We use data mining and predictive modelling techniques to target customers with better communications and make marketing more profitable.
A Small change can yield a high reward

Understanding your customer behaviour is essential for business success. It is more rewarding to segment your customers and forecast their requirements and interests based on the data you hold. Increasing customer satisfaction and CRM response rate.

We can use a host of predictive modelling techniques to identify patterns across data sources. Merging digital, call centre and transactional data to get an inclusive view of customer interactions and preferences.

Our analysts look beyond the obvious to allow you to create a joined up strategy, where customers are given the best experience at the right time. Making your marketing more profitable and allowing you to fine turn services across the board.

Customer Analytics

It is cost and time effective to recognise your most valuable, most engaged, loyal and highest spending customers and where to focus your marketing efforts. Also deciding on most likely prospects to target.

Identify naturally occurring clusters of similar users according to their behaviours and characteristics using granular behaviour, usage and demographic data.

We can then develop models based around these segments which can give you a clearer picture of where your revenue is coming from and where short falls may be and in turn  inform communications strategies.


Customer churn or attrition can lose your organisation revenue for two straight forward reasons. Firstly the loss incurred when that customer ceases to trade or subscribe with the business and the marketing budget spent in order to acquire them. Next the subsequent budget required to replace them.

Therefore identifying through predictive analysis whether a customer is likely to leave is brilliant for saving time, money and resources. Retaining an existing customer is less expensive than the spend required to attract a new one.

It is also better to develop accurately targeted retention offers or discounts for the identified segment. It is simply a waste to give a discount to a larger portion of your database than necessary or those customers who are a low flight risk.

If they have already left and are inactive is it worth winning them back? Or are you better off cutting your losses with that customer. If it is worthwhile pursuing them, you can determine what that customer is likely to need in order to return, their value to your business and make a sound judgement.

Using predictive propensity models we can segment, score and determine those most likely buyers. Using real-time and historic data we can predict those customers most likely to be susceptible to upgrading and the best time to make them that offer.

By looking at a customers purchase history, interests and demographic and other qualities we can segment those customers who would be interested in a deal by a similar brand or offer one that fits with their lifestyle.

Next- sell is similar, a segment who bought a particular item might be very likely to want the next item in that life cycle. Using a bicycle shop as an example. If a customer bought a tricycle for a three year old (knowing that they are a parent) in a years time you might target them with a “first two wheeler” bicycle offer.

Marketing Analytics

Attribution modelling is perhaps the holy grail of analytics. Which digital media channels are preforming best and driving sales?

Lynchpin look at “big data” at a granular level to understand the user journeys of your customers and where conversions are coming from.

Depending on your industry, service and product, a customer’s decision might be relatively fast and conversion funnel simple or it may be complex and winding, with many more touch points to take into account. PPC, Search, email, banner advertising and social media could all make a contribution.

Lynchpin can give your an accurate assessment of channel impact giving you the control to adjust where you spend your budget and optomise performance.

Econometrics Modelling looks at the impact of offline media channels including, still highly influential and important high reach media such as TV, radio and display. Also taking into account the external influences of seasonality and competitor activity on brands.

Looking at data from a longer time frame than attribution modelling Lynchpin can apply a predictive sales model  to measure the correlation of media weighted against sales.

We can  measure how successful campaigns have been and forecast future performance using “What If” scenario planning to set optimal spend levels.

Combined use of econometrics and attribution can give you valuable insight to inform your entire marketing strategy and ROI across channels.

Increase campaign response rates by using statistical user data to predict those users or subscribers most likely to respond positively to a particular campaign.

Use historical data to see how sub-segments have previously responded.  Decide which customers are best to target with a particular product or offer. Understand which marketing messages are working and the best time to serve various user groups.

These models can be continually refined in order to grow loyalty and provide the best customer experience. With large data sets even a small increase in positive reponse can yield high monetary gain.