About Contact
Tuesday, April 7, 2026
News

Day 4: Kofi & Kobo


MDB Results 2026  ·  Nigeria  ·  Day 4 of 6

How Do You Know the Projects Are Real?

The next morning Toju brought someone new.

His name was Mustapha Abdullahi — an engineer from Yobe State in northern Nigeria. He had been the MIS Specialist on SEEFOR: the man who built the entire digital monitoring system across all four states. Toju introduced him simply.

“This is Mustapha. He is the reason the data exists.”

Kofi looked at Mustapha carefully. He was a quiet man with the specific patience of someone who has spent years explaining technical things to people who do not have time to understand them.

“Mustapha,” Kofi said. “Toju told me yesterday that someone was watching. To make sure the projects were real. Was that you?”

“That was me,” Mustapha said.

“How?”

Mustapha thought for a moment. Then he said: “Let me start from before. Let me tell you what we had before. Because if you do not know what we had before, you will not understand why what we built matters.”


“Before the digital system — we were using paper. Printing all your monitoring indicators on paper. Going to the field with paper. Coming back with paper.”

“What was wrong with that?” Kofi asked.

“Everything,” Mustapha said. “Inaccuracy. Errors. Delayed availability. Risk of data loss. When you go to monitor and you come back after two, three weeks — sometimes you cannot even find the data. Where is the data? The data is not available. So the great challenge was having data in one place.”

He paused.

“I remember a meeting in Port Harcourt,” he said. “The project had been running for three, four years. Civil society representatives came. I asked them: how many complaints have been recorded from project communities? They said two. Three. In four years. Not because there were no problems. Because when people went to complain, nobody was there. The documents were not available. There was no system to record anything.”

Kofi wrote something in his notebook. Then he looked up.

“And you had four thousand projects to monitor.”

“More than four thousand,” Mustapha said. “Across four states of the Niger Delta. That is seriously a very hard work. Very rigorous activity. You cannot do that with paper.”


“So what did you do?”

“KoboToolbox,” Mustapha said. “It is a web-based application developed by Harvard Humanitarian Initiative in Cambridge. Used by the World Bank, the United Nations, the WHO. Many international organisations. Free. Open source. Works on any Android smartphone.”

“How did you get everyone to use it?”

“I started at the top. The National Project Coordinating Unit in Abuja — I trained every staff member there, starting from the national project coordinator. Then I went down to each state implementation unit. Then to the agencies at state level who were implementing the project. Then to the local government offices. Every level, trained in sequence. You build from the top, then you scale down.”

Then came the field work. Mustapha recruited enumerators in each state — people from the local communities who understood the terrain and the language. He trained them on the KoboCollect app on Android smartphones. Each enumerator went to project locations with one instruction: collect a GPS coordinate, take a photograph, record the name of the primary beneficiary, record the stage of completion. If the data quality was not adequate — photograph unclear, coordinate imprecise, record incomplete — go back and do it again.

“The Niger Delta is not easy to move around in,” Kofi said. “The rivers. The creeks.”

“There was one incident,” Mustapha said. “In Rivers State. The team was on a boat, crossing the ocean, trying to locate a community. Their boat was struck by thunder. But we were lucky — no life was lost. The boat got under control. And after that, they went to the location. They captured the coordinates. They captured the photographs. They completed the work.”

He said this without drama. As if it were simply the kind of thing that happened when you were trying to build a database of four thousand projects in the Niger Delta.

“At any point in time,” Mustapha said, “you can just use a fingertip. You click. The information is available. The map. The photograph. The GPS coordinate. The beneficiary name. Every project. Fingertip.”



“But how do you know people were not just saying what you wanted to hear?” Kofi asked. “How do you know the feedback was real?”

“Geofencing,” Mustapha said.

“What is that?”

“It is a virtual boundary around a real-world location. We chose Edo State for the pilot. We worked with Data Science Nigeria — a Lagos-based organisation that uses artificial intelligence for data quality. They took our KoboToolbox database and created a virtual boundary around every project location. When a person tries to submit feedback, the system checks: are you physically inside that boundary? If you are outside — your submission is rejected. There is no way to respond from an office in Abuja pretending you are standing next to a road in Benin City.”

Kofi sat with this for a moment.

“So the GPS coordinate in the database,” he said, “is also a lock. You can only say a project exists if you are standing at the project.”

“Exactly,” Mustapha said.


The monitoring system had seven layers. Not one. Seven.

