Representative Engagements
Work that
speaks for itself.
Every organisation that comes to Diadem has a data problem nobody has fully solved for them yet. These are the kinds of problems we solve and how we solve them.
01
Development, field data and programme management
Case Study One
From Scattered Data to Decisions That Actually Landed
8 field locations, 2,400+ survey responses, delivered in 6 weeks
The Challenge
A development organisation came to us with months of survey data from several field locations. The data was there, but it was difficult to use. Responses were inconsistent, formats did not match, and the programme team was struggling to get a clear picture of what was happening in the field.
What We Did
We started by fixing the collection process itself. We rebuilt the survey forms in KoboToolbox, adding skip logic and validation checks to catch common errors before they entered the dataset.
We then cleaned the existing data, carried out the analysis, and built Power BI dashboards that made the findings easier for the wider programme team to understand.
Scope of Work
- Survey design and KoboToolbox development
- Data cleaning and quality checks
- Quantitative data analysis
- Power BI dashboard development
- Field data and monitoring support
The Outcome
The team could see what was happening across locations without working through multiple files or relying on guesswork. Programme discussions became more evidence-led, and the dashboards gave the team something clear to take into stakeholder meetings.
KoboToolbox
Python
Power BI
Data Cleaning
02
Public sector, development
Case Study Two
One System to See Everything
12 field locations, 30+ indicators tracked, live within 4 weeks
The Challenge
A programme team was managing activities across multiple locations, but there was no single place to see what was happening. Updates came through WhatsApp, paper forms and phone calls. By the time everything was pulled together, some of the information was already out of date.
What We Did
We built a centralised monitoring system around the programme's own indicators and the way its field teams actually worked.
Reporting was standardised, tracking was streamlined, and the system was kept simple enough for field staff to use without needing technical expertise.
Scope of Work
- M&E system design
- Indicator and reporting structure
- Digital data collection
- Progress tracking
- Reporting templates
- Staff training and capacity building
The Outcome
The programme coordinator could open one system and see progress across all 12 locations, including activities completed, outstanding items and reporting gaps. Problems could be spotted earlier instead of waiting for the next round of reporting.
M&E Systems
KoboCollect
Excel
Capacity Building
03
Agriculture, development
Case Study Three
When the Data Was Always There But Nobody Was Using It
5 seasons of historical data, 4 dashboards built, delivered in 5 weeks
The Challenge
A small agriculture organisation had years of data sitting in spreadsheets. They had collected it season after season, but the information was rarely used beyond basic reporting. When it came to planning, much of the discussion still depended on experience and intuition.
What We Did
We cleaned and combined the historical datasets, identified the indicators that mattered most to the organisation, and built interactive Power BI dashboards to make the information easier to explore.
We also used Python to work through the larger datasets and identify patterns that would have been difficult to find manually.
Scope of Work
- Historical data consolidation
- Data cleaning and preparation
- Indicator selection
- Exploratory data analysis
- Power BI dashboard development
- Python-based data analysis
The Outcome
The organisation could use its historical data to look at trends, compare performance and support planning decisions. Instead of sitting in spreadsheets, the data became something the team could actually work with during planning meetings.
Power BI
Python
Excel
Data Analysis
04
Social development, health, agriculture
Case Study Four
Complex Surveys. Built for Reliable Data.
3 languages supported, 5,000+ survey responses collected, <2% error rate at close
The Challenge
Across projects in social development, agriculture and health, we have designed and deployed digital surveys for organisations collecting detailed information from people and communities in the field.
The questionnaires often need to handle multiple languages, branching questions, offline deployment, validation requirements and field teams working under pressure.
What We Did
We handled the process from instrument design to deployment. We designed the questionnaires, programmed them using KoboToolbox, ODK and SurveyCTO, built the required logic and validation checks, tested the forms in the field, made corrections where needed, and trained the teams responsible for data collection.
Scope of Work
- Digital questionnaire design
- KoboToolbox, ODK and SurveyCTO programming
- Skip logic and validation rules
- XLSForm development
- Field piloting and testing
- Enumerator training
- Data quality checks
The Outcome
The teams went into the field with tested digital instruments rather than improvised forms. At close, the datasets were structured and ready for analysis, with an error rate below 2% across the deployments represented here.
KoboToolbox
ODK
SurveyCTO
XLSForm
Training