Case Study

Service: Business Intelligence

Industry: Services – quality assurance, testing, inspection, and certification

Department: Marketing

Size: 40,000+ employees in over 100 countries

Revenue: 2-3 bn GBP / 3-4 Bn EUR (2020)

Customer Funnel

Create a fully automated marketing funnel dashboard in six months

Questions about this topic? Ask our Specialist!

Ben Neesom

Ben Neesom - Sereviso CEO Contact

Project Briefing & Objective

As an international company with business in over 100 different countries and more than 40,000 employees, keeping track of all the marketing efforts is a mammoth task. Maintaining an overview of the results achieved in each location and drawing conclusions for implementing processes or changing strategies on a global level entails a huge number of resources.

Automating data by extracting from different sources into a single source of truth is both time- and resource-saving. Knowing the customer’s behaviour in each stage of the funnel and how it changes by location makes it possible to allocate a budget to the actions, which achieves more engagement and value for customers and therefore for the business. 

Thus, the marketing department aimed to create a customer funnel dashboard that provides an up-to-date overview of all marketing efforts. This helps understand which efforts attract more leads and where they must focus sales efforts to an even greater extent. Sales effectiveness in activities and the conversion rate can be ramped up as a result.

“The better we understand our customer, the better we can allocate our efforts to achieve the highest return on investment. The customer funnel dashboard provides a unique overview in this regard.”


Low levels of collaboration

Management at the headquarters of large multinational companies often faces the challenge of keeping up-to-date with the status of different subsidiaries around the world. That is why it is imperative to build a collaborative dashboard that allows authorised users within the organisation to directly access the data insights to change, understand and analyse the data.  

Low data accuracy

The biggest challenges arise due to the huge amount of data saved in different places, missing files, modifications not being tracked, and gathering data from at least five different data sources. The improper organisation of this inevitably leads to data that doesn’t match, thus creating distrust. Ultimately, this forces the company to work at a slower pace, both in decision-making as well as in the overall processes.

Non-standardised KPI definitions

The different marketing and sales departments around the globe have different key performance indicators (KPIs) that are relevant for each location. While this may work well for each location individually, the responsible management for these locations will have difficulties tracking the overall performance, as non-harmonised KPI definitions have to be consolidated first – only then do they reflect a true picture when combined. This additional consolidation step is a typical feature of non-harmonised systems and KPIs, creating a lot of manual work that could be avoided if global standards are used and adhered to.

Lack of correct insights for decision-making

Finally, partly caused by the challenges listed above, not having a global view of KPIs makes the decision-making process a lot more difficult. Also, companies risk making erroneous decisions that could have been avoided by taking a holistic approach. Thus, it’s crucial to be aware of the current global status and how the customer funnel is performing on each level – globally, nationally, and locally. 


Project Approach

The project was divided into three main phases, with a total duration of approx. one year and three months.

Phase 1: Assess, understand, and define

This phase marked the beginning of the project. Its primary goals were:

  • Assessment of existing data & data sources, reports, and KPIs
  • Agreement of common KPI definitions
  • Design and agreement on an overall system architecture

Phase 2: Piloting

After having defined what to look at and how to report on it, the buy-in of the end-users and leadership is key to reaching a highly satisfactory result for all parties involved. The pilot phase consisted of the following elements:

  • Design sprints with various teams
    This part is crucial to ensure requirements and expectations are properly captured. The early involvement of the different stakeholders also helps the adoption at a later stage, as people feel more involved, making it their own project to some extent.
  • Definition of an overall design and usability concept
    Having a guideline for the creation of reports and dashboards significantly helps to create an appealing and consistent look and feel, leading to recognisability and further supporting the buy-in from users.
  • Basic system & infrastructure set-up
  • Development of mock-ups to demonstrate the functionality and usability

Phase 3: Development & deployment

In this final phase, the system becomes fully automated, all data pipelines are established, the mock-ups are turned into a final product and the tool is rolled out to the community.


Benefits at a Glance

  • Daily reports in less than one hour
    previously: reports in one week
  • High data integrity through a single source of truth
    previously: 16+ different data sources for the same topic
  • Holistic and consistent reporting globally, with drilldowns to countries and more
    previously: individual country reporting, large effort to consolidate
  • Globally standardised KPIs and definitions
  • An interconnected system with structured processes
  • Trend analysis over the past twelve months
    previously: no comparison with prior months possible
  • Latest technologies and features, also allow unstructured data sources

More Case Studies