Ever feel like you’re talking to a brick wall? Is the ‘wall’ to blame for not listening? Or could it be that you might be missing the mark? In fact, hitting a brick wall with your message is still a common weak point in B2B marketing. Trying to get through to places like “Restaurant Olive” or “Hotel Rainbow”, while they actually aren’t even reading your messages. They are essentially brick walls.
What are customer personas?
It is important to be aware that in B2B, it’s still about people doing business with people. We create customer personas to help you do just that. A customer persona is a person who symbolises a segment of your (potential) customers. We wrote a previous blog on creating and applying customer personas.
Developing a customer persona enables you to pull together all the knowledge already present in your company, in the marketing department and with the sales representatives, and using it create a fictional persona. Engaging with this ‘person’ helps you send out messages that are proven to be more effective. That means you can develop a better tone of voice, improve your word choice and make more powerful proposals. It’s actually a very familiar concept – in person you speak very differently to an 80-year-old woman than you do to a 5-year-old.
The dangers of qualitative data
When you sit down with everyone together to explore personas like this, to fine-tune them or to jump to the challenge, you are often talking about qualitative data. About feelings, ambitions, issues. That’s the kind of thing you hear from your customers. And that feeling also plays a major role in their decision-making process, provided you are able to properly assess it among your (prospective) customers. But how well can you measure that feeling? After all, do customers really lay all their cards on the table? Are they actually trying out a competitor’s product, or are they just saying that to get a discount? Are they just pretending to be a very sustainable business while they would never actually consider spending ten cents more on your sustainable packaging? And so there you are, with your customer focus and your warehouse full of products.
It’s also common to jump to the conclusion that the views of one specific customer – one you know very well – is a ‘reflection’ of the entire group. People can be too quick to generalise the insights they gain from just one person. Because you don’t hear or find out too much from that little guy in the corner who has been quietly using your products for years.
That’s why you should also have some quantitative data to help you fine-tune your customer personas. Because if it’s set up properly, it will give you a more objective representation of your customer base.
Fine-tuning your customer personas using quantitative data
There are several different models, all based on data that identify various types of customers. Personas like the ones described above are great for discovering new customers or targeting customers you don’t know that much about yet. But what if you had the luxury of getting some great sales data from wholesalers, or if you can keep a good record of your customer contacts yourself? You know when and how often you’ve gone the extra mile for customers. Or when they’ve been disappointed.
That data can be used to break down your customer personas in very differently ways. And you can use this data to give your personas a valuable update.
While it may be a kind of free for all, your website and social media can provide you with plenty of information about the persona you are dealing with. Is anyone following your page? Do they visit your page on a regular basis? Do they also sometimes interact on your social media posts? People don’t just visit your page multiple times for no reason. They don’t just view an above or below average number of pages just for the sake of it. It says something about how relevant you are to them, and that level of relevance is also extremely useful to take into account when completing a persona template. And even more obvious: Based on actual data, you know the channels you are reaching them on.
Based on interest in products or product groups, you can predict which other products are also of interest to a customer. Netflix and YouTube are well known for doing that. And players such as Amazon and Bol.com are also incredibly good at this, because they’ve put a lot of effort into making these models. But actually, it’s very simple to start.
Let’s say you’re a food manufacturer who has just developed a new flavour or product application. It’s best to recommend that product to customers who have already tried an earlier version, or who are still using it. “Hi there. You’re using a variety of oriental sauces already, this soy sauce is definitely for you.” That works brilliantly when you also make a relevant offer. But you will lose credibility if you offer this product to a Thai restaurant that makes its own sauces. For these customers, it would almost feel like an insult for them to be associated with a factory product.
You can use churn models to explore how likely you are to lose customers. There will always be a small percentage of customers you won’t be able to keep, for instance because they’ve had to shut down their business. We call that involuntary churn. But most customers who leave their suppliers do so voluntarily. In other words, they choose a competitor because they have a better offer, or because they’ve had an unpleasant experience with you. Churn models are completed using historical data. You look at why customers left you and what differentiates them from those who stayed. It could be that when you don’t have any sales representatives out in a particular area, the ‘churn’ is bigger. Or a complaints procedure could be the link between customers you recently lost.
You can use historical data to determine which criteria have a higher correlation with customers staying or leaving, and then put those aspects to good use in evaluating your personas. The image below shows some variables that could be interesting with this kind of model, but it is important that the variables that matter to you come from your own data.
How should you approach a customer with a low churn risk compared to a customer who is more likely to leave you? It all depends on the customer type. If you are almost certain that you can keep a customer, then you can safely try a cross- or upselling offer. But if you have a customer that is threatening to leave, then you might want to offer a discount on their regular order.
RFM model based on sales data
The RFM model categorises your customers according to three variables: Recency of the last purchase, Frequency of purchases, average Monetary value of each purchase. In this article, we will focus only on the first two as it makes sense that you will always go that extra mile for a customer that brings in more revenue.
In the diagram below, you can see how you can use sales data to identify 4 different types of customer. This is a very simplified illustration of the model. In practice, there are also many intermediate stages.
You can use sales data to identify your top customers, i.e. champions. They place frequent orders and made their last purchase recently. You want to treat them differently to sleepy customers that placed an order with you once sometime around the beginning of the year.
A champion needs to feel like a champion. They need to receive preferential treatment and get the feeling that they have really earned your appreciation. And if a customer – no matter whether they are a private individual, restaurant owner or catering company – threatens to leave, you need to find out why they have stopped placing orders with you and respond to this in a personal manner. A general persona-based approach will not be enough to keep all these customers.
An approach based on customer persona and customer data
Summary: customer personas are created using a mix of research, knowledge and experience. You can define customer personas most sharply by including CRM-data, sales data and quantitative research too.
Use the GROUP7 persona template to establish persona profiles. We use them in our own persona workshops too. These help you to create context for your persona so that you can address your customers in a more direct and personal manner. You can use our template to map the goals, challenges, expectations and desires of your target group and using some examples from your customer base at your own discretion.
Your communication with customers will become much more effective by responding – according to the data – to actual customer behaviour:
– What can I infer from behaviour on my website? (web analytics)
– Where is this (potential) customer positioned in the customer lifecycle? (RFM model)
– Which product is the most promising? (prediction modelling)
– Am I at risk of losing this customer? (churn model).
Combining personas with customer data will make your communication rock solid. Imagine we have a persona that we’ll call ‘William’:
You can identify a William when someone…
– regularly visits your website,
– is a satisfied customer, and
– has not yet placed a lot of orders with you.
On the other hand, there is also a William who…
– does not visit your website,
– is at risk of switching to another supplier soon, and
– is a ‘champion’ when we look at their purchasing data.
We decide that a campaign will be relevant to the ‘Williams’ by looking at their goals and challenges. The additional data determines, for example, how exactly the campaign will be carried out.
In the example above, the second group of Williams can be reached by phone – in addition to the digital campaign – in order to be able to keep this important customer.
What can you do now?
Don’t have any customer personas yet? Downloading our persona template is the best place to start. Use this aid to create razor-sharp customer profiles.
Do you want to find out more about how data is used? Or do you want to work with the customer data available to you and the models named here? Get in touch at firstname.lastname@example.org.