Using technology to help citizens

The success of the CFM demonstrates that technology can play a central role in helping us reform government.


Dr Umar Saif May 22, 2014
The author has a PhD from the University of Cambridge and worked as the Chairman of the Punjab Information Technology Board. He tweets @umarsaif

For many citizens of Pakistan, interacting with the government for even the most basic facilities, such as obtaining a domicile, registering a property or acquiring a driving licence becomes a difficult process fraught with uncertainty, corruption and bribery. The rigidity of government processes and lack of monitoring within government structures creates significant barriers between it and the citizens.

Information about pockets of corruption and bottlenecks in processes can easily slip through the cracks. Paying a bribe of a few hundred rupees is hardly frowned upon and is, in fact, routinely accepted as ‘speed money’. No one even bothers to report such petty corruption.

But why can’t the government set up mechanisms to solicit citizen feedback? Why can’t the government simply reach out to the citizen and ask them: You just interacted with a government office for a service, did everything go well?

Simple? Indeed. And that is exactly what we started doing almost two years ago.

Our system, called the Citizen Feedback Model (CFM), proactively calls the cell phone of a citizen to solicit their feedback about their interaction with the government. The data is automatically compiled, processed and analysed by the chief minister’s office. So far, over 4.4 million citizens have been contacted through the CFM and nearly 4,000 corrective actions have been taken on the basis of the feedback received.

Here’s how the system works: when a citizen accesses a government facility for a service, his/her cell phone number is logged into a central database along with the details of their transaction. Once this happens, the citizen receives an automated call in the voice of Punjab Chief Minister Shahbaz Sharif, asking for feedback, followed by an SMS from 8070. These feedback SMSs are then read and categorised, random calls are made by call agents and the results are analysed by a team of data analysts. The results are used to benchmark services across the province and ‘district scorecards’ are sent to the DCOs from the chief secretary’s office every week. The system includes automatic checks to ensure that the phone numbers being recorded are not repetitive or invalid, and historical trends are analysed to verify that the office is not under-reporting data. Bogus or incorrect numbers entered in the system negatively affect a district’s scorecard.

The CFM has several subtle features that have led to its scale up. First, it is not a complaint cell; we do not wait for a citizen to be dissatisfied enough to go through the process of lodging a complaint. In fact, the system does not receive unsolicited complaints; the system calls out only those citizens who have freshly interacted with the government for some specific service. The system is designed to gather data about specific parameters of service delivery and avoid generalised opinions. The CFM is not an individual grievance forum; actions against personnel are automatically initiated if there are repeated complaints, but each complaint does not trigger an open-ended inquiry. Each day, the system calls out between 10,000 to 30,000 randomly chosen citizens (who have freshly interacted with the government); the statistically significant and randomised data set minimises the chances of administrative biases. It also helps us control the volume of the feedback to a level against which the government can affirmatively take action. Presently, after about two years of operation, we have also started informing citizens about the actions taken on the basis of their feedback.

The nuanced features of the system have enabled us to keep ourselves focused on using it as platform for data-driven analysis to improve service delivery. Working with a great team of experts – most notably World Bank’s Zubair Bhatti, whose Jhang model (when he was DCO Jhang) inspired this system – has helped us avoid the caveat of the CFM becoming a traditional government complaint cell.

Currently, the CFM is operating with 17 government services in Punjab, including Police 15, Rescue 1122, emergency medical services, indoor patients at rural health centers, issuance of domicile, driving licence and character certificate, and district revenue branch operations. In the near future, it will be extended to cover more services.

Other provincial governments in Pakistan have already started adopting the CFM model. While we still have a long way to go, the success of the CFM demonstrates that technology can play a central role in helping us reform government.

Published in The Express Tribune, May 23rd, 2014.

Like Opinion & Editorial on Facebook, follow @ETOpEd on Twitter to receive all updates on all our daily pieces.

COMMENTS (21)

mazhar kaleem | 9 years ago | Reply

Umer Saif was running his own company and was heading up the PTIB. Either he was duffer to understand the obvious conflict of interest or he is utterly dishonest which is much more dangerous in positions of responsibility. He had made mass SMS sending system which he then infiltrated to all parts of Government. I am more than half sure that he must have copied the idea of that too from somewhere. He advertised himself as most young innovator under 35 on basis of simply that mass SMS sending and a failed SMS based social network. Crooks like him have are present in government science machinery who block other's good ideas, block them or copy them to portray as their own. ICT RnD fund had long list of discouraging good ideas and good people and made sure no economic activity takes place in country based on technology. This crook Umer Saif should be told to either leave his private companies or leave the Government positions.

Ayaz Jaffar Ahmd | 9 years ago | Reply Great system, but only 4000 corrective measures out of 4.4 million feedback defies the purpose of the project.
VIEW MORE COMMENTS
Replying to X

Comments are moderated and generally will be posted if they are on-topic and not abusive.

For more information, please see our Comments FAQ