data scientist


This page provides a comprehensive narrative of my career to date.

It covers three distinct stages, starting with Education. The next stage is Project and Programme Management, followed by the last and current Management Consulting and Data Science.

Please click on the following links to jump to a relevant section:


  • Project & Programme Management describes project and programme management roles at Orange UK and Convergys.

  • Data Consultant describes various contract roles throughout the UK, where I increasingly specialised in data science and services as a core capability of my consultancy.


Early years & Education

I was born and raised in Cape Town.

I excelled at music and reached licentiate level by the time I graduated from school. During my year of compulsory military service, I was selected to an elite group in the South African Air Force (SAAF) and performed 255 concerts on tour throughout South Africa.

After completing national service, I remained in Pretoria and joined the SAAF Band. It paid very well for a 19-year-old and allowed enough time to focus on studies too.

In 1994 I started a degree in Marketing and Financial Management at the University of Pretoria and transferred to the SAAF Magazine, Ad Astra, as Marketing Coordinator. I spent my final year at the SAAF Gymnasium (Training Academy) as a Public Relations officer.

I resigned the SAAF upon completing formal education and returned to Cape Town for a few months before travelling to the UK.


BCom (Hons) Marketing & Financial Management 1994 - 1998

I studied at the University of Pretoria and majored in Marketing and Financial Management, completing an Honours degree in 1998. A South African Honours degree is an optional fourth and additional one-year, postgraduate qualification where intake is highly selective.

I completed two research projects as part of this degree, including:


Project & Programme Management

In March 1999 I emigrated to the UK, initially settling in London.

Temping 1999

I took on temporary work as an Accounts Assistant after arriving in the UK. I planned to complete training as a Management and Cost accountant; however, the lure of variety within the booming telecommunications industry proved too high.

Orange 1999 - 2006

I joined Orange Network Planning in London as a Project Coordinator.

Within a year I was promoted to Project Manager, a PMO role primarily focussed on providing management information in support of regional delivery. I discovered a love for reporting and fast became a Business Objects superuser, employing increasingly advanced MS Excel and Access with VBA to deal with more complex business problems.

It came naturally to me, and I developed an insatiable appetite for learning about data management and analysis.

A few highlights from this role include:

  • I designed and delivered a tracking database and spearheaded a focus group that improved a business process by following the six sigma approach. This reduced abortive and additional costs of over £5m nationally in the first year alone after implementation.

  • I developed a management reporting approach that improved the end-to-end delivery of the 3G programme of works. It optimised overall benefit by aligning delivery across multiple departments2, capitalising on earned value.
    It was rolled out nationally and used by leadership to gauge the status, progress, convergence and potential of the programme.

In 2002 I was promoted to Regional Facilitator (Programme Manager) in the North West, an area covering Northern England, Scotland and Ireland. I simultaneously commenced the Orange Senior Management Development programme.

I was responsible for regional governance and managed operational reviews with the leadership team.

Part of the North regional senior management team, I facilitated the delivery of several access network programmes across various regional and central functional departments.

It was in this role where I first started to programmatically sequence and model both concurrent and consecutive activities within workstreams, generating a programme schedule to harmonise and optimise delivery between multiple and independent constituent teams.

I won an award for leading the successful, cross-functional and end-to-end delivery of the Northern 3G launch programme of works.

This role, more than any other, honed my skills to influence and persuade and managing upwards.

In 2004 I took on responsibility for Facilitation (renamed as Programme Control) nationally, taking on direct reports and gaining line management experience.

I used modelling more widely to plan and forecast programmes and quantify risk. It kick-started an initiative to apply marketing segmentation to London boroughs in a bid to prioritise delivery. It served as a precursor to the Value to Cost Ratio, a weighted set of attributes resulting in a ranking score, used to determine budgetary spend on the Access Network to maximise benefit.

Convergys 2007 - 2010

I was responsible for managing multi-million-pound programmes developing, customising and implementing SS7 and SIGTRAN based IVR systems and Value Added Solutions (SMSC, Voicemail, MMS) throughout the EMEA/ APAC regions from inception to completion.

