Improving e-mail conversion with segmentation: questions to answer
The recent Conversion Report from Econsultancy mined the data from more than 700 respondents which included both client-side (in-house) organizations who want to improve their conversion rates, and agencies, vendors or consultancies (supply-side) who are involved in trying to improve conversion rates for their clients.
For this post let’s focus on segmentation. Now, here’s something which we’ve known for years enhances relevancy and ultimately conversion rates and yet still we see this technique being either ignored completely or not employed to its full potential.
From the report:
Demographic (39%), geographic (36%) and behavioural (33%) data are used most often as a way of segmenting visitors and customers, according to our client-side survey.
A higher proportion of supply-side respondents (54%) say their clients use demographic data than indicated by the companies themselves. Similarly, just half of agencies (46%) say their clients segment by geography, compared to 36% of companies.
Segmenting using preferences, interests and hobbies is more widely seen by agencies than it is by client-side respondents (26% compared to 20%).
Some 13% of company respondents say they do not do any segmentation, and 15% of agencies say their clients are not segmenting their visitors and customers. There are clear opportunities for companies to make better use of advanced targeting techniques, and for agencies to help their clients explore the different techniques used for segmenting customers.
It goes on to indentify a number of segmentation options and the percentage of those companies vs. agencies using these techniques.
Some of the more familiar segmentation types would be:
- Demographic
- Geographic
- Behavioral
There are certainly more ways to segment your data, such as:
- Customer Engagement
- Preferences, interests and hobbies
- Transactional/RFM (Rececency/Frequency/Monatery Spend)
- Propensity models (Churn/Cross-Sell/Future-Sell)
- Media Interaction
- Psychographic/Attitude/Satisfaction
How many segments do you look at?
With both companies and agencies just over half look at two to five segments. Not surprisingly the report also found that, on average, organizations whose online conversion had improved over the previous 12 months looked at twice the number of segments as those organizations whose online conversion had not improved. More targeted means more relevant and that should mean better results all around.
There are really 3 important questions to ask:
Are you segmenting?
This is a key factor for success especially in email marketing! Personalization is the tip of the iceberg here. Segmenting allows you to be more targeted with your message and more relevant! More relevance means a better engaged subscriber base and that means more potential for conversions and fewer unsubscribes.
Are you segmenting correctly?
Are you focusing on segmenting based on data which will increase conversion rates? Define what you really need to know about your customers to enhance their experience and your ROI.
Are you looking at enough segments?
The report does seem to draw a correlation between conversion rates and the number of segments tracked.
Gathering segmentation data
Finally, how do you gather this data? The first step is the subscription/sign up form. Never ask a person for more information than you will absolutely use right away to the benefit of the subscriber. According to some recent surveys, for every additional field of data added, you lose 6% of sign ups. Keep your sign up process under 45 seconds.
Remember, you can always use your newsletter and landing pages/preference centres to ask for more data from your subscribers/customers as your relationship goes forward. Don’t ask someone to marry you when you are only on your first date. And don't ask more than you really need to know...
Finally this: the report and this post depart from the email perspective. Try to look at personalization, segmentation and data from the customer perspective, which is by definition cross-channel...
- Category:


Recent comments
1 year 49 weeks ago
1 year 49 weeks ago