Marketing Response Understanding the complete marketing ROI for direct seguros
Google Confidential and Proprietary
1
Background Direct Seguros as part of the AXA, Europe’s largest insurer, provides a range of insurance solutions for the Spanish market. It also has an extensive network of direct insurance businesses in France, Italy, Poland and Portugal.
Established in 1996 as an insurance company specializing in car insurance, it is now one of the leading insurance companies in Spain, marketing its products directly to consumers (no aggregators/brokers etc). Direct Seguros has adopted the web channel completely and now accepts applications via both online and its call centre.
In 2007, the company extended its insurance offer to include motorbike insurance among its core offering. Google Confidential and Proprietary
2
Objectives and Methodology Objectives
• Identify the key marketing drivers that generate new applications
• Quantify the effectiveness and efficiency of these drivers • This applies to both applications made via the call center as well as those coming from the Internet. • Understand the role played by website traffic along with the media (offline and online) in generating new applications.
Methodology
• Using Direct Seguros’ internal and proprietary data (October
2007- June 2010)
• Application data (split by call center, Internet) • Price/rate • Website data and analytics • Direct marketing and TV advertising investment
and data from other sources e.g.: • Google search volume • Economic indices
Various econometric models were built to understand the return on investment of communication media and marketing. Google Confidential and Proprietary
3
3
13/10/2007 03/11/2007 24/11/2007 15/12/2007 05/01/2008 26/01/2008 16/02/2008 08/03/2008 29/03/2008 19/04/2008 10/05/2008 31/05/2008 21/06/2008 12/07/2008 02/08/2008 23/08/2008 13/09/2008 04/10/2008 25/10/2008 15/11/2008 06/12/2008 27/12/2008 17/01/2009 07/02/2009 28/02/2009 21/03/2009 11/04/2009 02/05/2009 23/05/2009 13/06/2009 04/07/2009 25/07/2009 15/08/2009 05/09/2009 26/09/2009 17/10/2009 07/11/2009 28/11/2009 19/12/2009 09/01/2010 30/01/2010 20/02/2010 13/03/2010 03/04/2010 24/04/2010 15/05/2010 05/06/2010 26/06/2010
Background Direct Seguros accepts applications both online and offline via its call centers.
64% of Direct Seguros applications are made offline via the call center, while 36% are made online.
Call Center Internet
36% Call Center Internet
64%
Google Confidential and Proprietary
4
Call Centre What are the drivers of offline call centre applications?
Google Confidential and Proprietary
5
Drivers of offline Call center applications 7.4% of first call center contacts come as a result of visiting www.directseguros.es. And paid search increases these contacts by a further 1.4%.
8.8%
of the
contacts of call center come from online sources TV advertising 1.4% Call Center Contacts
Direct Marketing 0.6%
Web visits 7.4%
Price/rate 23.4%
Paid Search 1.4%
Holidays and seasonality 2.1%
Base level 63.6%
Google Confidential and Proprietary
19/06/2010
22/05/2010
24/04/2010
27/03/2010
27/02/2010
30/01/2010
02/01/2010
05/12/2009
07/11/2009
10/10/2009
12/09/2009
15/08/2009
18/07/2009
20/06/2009
23/05/2009
25/04/2009
28/03/2009
28/02/2009
31/01/2009
03/01/2009
06/12/2008
08/11/2008
11/10/2008
13/09/2008
16/08/2008
19/07/2008
21/06/2008
24/05/2008
26/04/2008
29/03/2008
01/03/2008
02/02/2008
05/01/2008
08/12/2007
10/11/2007
13/10/2007
Consumer demand, economic factors etc.
6
Drivers of offline Call center applications The website generates 7.4% of applications. Paid search contributes 1.4% of applications 89% of applications are not
7.4%
influenced by marketing 23%
64%
1.4%
via online
Only 11% of applications are influenced by marketing.
Seasonality
1.4%
8.8%
2%
The base level includes the effect of consumer demand, motor vehicle sales seasonality, the economic situation etc.
Base level
0.6%
Price/rate
Direct Marketing
TV Advertising
Of which 8.8% come from online
Paid Search
Web traffic
Google Confidential and Proprietary
7
Website applications What are the drivers of online applications?
Google Confidential and Proprietary
8
Drivers of online website applications The web unsurprisingly generates the majority of web applications
65% of applications are influenced by marketing.
