Story
Marketing
Machine Learning

Marketing Analytics: collecting out-of-the-box data

By Jordy de Jong | Invalid Date

So, you already collect basic information from your company's customers? Then you're on a roll! Now let’s move on to the next steps: from machine learning to real-time personalisation. Sounds difficult? It's not! Let's get started.

This article is a continuation of our blog "Where to start with Marketing Analytics" It explains exactly how to collect the basic information from your customers: what you can do yourself and what you can leave to the application.

Put your technical glasses on

Brace yourself, because from now on things will get a lot more technical. We delve deeper into all the useful tools, but this might require some prior knowledge. Do you need some extra help with that? No problem, you can always contact us.

Grab what you have and make it smarter

You already know what your customers do on your website and now you can do a few more things to collect extra data: teach computers how to learn, contact customers directly and / or apply personalisation. These are all things that are probably already (partly) in your current service. But now we make things smarter!

1. Machine learning (ML) Although the initial preparation work is considerable, you will seriously benefit from Machine learning. You can use it for:

  • Conversion Rate Optimisation (CRO): which customers have a high chance of converting and ML can then take the right actions automatically.
  • Customer Lifetime Value (CLV): how much are you likely to earn from a potential customer and automatically make investment in them by retargeting, for example.
  • Product Management: the Google Vision API looks at the colours of your products and can recognise similar products.
  • Sentiment: recognising the tone of a text is a challenge, but the Google Natural Language Processing API can handle it. It can recognise whether a review is positive or negative and immediately incorporates this into the collected data.
  • Social media: you want to know how your products are used but don't want to endlessly search hashtags and tweets about your stuff. That’s no longer necessary! Thanks to Google Vision and NLP APIs.

2. Customer satisfaction Happy wife, happy life: but with customers. When people contact you, talk about you and end up on your site, then you can collect the data.

  • Automated customer service: chatbots are useful, but Conversational AI are next level. Quickly resolve customer issues by having a computer respond the way a person would.
  • Find-ability: customers with a specific idea in mind want to be able to find it easily. A Product Search API ensures that similar products can be found through photos.

3. Real-time personalisation If you already have a lot of information about your customers, you might as well apply it to their visit immediately. Use this information so that the chance of conversion is even greater (CRO).

  • Data-driven customer segmentation: this creates personas based on the given data. These are then matched to customers and personalised for them.
  • Recommendation engine: the right recommendations for visitors ensures a longer time on the site and hopefully a conversion.
  • Advanced retargeting: address interested (new) customers with a high chance of conversion.
  • Price optimisation: what price do you show when and to whom? This can be precisely determined based on personalisation.

The finishing touches

If you take all these steps, you will end up with a lot of data. If the customer is still not happy and satisfied after that, then we're all out of ideas. Well, actually... you might want to ask yourself if you're going to make this technology yourself or buy it? We will discuss these advantages and disadvantages in the next blog: Marketing Analytics: making it yourself or buying it?

Jordy de Jong


E: jordy.dejong@incentro.comT: +31614184194
next story
Where to start with Marketing Analytics?
By Jordy de Jong | April 19, 2019