I built an R Shiny decision-support tool that compares four urea rate/timing combinations across five mega-environments, districts and seasons — joining field-trial data with SoilGrids soil and CHIRPS rainfall. Soil pH carried about 42% of variable importance; urea treatment 28%; district 18%.
Split urea at V6+V10 gave the best observed yield in the Eastern Savannah, 7.45 t/ha. Treatment differences were significant in 4 of 5 environments, and the yield model (R² 0.623, RMSE 1.24 t/ha) was assessed honestly: good for ranking strategies, not for promising a farmer a number.
LIVE DEMO
Optimal urea rate/timing by environment, yield drivers, an honestly-assessed yield model and 600+ mapped trial locations.
test · password testBaseline survey of 1,628 households turned into an 11-module dashboard with maps, disaggregation and downloadable tables.
test · password test
LIVE DEMO
Real-time KPI tracking during data collection: completion rates, data-quality flags and daily enrolment across sites.
LIVE DEMO
HIV testing, PrEP uptake, SRH services and mental-health indicators for adolescents aged 10–24 across programme sites.
LIVE DEMO
IPC classification, Food Consumption Score and livelihood indicators to target food assistance faster.
Poverty disaggregated into education, health and living-standard dimensions to target interventions beyond income.
How donor funds flow across Kenya's counties and sectors — spot under-funded areas at a glance.
Medicine demand by county, product and package, with weekly, monthly or yearly quantity predictions.
PART 1Maize Case Study
Comparing four urea rate/timing combinations across Rwanda’s mega-environments, districts and seasons — and why fertiliser recommendations must be location-specific.
PART 2Maize Case Study
Soil pH accounted for roughly 42% of variable importance in a Rwanda maize yield analysis — ahead of urea treatment (28%) and district (18%).
PART 3Maize Case Study
Turning maize trial research into an interactive R Shiny decision-support application built around the questions extension officers actually ask.
PART 4Maize Case Study
A linear yield model with R² = 0.623 and RMSE = 1.24 t/ha — and why statistical significance alone doesn’t make a good prediction tool.
Monitoring & Evaluation
In the context of Kenya’s healthcare sector, absorption rate refers to the percentage of allocated funds that a county effectively uses for its intended…
Health & Demography
Ethiopia has made significant strides in improving child nutrition over the past two decades. However, child stunting remains a pressing issue, with many…
“Data has little value sitting in a spreadsheet. The value comes when it changes a decision.”
That is the test I hold every engagement to — whether it's an evaluation design, a monitoring framework or a dashboard. Read: from analysis to decision support ›
Surveys & Field Data
In today’s world, collecting data efficiently is critical for research and decision-making. Whether you’re conducting surveys in remote areas or urban…
Surveys & Field Data
Creating well-structured surveys is key to collecting accurate and useful data. With KoboToolbox, you can design surveys that are easy for both…
Surveys & Field Data
One of the best features of KoboToolbox is its ability to automate quality checks during data collection. This ensures that the data you collect is clean…
GIS & Climate
Integrating citizen science data into conservation research: GIS, climate covariates and regression modelling in R.
Data Analysis
The global wildlife economy plays a crucial role in fostering sustainable development, preserving biodiversity, and providing economic benefits to local…
Free 15-minute discovery call · written proposal & quote · most dashboards delivered in 2–4 weeks with training and documentation.
DASHBOARDS FROM
Typical delivery 2–4 weeks
Services & process ›Scientific Reports (Nature Portfolio), 2025.
Read the article ›