7.1% 25.2%
35% of applications are not
28.2%
controlled by marketing
28.3%
1.5%
5.5%
3.2%
1.0%
61% driven by online Base level
Seasonality
External factors
Direct Mail
TV advertising
Organic Search
Paid Search
Direct web visit
Google Confidential and Proprietary
Marketing ROI Return on media spend
Google Confidential and Proprietary
10
Understanding the effectiveness of marketing spend Due to the natural variance in both spend and sales levels over the analysis period a response curve can be drawn to understand the response or ‘elasticity’ of each marketing driver.
4,800
Media X Weekly Sales (Units)
4,700 4,600 4,500 4,400 4,300 4,200 4,100 -
50,000.00
100,000.00
150,000.00
Weekly Media Spend €
200,000.00
250,000.00
300,000.00
350,000.00
Google Confidential and Proprietary
11
Understanding the effectiveness of marketing spend ROI of each media is dependent on the point at which investment is at on the curve
4,800
Media X Weekly Sales (Units)
4,700
Lower investment high ROI
4,600 4,500
Higher investment, same media, but lower ROI
4,400 4,300 4,200 4,100 -
50,000.00
100,000.00
150,000.00
Weekly Media Spend €
200,000.00
250,000.00
300,000.00
350,000.00
Google Confidential and Proprietary
12
Return on Marketing investment
The ROI of each media is based upon the media investment and return .
Paid search is the most cost-effective medium; very significantly more cost effective than direct marketing and TV advertising.
Online ROI (website) Offline ROI (call center)
26.76 55.4 17.90
0.47
1.62 1.83 Direct Mail
0.38
TV
Monetary value of an application and actual returns not shown due to client confidentiality – data indexed against total media average ROI
Branded paid search
Generic paid search
Google Confidential and Proprietary
13
Understanding Digital
Google Confidential and Proprietary
14
Drivers to www.directseguros.es Search is the primary driver of incremental web traffic. Also significant evidence of offline media driving web traffic
TV
5%
Direct Mail
5%
Paid
21%
Organic
40%
Affiliates
6%
Seasonality
3%
Base
22%
This is the modeled decomposition of incremental response (i.e sources of incremental web traffic) – not an observational referral source analysis .
Google Confidential and Proprietary
Effect of position in paid search Improving position in paid search, results in a noticeable increase in applications, both online as well as offline at the call center.
3rd position
Average paid position
Online application increase %
Call center applications increase %
Generic, auto and brand search terms
Direct Seguros average paid position
1.0
2.7%
1.7%
1.7
2.1%
1.3%
3.2
-
Google Confidential and Proprietary
16
Drivers of Branded queries Generic paid search impressions, contribute an incremental 3.2% to total branded query volume
Tv Advertising 2%
Generic paid search impressions 3%
Car registrations 11% Seasonality 2% Direct Marketing 1% Base online Queries 81%
Google Confidential and Proprietary
17
Drivers of incremental AdWords clicks Significant evidence of Offline to Online synergies.
6%
Ad position on page increases clicks by
9.6%
% Uplift in Paid search clicks
2%
Competitor Ads on the page increase clicks by
1.4% Google Confidential and Proprietary
Conclusions
Google Confidential and Proprietary
19
Summary and conclusions
1
The Online to Offline effect is sizeable. 8.8% of offline conversions occur due to online activity
2
Paid search is the most cost-effective way to increase insurance applications: over ten times more efficient than other media
3
Improving position in paid search increases applications online by 2.7%, and offline via the call center by 1.7%
4
Considerable TV and search synergy – TV incrementally drives 6% of paid search clicks
20
Post Research Media Spending
Google Confidential and Proprietary
21
Direct Seguros Media mix pre/post research Before research: January-June 2010 PRINT 0.1%
After research: July-December 2010
PAID SEARCH 8.9%
RADIO 3.0%
PRINT 0.2% RADIO 4.9%
MKD 7.0%
PAID SEARCH 17.0%
MKD 4.9%
TV 81.1% TV 73.0%
What they should really have done July-Dec 2010 Optimal media mix obtained from econometric models PAID SEARCH 20.3%
MKD 6.9% PRINT 0.9% RADIO 10.3%
TV 61.5%
Please note that the Optimal recommended mix does not take into account natural media constraints and therefore might not be possible to apply
MKD = Marketing Directo = Direct Marketing
Google Confidential and Proprietary
22
Testimonials
Olivier Lamarque – Marketing Director Direct Seguros Spain
Thanks to Google’s search data and our own advancements in qualified lead measurement, we are able to understand how all of our marketing channels are working. We understand the relationship between offline and online both in terms of our media efficiency but also our consumer and their preference to shop multichannel. We now know the true ROI of each of our media channels and have already optimised for performance.
