- About Us
- Customer Cases
Advertisements are often tailored to individuals or groups of similar users, but they don't always focus on identifying the value the customers bring to the business. Customer lifetime value will help you determine the behaviours and characteristics of your most valuable customers and understand how they interact with your brand. By consolidating data and applying machine learning you can discover what value each customer brings to the business and run data-driven campaigns on your own and paid channels to increase revenue. You can identify the most profitable customers not just historically, but in the future as well by predicting the customer lifetime value (pCLTV).
Optimise your ad spend by targeting people that bring the highest value to your business. To be able to determine who those customers are, it is crucial to have accurate data that are being delivered real-time into systems that can activate them. Combine information from web traffic, purchases, email clicks and other sources to reach the audiences that really bring value to you.
Data from Google Analytics and real-time transaction data will be used as the basis to score customers immediately after each event they create. Based on session-data from Google Analytics and real-time transaction data the pCLTV is calculated for a customer using a predictive model immediately after each relevant event they trigger. This pCLTV score is pushed to Google Analytics (via the measurement protocol) and used to determine if the user is added to an audience.
Based on pCLTV you can set up audiences and target them effectively via marketing campaigns in Google Ads, Google Marketing Platform or social media. Use the information also to create similar audiences and improve reach in those channels or use predicted CLTV to optimise the audiences that you want to target in your campaign
Similarly to creating audiences based on high CLTV in Google Analytics, you can optimise bidding for paid media using a target return on ad spend (TROAS) based on pCTLV. In order to reliably predict CLTV, detailed transaction information like product-margin, discount and returns, have to be used.
Move customers along the stages of the customer journey, using predictions about the likelihood to engage or purchase propensity synced in near real-time into marketing audiences. Focusing marketing activities on customers who are most likely to do the desired behaviour, based on their most recent state, will result in more relevant customer interactions and greater ROI for marketing campaigns.
Identify and prioritise the most attractive visitors to your website and invest more in the effort to nurturing them.
Optimise visitor engagement and communicate with visitors in a meaningful way. Stimulate more return visits, more page views and more time spent on the website. For the modelling of engagement, we use the Google Cloud Platform, BigQuery, clickstream data tracked by Google Analytics, to prepare, train and test our models and deploy readily on GMP. Used for effective retargeting across marketing channels as well as the development of similar audiences for more intelligent customer acquisition.
Understand which customers are likely to purchase in the next, say 15 days, using the likelihood to convert the propensity model. Similar to the Visitor Engagement model, it is built on Google Cloud Platform, BigQuery using clickstream data tracked by Google Analytics across session, purchase, discount, location feature types. Target high probable customers via audiences for moving the customer along the customer journey.
We specialise in building headless data CDPs where systems are loosely connected through APIs. Our data CDP can integrate all the different data sources like Salesforce Commerce Cloud, Marketing Cloud, Service Cloud, Selligent, Point of Sale Cowhills, Exact, Google Marketing Platform products, Tableau and any other sources you might think of. The possibilities are endless and this is what makes our solution unique.