Subscribe to our mailing list

* indicates required
Close

Market research, survey analysis and productivity.Subscribe Now

Speak Your Customer’s Language: How You Can Use Segmentation to Create Effective Customer Profiling

DataCracker - Use segmentation to create effective customer profilingYour customers aren’t all the same. Segmenting customer lists and creating customer profiles continues to be a top priority for marketers. By tailoring messages to different groups of customers you could be opening the door to higher conversion rates, customer engagement and sales. If you feel like your business isn’t doing enough segmentation of your customer base then read on to find out how you can start creating your own useful customer profiles.

Why bother with segmentation?
No matter what product or service you sell chances are you don’t have a “one message fits all” approach that will resonate with every single one of your customers. The solution is to spend some time understanding the key similarities amongst sub-groups of your customer base or email list.

Let’s take the simple example of segmenting by customer’s content consumption interests. Say you’re a yoga instructor that sells yoga training courses and related yoga products. You can start to collect data on the preferences of how your list prefers to consume information. Then get busy crafting a tailored message for each of the segments you identify. You might learn that 30% of your list prefer reading educational blog posts but 70% prefer more visual-based learning via video. By segmenting your list you now know that it could be a useful marketing strategy to spend more time on creating visually-aided content. Creating a series of videos or webinars where you can visually promote your yoga courses and products could deliver more sales.

Segmenting: Your sales and marketing decisions can be based on multiple types of segments. For example the types of questions you might look to research might include:

1. Demographic data
Location. Age. Male or female? What income bracket? Married or single? Children or no children?
2. Buying preferences
How do your customers prefer to transact? Online, Over the phone or in person?
3. Where do they hang out online
Where did they hear about you? Do they prefer emails or engaging on social media? Or both?
4. Identifying problems
What are their pain points and frustrations in relation to your industry?
5. Identifying attitudes and behaviours
How much time do they spend online? What other interests do they have?

The collection of the above types of data has the ability to show you that all of your customers or potential customers have slightly different wants and needs. Your job is to take this data and to look for the insights that will identify these wants and needs. The segmentation tools in DataCracker will help you unearth these insights. You can then have some fun with your insights by creating customer profiles.

Some basic segmentation of the above data leads us to create two profiles for our earlier yoga business example:

  • Martin “The Email Guy”: Martin is representative of the segment of customers that strongly identify with a weekly email newsletter that keeps them up to date with tips and new product offerings.
  • Mia “The Video Lover”: Mia hates emails. She loves spending time on YouTube and is happy to invest an hour in a webinar if it will teach her some new yoga routines.

Reaching out: Now you need to decide on the best way to reach your segments and engage with them. Some questions that will help you achieve this include:

  • What is the best way to communicate with this customer?
  • What key terms and phrases will resonate with this customer?
  • What might be some potential barriers to them buying? How can we address these?

You’re Not Alone
In case you’re wondering whether this type of segmentation and profiling is just a flash in the pan we’ve got some stats to highlight it’s importance. A recent Econsultancy/Adestra Email Marketing Census report based on a survey of 1,329 agency and client-side respondents showed that:

  • 73% carry out segmentation
  • 46% plan on implementing more advanced segmentation
  • 38% plan on implementing behavioural targeting

Use Segmentation to Create Effective Customer Profiling

Segmentation won’t come without challenges. But by taking advantage of good survey and analysis tools you can begin to collect data from your list and unearth some valuable insights. Start with small experiments. Segment your customers into some basic profiles to begin with and measure the success of your open-rates, engagements and revenues. Can you afford not to?

Image courtesy of jscreationzs at FreeDigitalPhotos.net

About Luke Ryan

Luke Ryan does marketing for DataCracker. He writes on the blog and works on community and coverage. He loves to answer your questions about all things DataCracker.

Comments

  1. Hi Luke. Good article. Identifying attitudes and behaviors is not as easy as it sounds as people are normally not aware of their own attitudes and behaviors. Researchers need to ask quite a few questions to identify the real motives and feelings behind. I’m interested in knowing more how DataCracker handles the data complexity and segments customers into groups. Thank you.

  2. Segmentation is a generic term. It’s normal usage is the simple application of filters based on existing categories. For example, to filter your data to only look at male or female data. DataCracker supports the simple concept of segmentation via creation of filters, and also cross-tabulation (looking at “What’s your favourite soft drink” broken down by “Gender”). This is called a priori segmentation.

    Sometimes there is no 1-to-1 mapping between survey questions and segments. This is where a more advanced statistical technique called Latent Class Analysis (LCA) actually finds un-named/un-categorised segments (or “classes”) in your data, based on similar answers to a survey question. DataCracker exposes this advanced technique in a very simple way, automatically discovering segments. For example, it could find 3 segments of people with strongly varying answers to “Describe your top 10 factors in deciding to purchase at Amazon”. Once you’ve found these un-named segments, you can then try and discover who these people are by comparing their answers to other survey questions.

    There’s more information on the different ways to form segments here: http://surveyanalysis.org/wiki/Approaches_to_Forming_Segments

    DataCracker is the only software I know of that exposes latent class analysis in a way that a non-expert can use.