Google Confidential and Proprietary
23
Appendix
Google Confidential and Proprietary
24
Modelo Ajustado 19/06/2010
22/05/2010
24/04/2010
27/03/2010
27/02/2010
30/01/2010
Primeros Contactos Call Center OT
13/10/2007 03/11/2007 24/11/2007 15/12/2007 05/01/2008 26/01/2008 16/02/2008 08/03/2008 29/03/2008 19/04/2008 10/05/2008 31/05/2008 21/06/2008 12/07/2008 02/08/2008 23/08/2008 13/09/2008 04/10/2008 25/10/2008 15/11/2008 06/12/2008 27/12/2008 17/01/2009 07/02/2009 28/02/2009 21/03/2009 11/04/2009 02/05/2009 23/05/2009 13/06/2009 04/07/2009 25/07/2009 15/08/2009 05/09/2009 26/09/2009 17/10/2009 07/11/2009 28/11/2009 19/12/2009 09/01/2010 30/01/2010 20/02/2010 13/03/2010 03/04/2010 24/04/2010 15/05/2010 05/06/2010 26/06/2010
visitas 02/01/2010
13/10/2007 03/11/2007 24/11/2007 15/12/2007 05/01/2008 26/01/2008 16/02/2008 08/03/2008 29/03/2008 19/04/2008 10/05/2008 31/05/2008 21/06/2008 12/07/2008 02/08/2008 23/08/2008 13/09/2008 04/10/2008 25/10/2008 15/11/2008 06/12/2008 27/12/2008 17/01/2009 07/02/2009 28/02/2009 21/03/2009 11/04/2009 02/05/2009 23/05/2009 13/06/2009 04/07/2009 25/07/2009 15/08/2009 05/09/2009 26/09/2009 17/10/2009 07/11/2009 28/11/2009 19/12/2009 09/01/2010 30/01/2010 20/02/2010 13/03/2010 03/04/2010 24/04/2010 15/05/2010 05/06/2010 26/06/2010
19/06/2010
22/05/2010
24/04/2010
27/03/2010
27/02/2010
30/01/2010
02/01/2010
05/12/2009
07/11/2009
10/10/2009
12/09/2009
15/08/2009
18/07/2009
20/06/2009
23/05/2009
25/04/2009
28/03/2009
28/02/2009
31/01/2009
03/01/2009
06/12/2008
08/11/2008
11/10/2008
13/09/2008
16/08/2008
19/07/2008
21/06/2008
24/05/2008
26/04/2008
29/03/2008
01/03/2008
02/02/2008
05/01/2008
08/12/2007
10/11/2007
30000
05/12/2009
07/11/2009
70000
10/10/2009
80000
12/09/2009
15/08/2009
18/07/2009
20/06/2009
23/05/2009
25/04/2009
28/03/2009
28/02/2009
Modelo Ajustado
31/01/2009
03/01/2009
06/12/2008
08/11/2008
11/10/2008
13/09/2008
16/08/2008
19/07/2008
21/06/2008
24/05/2008
26/04/2008
29/03/2008
01/03/2008
02/02/2008
05/01/2008
08/12/2007
10/11/2007
13/10/2007
13/10/2007
Models and settings
25000
94.2% 92%
20000
15000
10000
5000
0
Modelo Ajustado
90000
Modelo Ajustado
Primeros Contactos Internet OT
82%
93.4%
60000
50000
40000
30000
20000
10000
0
Google Confidential and Proprietary Clicks_Paid_search
25
Interconnections between Direct Seguros business drivers Caption
Online World Online search
Offline World
call center
+ Driver
TV advertising + Direct marketing
- Driver
SEM (Paid search)
Call center
Web traffic
Online (Contacts)
(Contacts)
Base level, price, seasonality registrations
Drop in car sales, crisis