Some key points from this position:

  • Full budgetary responsibility of up to USD 2.5m
  • Customer facing, often at an executive level
  • Matrix management of teams of over 20 people internally and 3rd party suppliers, including remote teams.
  • Managing complete programme lifecycle, including design, software development, manufacturing, testing, installation, commissioning & network integration
  • Producing Statements of Work and Business Cases to support pre-sales bids
  • Coordinating implementation of solutions with resellers, including Ericsson and Huawei’s complex and extensive core network rollout and upgrades

Data Consultant

The next section covers the contracts that I took on as an independent data consultant.


Some general observations about the role:

  • Visualisation
    I initially focussed on generating management information only but soon realised the importance of visualisation, and have focused on improving that skill since.

    During this time I invested in Tableau Desktop as it produces standardised, high-quality and straightforward visualisations with little effort. Coding up a visualisation (using R or Python) is flexible and highly customisable, but it takes a lot more effort and thought all things being equal.

    I found it very useful in consultations, specifically during collaborative munging and visualisation as it is often challenging to keep typically non-technical (data) participants engaged during live coding sessions.

    Geospatial computation and visualisation is an invaluable skill and deliverable, especially given the nature of telco programmes. I perform many calculations and mapping programmatically using R3 (e.g. libraries like rgdal and spdplyr) and use the excellent GQIS for on-the-go visualisation in addition to Tableau. Tableau has an ongoing initiative to improve its mapping offering and have released excellent features during the past few years.

  • Programming
    At first I ran simulations using MS Access, VBA and ActiveX Data Objects. It executed serially, and some would often take 24 hours to complete!

    I quickly moved to use SQL (exclusively for over 3 years) but eventually found R a lot more scalable and efficient, not just for modelling, but other deliverables too. I progressively transitioned to R until it became my tool of choice for prototyping and rapid development.

  • Machine Learning
    During 2016 I started using python more extensively leading up to an 8-week course in Natural Language Processing in preparation to create a classification algorithm. Today scikit-learn is my default machine learning library although I am working towards using the TensorFlow framework.

  • Continuous Learning
    I have spent considerable time analysing data, focussing on prediction and inference. It naturally led me to the emerging Data Science field which, at the same time, had been exploding with powerful and accessible tools and techniques.

    Using an increasingly broader and more advanced range of solutions ultimately fuelled my a thirst to gain extensive and more in-depth knowledge in all crucial aspects of the discipline. I took a sabbatical to finish up courses in all foundational aspects of data science, bringing to an end a journey that started a few years ago. More accurately, it marks the start of my education, albeit on a solid footing.

    Data Science is a vast and evolving field with many disciplines. I can never know enough, so continuous learning has become a way of life.

    I am an active member of a few tech groups in Manchester and attend regular meetups in the city.

Nokia 2011 - 2013

In 2011 Nokia approached me to help demystify a convoluted KPI improvement programme. Its complexity made it difficult to demonstrate to the customer with sufficient confidence that the project was under control and on target.

I created and maintained various modelled scenarios of programme delivery, and it was used to prioritise, plan and control multiple work streams. The last of the outstanding target KPIs turned green just outside of a week of the deadline. This result would have been unlikely without systematic planning and prioritisation.

I set up a reporting solution for the Joint Benefit Project (JBP500) and completed optimisation analysis to use existing spare transmission tactically to expedite the programme. Using the spare capacity helped Nokia avoid significant liquidated damages and achieve the delivery target within the agreed time.

I created a reporting suite to manage the London Olympic programmes. An emphasis on management by exception allowed us to focus on the right things at the right time.

In 2013 Nokia commenced work on a significant, complex and large-scale programme without a suitable tool to manage the work. I got the ball rolling with a few tactical solutions but ended up designing, building and implementing ‘Nisaba’, a web-based workflow, data concentrator/ warehouse and reporting solution used to manage the 2G Refresh and the 3UK Buy-In and Free Carrier LTE (4G) programmes.

Nisaba was built using a LAMP stack and hosted on AWS, using a RabbitMQ as messaging and queuing solution for exports.

I specialised in MySQL, creating and configuring databases, provisioning users and setting privileges. I designed database structures and created normalised tables and optimised queries. I wrote Stored Procedures, Triggers and Functions to automate the solution and add features in conjunction with the application.

My team and I continued to provide the SaaS solution after I left Nokia as a consultant. It supported a few hundred users for almost 4 years, despite being introduced as a tactical solution only.

In 2015, Mobile Telecom customer MBNL praised the ease and speed of reporting generated from the Nisaba Database. The feedback was the tool provided, fast delivery with agility provided by changes of data input and output.

