Mirek Cerny | Data Advisor
Mirek Cerny | Data Advisor
Anonymised client: a B2B automotive trading company operating in the Czech Republic with business activities across the EU.
The client had extensive customer and sales data, but the sales team did not use it systematically when deciding whom to call, when to call, and what exactly to offer. Most of the sales activity was carried out remotely through phone-based sales calls.
Before the change, the client generated approximately EUR 13 in average revenue per sales call.
The goal was not just to increase the number of calls. The key question was how to generate higher business value from each call.
The main questions were:
which customers should be called first,
when the right moment to contact them is,
how often each customer should be contacted,
what offer the salesperson should open during the call,
how to turn data analytics into practical recommendations for the daily work of the sales team.
The solution was based on connecting customer, sales and phone-call data.
The main data inputs included:
purchase history from the ERP system,
records of completed calls from the phone system,
cleaned phone numbers and contact details,
purchase frequency,
time since last purchase,
historical revenue,
time since last contact,
number of completed calls.
Based on this data, a recommendation model was created to prioritise customers by likely business value and help salespeople decide whom to call and what to offer.
The model segmented customers by business priority and recommended when and how often they should be contacted. It was not a static list, but a practical plan for the daily work of the sales team.
The most important factors included:
time since last purchase,
purchase frequency,
annual revenue,
time since last contact,
phone-contact history.
The predictive model recommended several specific items or vehicle types that the salesperson could offer to the customer.
It was based on historical customer purchasing behaviour and worked in a similar way to recommendation models used in e-commerce.
The recommendations were made available to salespeople through daily outputs and interactive dashboards. Selected information was also integrated into the ERP and CRM systems so it was available directly within the sales team’s normal workflow.
The solution ran on a dedicated data warehouse and used, among other components, Microsoft SQL Server, Power BI and reporting outputs for daily sales instructions.
After the recommendation model was introduced, the average revenue per sales call increased from approximately EUR 13 to approximately EUR 21.60.
This represents an increase of approximately 66% in revenue per B2B sales call.
The result was achieved even though the quality of contact data was not ideal and the sales team used only part of the recommendations in practice.
The solution helped the client:
prioritise sales contacts more effectively,
increase the commercial value of phone-based sales calls,
give salespeople practical recommendations for their daily work,
connect data from ERP, phone systems, CRM and reporting,
move analytics from dashboards into concrete sales actions,
gradually extend the recommendation logic into other parts of the sales process.
This case study shows that the commercial performance of phone-based teams can be improved significantly when data is used not only for reporting, but to guide everyday work in a practical way.
The core of the solution was not technology alone. The key was to connect customer, sales and operational data, identify useful patterns and translate them into concrete recommendations for the people running sales conversations.
The same principle can be applied in other areas where companies work with a large number of customer interactions — including sales teams, customer care, call centres and retention activities.
In similar situations, it makes sense to build the solution gradually in layers: from reliable data and clearly defined metrics, through analytical and predictive models, to recommendations that help people make better decisions in a specific operational context.
Send me a short description of what you are dealing with, what phase the project is in and what result you need. Together, we will choose the right form of support — from consultation through solution design to expert oversight or preparation of more advanced analytics.
Mirek Černý
Data Advisor
