Grow your business with data-driven segmentation
Many companies find it challenging to segment their data and have quality, relevant data to use and gain insight quickly. Gartner’s research says that 63% of digital marketing leaders continue to struggle with “delivering personalized experiences”, while only 17% are using artificial intelligence and machine learning to align with their customer acquisition and retention goals.
To offer solutions, our last article on segmentation covered how companies can understand customers better, make business processes more efficient, and what caveats to consider when operationalizing their segmentation.
In this article, Alex Schouleur, Data Scientist, breaks down some segmentation best practices and how you can take your segmentation to the next level.
“Segmentation leads to smarter decisions. It ensures you’re taking the best commercial action for each client which will increase their value and loyalty. It also helps you focus your efforts where it matters.”
Segmentation and why it matters
Why do we need to focus on segmentation?
We’ve seen too often, and maybe have been on the receiving end of, a marketing campaign with no personalization, except, perhaps, your first name. That lack of granularity can even be seen at the core of businesses’ strategies. They only have one big undifferentiated mass of ‘customers.’ This is the old way of doing things.
Business leaders are looking for a better approach. First, it begins with the client’s needs. Then, depending on the needs you identified, sending the appropriate message. So, identifying particular needs and preferences for specific groups can derive better results, benefitting every part of your company.
Segmentation leads to smarter decisions. It ensures you’re taking the best commercial action for each client which will increase their value and loyalty. It also helps you focus your efforts where it matters.
In what ways has client behavior evolved in the B2B space?
Expectations of B2B customers have certainly changed. Business managers expect messages to be personalized and attentive while knowing how to recognize a mass-automated message. Marketing also needs to adapt to the increasing amount of information on businesses available online and target accordingly. Business clients expect you to search for it. If you don’t, it gives a poor impression, while wasting your time and money with bad targeting.
Back to basics
Okay, so how can companies begin to understand their customer segments?
There are two main ways. The first is the more traditional way, often where marketing departments split their contact list according to specific criteria, or run wide-scale customer surveys and have other direct customer interactions. But now, companies need more granularity and feedback across departments and sources for improved personalization and customer intimacy.
As a second alternative, rigorous quantitative techniques allow you to go further and to have a better understanding of your customer. Machine learning algorithms can group companies or contacts on a mathematical level where the algorithm recognizes them as very similar to each other, forming a cluster.
At Agilytic, what we do is combine those two methods to apply segmentation capabilities based on advanced analytics. We begin with a technical challenge, then give meaning to the findings with business intuition while taking feedback into account. Segmentation is an iterative process. We may run the algorithm once, then try to understand it from a business perspective and perfect it to ensure it’s useful.
How can firms start with gathering internal information for segmentation? What are the standard methods you’ve used in projects?
Teams need to know their customer segments and critical indicators and dimensions. In B2B environments, we investigate company-specific “firmographics” to see what business-specific factors matter more.
These firmographics can be organized into a database, CRM, or even a simple Excel file to include company name, location, business size, employee number, turnover rate, the language used, and much more. Additionally, you can tag the industry type, billing (e.g., recency, frequency, average basket/purchase size), customer interactions, and behaviors (e.g., late on purchase). With this groundwork, you can already have a good segmentation.
Simple, straightforward methodologies to gathering internal data can generate a lot of quick wins already. But whatever approach we take, it needs to be actionable.
Let’s take it to the next level
How can firms start with gathering external information for segmentation? What are the standard methods you’ve used in projects?
Externally, companies often tend to limit their options without realizing the possibilities. External sources can be neglected but bring a lot of value to the table. Usually, it’s not a training problem but an awareness problem.
We can guide teams on which data we recommend obtaining and how to get that data. External data useful for B2B teams are turnover, profit, number of employees, balance sheet size, profitability, and other financial information. Additionally, geodata can make links between coordinates, regions close to eachother, interest points, highlight average revenue or income in certain areas.
Using accessible, open, online sources to feed the segmentation and qualify leads ensures you’re spending valuable time on the rights prospects and that you can send them the right message. That’s where web scraping comes in.
There are two main ways for web scraping; the first is to develop code with Python or use dedicated paid web scraping tools. For us, we keep things simple and try to deliver what has the most value. It doesn’t make sense to reinvent the wheel. Python usually works best for what we need.
Can you explain why a segmentation project may be a good ‘springboard’ to other projects that serve a business’s strategy?
The purpose of segmentation is to derive actions and help your business grow. Depending on your segments, you should have a clear idea of how to adapt the delivery and approach of your communications.
Data segmentation is a great place to start as it enables a lot of other projects. It can help you forecast behavior, build a recommendation engine, signal customer churn, and much more. Segmentation will also help you structure and develop your business strategy by providing a common language to describe your customers' groups.
What you do with your newly-found segmentation knowledge is essential. After a segmentation project, we take the time to describe how to use the segmentation can deliver ongoing value.