For me as a PM, delivering customer reporting on mobile network multi technology projects, NISABA as a tool has allowed simple views from complex cross work stream data. The reporting functionality was very simply tailored to meet the customer needs with instant updates to many of the output reports. The ability to access the tool securely from a number of different devices means we can manage data on the go. The stability of the platform in the last 3 years has also been excellent.

Andy Scott, Customer Project Manager - Nokia Networks

CTIL 2013 - 2016

Towards the end of 2013, I was headhunted to take over from Accenture as their contract came to an end. As part of the Beacon PMO, my primary and first project was to model programme delivery and create scenario and baseline plans of the Beacon and London programmes.

How was the information used?:

  • Vodafone used the predictions to launch regional (key cities and town) marketing campaigns in pace with the 4G Network rollout throughout the UK.

  • Generate budgets and targets, agreeing on corresponding budgets within Telefónica and Vodafone.

  • Organisational and resource planning, including proposals to the board to restructure the organisation per the Beacon programme life-cycle

Highlighting a few other projects:

  • Conducting an extensive and in-depth analysis of CTIL Estates rent and rates, producing time-series forecasts and providing information to set requirements for the network maintenance RFQ.

  • Implementing ‘AIR’, a tactical IT solution which extracted, transformed, concentrated and combined data from various sources providing a feed to the CTIL Acquisition workflow tool (SOL). It exposed privileged and secure access via a web interface and was used by the CTIL London Acquisition team as a case management solution.

Huawei 2016 - 2017

My primary role was to interface with the customer and provide reporting and analytics. I mapped all major operational processes using the BPMN2.0 standard.

My main achievement during this time was providing management information and steer to a successful turn-around project dealing with a significant backlog of quality issues. By the end of the calendar year, we closed down 92% of quality issues, with performance and status well within acceptable tolerances and work-in-progress pots.

Vodafone 2017 - 2018

Vodafone headhunted me for the position.

Some aspects of the role included:

  • Regional governance Reporting and Analytics
  • ETL solutions to import and transform high volume and complex 3rd party Network Performance Data every fortnight, and more frequently during the annual P3 drive surveys. This activity would complete within minutes, previously taking several engineers a week to perform, with savings estimated at c. £200k p.a.
  • Network metrics anomaly detection and association with potentially causal events
  • Financial, planning, budgeting and reporting. Vodafone was able to recoup at least £3m of overspend. I contributed by algorithmically auditing and classifying anomalous entries in extensive itemised purchase order schedules.
  • Business process analysis and mapping using the BPMN2.0 standard
  • Bespoke network analysis and supporting media used in corporate customer sales pitches

My headline achievement was developing and producing automated packs for Sales and Marketing. It identified and classified opportunities, using predictive clustering of network programme outcomes, which were then geospatially associated with existing and prospective customers.

Sabbatical 2018 - 2019

In December 2018 I took a sabbatical and completed training that gave me a thorough and exhaustive foundation in all essential aspects of data science. It marked the culmination of a journey that started many years ago.

I did not read mathematics at university even though majoring in highly numerate subjects like Statistical Research and Financial Management. Linear Algebra and Multivariate Calculus, in particular, are essential for a deep understanding of algorithms. In addition to mathematics, I aimed to refresh and update my knowledge of Statistics and broaden and deepen other machine learning aspects too.

Some of the courses and training included:


This phase follows my Sabbatical, relaunching my career from Telecommunications (Access Networks) contracting to permanent positions in Greater Manchester.

Co-operative Insurance Services Ltd 2019 - 2020

My primary role with the Co-op is to validate the transition and consolidation of legacy data, with subsequent abstraction, into a central repository, made up of Azure services. Once validated, the emphasis is on consuming the data, including creating marketing effectiveness reporting to propensity modelling.

ANS 2020 - current

ANS is the textbook example of what the Northern Powerhouse is attempting to achieve. I am committed to doing my part into growing Manchester into a thriving city, and my role as Data Developer with ANS allows me to contribute towards this grand vision.

I am building ANS’ Data Practice, leading a team of data scientists and engineers, to provide state-of-the-art, cloud-based, data services.

  1. A Multiple-Item Scale For Measuring Consumer Perception study (Parasuraman, Zeithaml and Berry)↩︎

  2. E2E: end-to-end management↩︎

  3. Geocomputation with R is a fantastic reference↩︎