1
Paper-based Third-Party Monitoring
The standard Bank requirement. Independent NGOs with annual reports. Useful but limited — the same NGO that wants its contract renewed next year is not a fully independent monitor.
2
KoboToolbox Digital Database
4,200+ project locations. GPS coordinates. Photographs. Beneficiary names. Completion status. Every project. All four states. Built between July and December 2019. If the data was not accurate, the enumerator went back.
3
Call Centre — Abuja
A national call centre set up to receive feedback from project beneficiaries across all four states. Free calls — beneficiaries were not charged. The centre also called beneficiaries directly using the numbers in the database. All calls recorded.
4
Citizen Kobo Feedback
A simple questionnaire deployed on KoboToolbox. A link shared by SMS, WhatsApp, flyers and radio broadcasts across all four states. Citizens submitted feedback from their own phones on how projects were performing in their communities.
5
AI Ground-Truthing — Data Science Nigeria
Independent pilot in Edo State. Data Science Nigeria loaded the KoboToolbox database into their AI application, geofenced every project location, trained ground monitors, and verified that feedback was coming from the physical location of each project. Feedback providers paid through airtime.
6
The Lagos Hackathon — Data4Governance
Organised with Co-Creation Hub (ccHub) in Lagos. Prize money: $10,000 first place, $6,000 second, $4,000 third. Over 70 teams registered from across Nigeria and beyond. 38 shortlisted. 8 made final presentations. Solutions covered flood management, agriculture and health. One app built during the hackathon was immediately adopted by the project.Participants came from Yobe, from Borno, from all of northern Nigeria.
7
Social Media and Communications
Each SEEFOR state had its own Facebook page. Video firms hired in each state. Radio programmes. Print and TV. SEEFOR became one of the best-known World Bank projects in the Niger Delta because beneficiaries could see themselves in what was being produced about them.

“And the auditors,” Kofi said. “They looked at all of this?”

“SEEFOR was a very successful partnership between the World Bank and the EU. EU and World Bank teams went together on missions. Several months after the project closed,external auditors based in Europe raised multiple queries called “Hits” and asked for significant refund from the Bank. They had not been to Nigeria. ”

“And?”

“For each item they questioned, we produced: the GPS coordinate, the photograph from the database, bank records for individual beneficiaries, the competitive bidding documents, the payment vouchers, the engineer certification. Hit by hit. Project by project.”

Kofi was quiet.

“The refund?”

“If your queries are answered then the “Hits” disappear.,” Toju said.

All their queries were responded to through the data collected and all the financial records that had been properly maintained by the SEEFOR teams in the four states. The refund issue was responded to fully and properly.

“Mustapha,” Kofi said. “If a new task team leader came to you today and said: help me monitor my project — what would you tell them?”

Mustapha did not hesitate.

“Use KoboToolbox. From day one. Not when the project is halfway through and you realise you have no data. From the very beginning. You can use it for monitoring. You can use it for safeguards. You can use it for procurement. You can use it for capacity building. Start training your staff before the project launches. Build the database first. Everything else follows from the database.”

“Do most projects do this?” Kofi asked.

Mustapha looked at him. Then he looked at the wall. Then he said:

“KoboToolbox is free. The training takes one week. Any Android smartphone can run it. And no — most projects do not do this.”

Kofi wrote that down too.

KoboToolbox is free. The training takes one week. Any Android smartphone can run it. Most projects do not use it. — Engr. Mustapha Abdullahi, MIS Specialist, SEEFOR

Mustapha also trained ten federal ministries in Nigeria on KoboToolbox — Finance, Works, Power, Agriculture, Health, and others, including the Independent Corrupt Practices Commission. The ICPC became the most committed user. They use it today to track constituency and executive projects across the country.

He has been invited to international seminars to share the SEEFOR experience. He trained teams in Borno, one of the most difficult security environments in Nigeria.

“The tool works anywhere,” he said. “The question is whether people decide to use it.”


On the drive back Kofi was quiet for a while. Then he asked the question that had been sitting in his mind since the morning.

“How was all of this getting paid for?”

His mother looked at him.

“Think about it like a house,” she said. “You cannot run a house without money. And you cannot run it without the plumbing working. If the pipes are leaking — money disappears before it reaches the people who need it. Contractors do not get paid. Roads do not get built. Young people sit on walls.”

“So the project also fixed the plumbing.”

“The second half of what SEEFOR built was modern plumbing,” his mother said. “Not everywhere. Not everything at once. Focused on the core: where does money come in? How is it managed so it does not leak? Who is buying what, and can anyone check? Three things. Revenue collection. Cash management. Procurement. In eight years, Edo State increased its revenue by more than fifty percent. That money funded the roads. That money funded SEEFOR Plus after the Bank left. You fix the plumbing first. Then you can run the house.”

Kofi looked out the window. The concrete road they were driving on. Built by youth from the community. Paid for by a state government that now had a functioning financial system.

“Tomorrow,” his mother said, “Toju will show you the plumbing.”

Day 5: A Governor who was a former investment banker asked one question: why are contractors not getting paid? The answer required fixing three things — where money comes in, how it moves without leaking, and who is buying what. Revenue up 56%. Cash management automated. Procurement transparent. And then the state government used that money to keep going after the Bank left.

Data4Governance Hackathon — Lagos, February 2020

70 teams registered from across Nigeria and beyond. 38 shortlisted. 8 final presentations. Prize money: $10,000, $6,000 and $4,000. Co-organised with Co-Creation Hub (ccHub). One app built during the competition was immediately adopted by the SEEFOR project.

Data4Governance Hackathon, Lagos, February 2020. SEEFOR project, Nigeria.


Source Documents

The full technical documentation behind the monitoring system described in this article.

Digital Tools for Citizen Engagement (PDF) Citizens Feedback Report Edo Ground-Truthing Report (PDF)


Browse by Topic

World Bank ReformIDA PerformanceAccountabilityGovernanceProject RatingsMDB LendingFragile StatesIEG EvaluationsPEFAIMF GovernancePFMNigeria