A few years ago, a citizen corruption monitoring platform called Zabatek emerged in Egypt. Zabatek was a great idea but it was up against a phenomenon that many new initiatives of this type face: it’s much easier to set up a new platform than to find the resources, time and energy to keep it going over time.

This year, the same team introduced a new project. Called the MorsiMeter, it doesn’t require citizen reports to provide value. MorsiMeter aims to achieve transparency and accountability without getting a critical mass of citizen participation. Similar to a Chilean project called Del Dicho Al Hecho, its core team takes public information about President Morsi’s campaign promises versus presidential actions, analyzes it and presents it for anyone who is interested.

Why the shift? Partly, I would imagine, because of lessons that were learned with Zabatek.  I spoke with the team’s Amr Sobhy recently and asked him what those lessons were. Here are the takeaways:

1. Citizen participation and transparency projects do not keep themselves going on their own. They need human resources.

Amr’s team is an informal band of interested citizens who also have tech skills. Everyone has a full time job which makes it harder to find people with the time to undertake important tasks. This sheer lack of human resources had a lot to do with the shortcomings of Zabatek: they invested a lot of time in the beginning, thinking that it would keep itself going, and realized pretty soon that it would not.

2. If anyone who initially sends in reports never hears back, they are unlikely to return.

People at global organizations who study transparency and citizen participation projects from 10,000 feet above ground level (I’m including myself in this group) like to talk about feedback loops. Amr and colleagues’ experience underlines the importance of providing feedbacks to people who have submitted information or opinion. He quickly realized that, as he put it, “the user experience is broken: if someone reports and nothing happens you lose them, and you invest much more effort in either getting them back or getting someone new to report.”

3. Even if you choose not to run a marketing campaign to attract participants, you should still be prepared to convert new visitors into power users.

As Amr said, “it was not our goal for everyone to know about it, but to see how it would work with just the amount of people we have…” As a result, no effort was made to run a public outreach campaign for either Zabatek or MorsiMeter. Zabatek, however, was featured on a few television shows. This coincided with a spike in traffic that the team wasn’t prepared for. This meant a lost opportunity for engaging new participants who could have been information sharers or even volunteers.

4. Pivot! When there’s not enough citizen or institutional participation – consider shifting goals.

A project like Zabatek is only a resource if many people share information, and many people will not share information if they do not think it will lead to a response. Because it proved so difficult to elicit responses from the government on Zabatek, they changed their goal from “we will take citizen reports to the government” to “we will provide a platform for citizens to connect with other citizens that have the same problem.”

Pivoting is smart, but this particular new direction still required significant staff time. With their next project, MorsiMeter, they made an algorithm so it would be easy for the core team to update the project, and so that they could automate the updates where possible. According to the algorithm, an event is marked as a work in progress when there’s any initial relevant action and marked as completed when more than one source, in different geographic locations (preferably on different periods of times), say so.

While their algorithm cannot be 100% accurate, it does allow the project to exist even if it’s not regularly updated. This shift in tactic addresses the challenges (listed above) that the team first encountered in its experience with Zabatek.

5.  Reward participants by adding features that reflect their opinions or suggestions.

After it had been around for a couple of months, MorsiMeter added a voting feature for each item marked as “in progress” or “done.” This allowed them to get people to rank difference tasks according to their perception of its importance. From this data, they calculated a “general satisfaction percentage” for each item. The feature provides a way for citizens to influence the platform and, if anyone from the Morsi administration is looking, to influence governance.

Amr’s experience spotlights an important point: the more a platform manages to remain informative without reaching a critical mass of of citizen participation, the more value it holds for both viewers and participants. How many citizens does it take to create a valuable transparency website? Only a few.