Our Work

Selected Projects

A sample of the research and analytics work we’ve delivered — across business strategy, medical imaging, national survey data, demand forecasting and international relations. Each one shows the method, not just the headline.

Business & policy writing Statistical modelling Python & machine learning Primary survey research

AT Acumen Research

Division 01

Written work that has to hold up to an examiner, a reviewer or a board — strategic audits, comparative analysis and academic argument, researched properly and written to be read by people who will push back on it.

01 — Adidas AuditStrategic audit of Adidas
Professional Report Writing

Strategic Audit of Adidas: External & Internal Business Analysis

A full audit of Adidas’ competitive position, benchmarked against Nike. PESTLE and SWOT analysis fed a set of distribution recommendations — including widening digital reach through Google Shopping and JD Sports.

PESTLESWOTCompetitor benchmarking
02 — Security EssayAcademic essay on security and conflict
Thesis & Proposal Writing Support

Academic Essay on Security, War, Peace and Conflict

An essay on the contested meaning of “security” in international relations, weighing realist, poststructuralist and non-traditional readings against one another, with the War on Terror as the case study for how those definitions shape policy.

International relationsCritical analysisCase study method
04 — Electric DiggersElectric versus diesel mini diggers analysis
Professional Report Writing

Electric vs Diesel Mini Diggers: Cost & Sustainability Analysis

A comparative analysis written for a construction audience — emissions, fuel cost, operational performance and the regulatory picture, set out to support a switching decision rather than to advocate one.

Comparative analysisCost modellingSustainability

Need work like this?

These projects came out of three services in the Research division. Each page sets out what’s included, how long it takes and what you get back.

AT Acumen Analytics

Division 02

Data work where the method has to be defensible — the right test for the design, the right model for the question, and results explained in language a non-specialist can act on. From single survey datasets through to deep learning on medical images.

03 — Cartilage MLDeep learning cartilage segmentation in MRI
Big Data Analytics, ML & AI

Deep Learning for Cartilage Segmentation in Knee MRIs

How convolutional neural networks can segment knee cartilage automatically to support osteoarthritis diagnosis — with the emphasis on keeping the method reproducible without specialist hardware or heavy programming.

CNNMedical imagingImage segmentation
05 — Gender AnalysisSocio-economic gender analysis of ONS data
Big Data Analytics, ML & AI

Socio-Economic Gender Analysis of UK ONS Data

Gender disparities across education, employment, ethnicity, marital status and unpaid care, analysed from Office for National Statistics data in Python — cleaning and preprocessing, KMeans and hierarchical clustering, PCA for visualisation, and Random Forest modelling.

PythonClusteringPCARandom Forest
06 — Demand ForecastingQuarterly demand forecasting and decision analysis
Data Analysis & Interpretation

Forecasting and Decision Analysis for Quarterly Demand

Quarterly demand projected with moving-average and static forecasting, accuracy tested through MAD and MAPE. Regression identified the demand drivers; decision-tree analysis set expected return and present value against each option.

ForecastingMAD / MAPERegressionDecision trees
07 — Protest SurveyPrimary survey analysis of protest behaviour
Data Analysis & Interpretation

Primary Data Analysis of Rebellious Behaviour

A primary survey study of public willingness to take part in peaceful protest. Large survey datasets cleaned and coded to a structured scheme, with index reliability confirmed by Cronbach’s alpha and factor analysis before any interpretation was drawn.

Survey designCronbach’s αFactor analysis

Your project could be the next one here

Tell us what you’re working on and we’ll come back with an honest view of what it needs — method, timeline and cost — before you commit to anything. Your first 15 minutes are free, with no obligation.