to ‘know their stuff’ Manufacturer’s website Web independent reviews General web Newspaper advertising

*Car category

i109 i160 i142 i101

i139 i196 i216 i119

Prospects seek out more information that Skoda buyers Source – TGI/NCBS

Example 3: Mars Food segmentation

-Penetration -Headroom -Switch -Basket -Trends

Know Where Example 4:You EFDStand – Establishing the job for comms

Example 5: Helping clients make sense of their own data Top 10 Drivers of Brand Preference (all brands) Understands its customers needs A brand you can trust Always first with innovative ideas and products A brand that will be popular in the future Stands out from other brands Cares about its customers Makes products that feel good in your hand Has beautifully styled products Makes phones you are happy to be seen with Honest

0.25

0.2

0.15

0.1

0.05

0 Q1 '07

Source: On-Track

Q2 '07

Q3 '07

Q4 '07

Q1 '08

Q2 '08

The conclusion was that showing and demonstrating the sexiest Nokia phones was required Nokia’s highest-scoring attributes

Nokia Loyalists

n=12 4 Returners

• Makes products that are easy to use

• Makes products that are easy to use

• Makes reliable products

• A brand you can trust

• A brand you can trust

• Makes reliable products

Nokia’s lowest-scoring attributes n=43 n=14 3 Defectors Other-Brands 0 • Has the latest features and functionality

• Stands out from other brands

• Stands out from other brands

• A socially and environmentally responsible company

• Stylish and sophisticated

Brand Preference

n=222

83 77

60

45 39

1 to 6

8

9

Handset Satisfaction (1-10 scale)

• Makes phones you are happy to be seen with Source: Synovate

7

Source: Bus Science

10

Example 6: Remington & data generated insights

Brand insight

Consumer insight

But there is a lot more that we need to get our heads around in this area

Reversing back to source

Did they do some groups?

Nope they used customer data

And did loads more stuff with it

So we got to the hypothesis of Polo Woman

Enter binns

Next steps • Q+As – Millward Brown – TNS

• The Binns Sessions – Data workshops

• Open to us to add………….. • We Love Data progress check