How to predict people’s buying habits

How to predict people’s buying habits

This article revealing five steps to using customer data to predict people’s buying habits was published in Technative and in print in Ad Map.

The key to converting potential customers to actual customers requires ensuring your marketing messages reach them when they are ready to buy your product or service. Sirdata’s Harvey Sarjant reveals the route to using your customer data more intelligently to reach your audience in the right place at the right time.

The secret to successful marketing is mapping out the customer journey to identify the point at which people or businesses are ready to buy and then making sure you reach them at this stage. Whilst organisations are getting better at targeting the right audience, few are actually hitting them at the right moment, which wastes valuable marketing time and budget.

Here are five ways to hone your campaigns on purchase intent through using data more intelligently:

1. Define a clear goal

Every campaign you run should have a goal from the outset, but the clearer and more specific this is the easier it will be to define the key data you need for effective targeting. It will also provide you with something more concrete to measure so that you can gauge the success of the campaign more easily.

A goal such as “Grow retail purchases”, for example, may seem sufficient, and it is fine as a general aim. However, it is too broad for a specific campaign. Look at your business plan for the coming months and pull out a key objective to focus your campaign around, such as “identify which customers will buy a barbecue within the next 21 days with 80% accuracy”. This gives a sharper edge to your work.

2. Think like your customers

Rather than thinking of your audience as homogenous groups, which segmenting does, see them for who they really are – real people. Put yourself in their shoes and consider the factors that will motivate them to by your product or service. Think about what mindset they need to be in and how they came to this point, then map out what their journey to buy might look like.

This will give you a better idea of the kind of data you will need. Purchase history will give you an idea of who has bought what and when, but to define intent you’ll need to draw on information that gives you an idea of how they are currently behaving and thinking, such as social data.

3. Collect the necessary data

With your goal clearly defined, it’s time to identify the data that you’ll need to make your campaign a success. Start with purchase history, and then make sure you collect relevant data from across all your communication channels: search, social and programmatic.

Carefully analyse the journeys customers have taken to purchase, identifying key signals of intent to find out if a pattern or trend emerges. Try to track down the trigger that set them off on the path to making a purchase. Also find out if the pattern changes for different products and in different jurisdictions, or due to any other factors.

Depending on the skills you have in house, consider partnering with a reputable data modelling agency. As well as helping with the data collection and analysis process, they will also be able to advise on how much data you’ll need as this could vary from two years’ worth down to a single month.

4. Build predictive data models

Work with your data analytics and data modelling partner to apply intelligent techniques to build a more complete picture of the customer journey so that you’re better equipped to gauge intent. Building predictive models is an iterative process, so start with a simple one first, as this will be faster to turn around, and you’ll be able to gauge the results more quickly.

By first identifying the right mindset and intent signals, then establishing the right place and right advert, intelligent data modelling improves the performance of your campaigns, delivering a clearer understanding of what your audience is going to buy and helping you identify which product and price point to communicate.

Gaining a greater understanding of audience intent in this way will also enable you to use advertising to guide people along their buying journey by giving them the information they need when they need it.

5. Test your model

The final stage of the process is arguably the most important as you don’t want to compromise the integrity of your data by running your model without first testing it. Carry out the testing of your predictions on a sample dataset that has not been used in developing the model to see how it performs before applying it to your live data.

This should give you a reliable assessment of how your model will work in a real world setting. If all goes well, you will be ready to run your model and identify an audience that meets your original objectives with real intent to purchase.

Summing up…

Predicting the purchasing intent of your audience through intelligent data modelling enables you to deliver advertising and marketing campaigns that are more relevant and interesting. This in turn drives engagement, response and conversion, increasing revenue, and helps you get more from your marketing budget. What’s more, it improves the customer experience as you are only reaching people when they actually need you.

Harvey Sarjant is Managing Director at Sirdata

(This article was co-written by Three-sixty)