Collect Data with Scale in Mind
What simple and repeatable processes do I need to have in place to enable data collection on a larger scale?
As social enterprises ready for scale, they must consider the types of data they will be able to collect through repeatable processes that can be cost-effectively and efficiently implemented by staff, partners, or others. Without this focus, more manual or resource-intensive data collection processes that may have worked well in pilot stages will slow an enterprise down and require unsustainable resources (whether financial, human capital, or otherwise) as the work scales. Finding the right balance of scalable yet robust data collection is critical for scaling success, as Dichter, et al. emphasize in “The Power of Lean Data”: organizations should move toward “use of methods and technologies for data collection that favor efficiency and speed while maintaining rigor.”
IN ACTION: Pratham, using simple, repeatable, and timely data collection
With an M&E team of more than 70 people in India alone, one might think that Pratham’s data approach is highly complex. On the contrary, because Pratham has so many people involved in data collection, it has strongly prioritized a data approach that is simple, easily repeatable by multiple stakeholders, and scalable. For example, one of Pratham’s core programs, Teaching at the Right Level (TaRL), involves simple learning assessments that are quickly and easily used by teachers to place students in appropriate learning groups and track progress throughout; this same data is also aggregated at various geographic levels for teacher monitors and education officials to use. For the teachers, the simple assessments are highly aligned with accomplishing their work (i.e., in delivering value to students), so it does not feel like an added or irrelevant exercise. While Pratham is directly involved with delivering TaRL through its own learning camps, it has also scaled by partnering with state and district administration to have TaRL delivered by government teachers at government schools. To ensure that the relevant information is collected consistently and in a timely manner across this growing cohort — and to ensure that the other stakeholders have timely access to relevant aggregated data — Pratham developed an Android application for teachers to input information and receive reports.
Top Tips for collecting data with scale in mind
- Keep it simple & repeatable.
Where possible, data collection processes should be simple and easily repeatable by someone with limited training (as different stakeholders may be collecting the data). For example, MiracleFeet has identified five simple indicators that can be easily collected and reported from the clinics in the field, such as average number of casts per patient (target of 4–6), percentage of patients receiving a specific surgical procedure as part of the treatment process (target >80 percent), and treatment completion rate. These metrics are easy to collect, do not require subjective interpretation but rather a simple yes/no for surgical procedure and a tracking and count for number of casts and completion, and are easily reported at appointments in real-time through a mobile app used by patient-facing personnel. Living Goods emphasizes the importance of incorporating elements in data collection tools that minimize data entry mistakes — such as setting ranges that a reported number must fall within and limiting open-ended questions. - Leverage technology to simplify data collection.
Many organizations have taken advantage of the prevalence of mobile phones to create phone-enabled templates or apps to ease and standardize data collection. CAMFED turned to a mobile monitoring app in 2012 when “paper-based data collection [had become] less and less viable” due to its scale. The primary challenges mobile collection addressed for CAMFED were reducing time to receive data and ensuring data did not get lost. The tool was able to replace paper records which had to be sent from rural schools to CAMFED offices. The mobile app also makes the step of inputting data (from paper records) redundant. Efforts can instead be focused towards unlocking data insights. Find more on leveraging technology to activate data collection at all levels of the organization here [LINK TO Activate data use at all levels]. - Make data collection a passive activity.
Rather than collecting data only through surveys which require active engagement from both its team and texters, Crisis Text Line tries to find data that is inexpensive and easy to collect, tied to existing customer touchpoints, and automated, when possible. In practice, this means that Crisis Text Line defaults to automated data collection, including text messages with metatags (e.g., time stamps, assigned risk level). Other organizations can embed data collection into their customer touchpoints, such as collecting data while onboarding a customer or automatically tracking purchasing behavior. Leveraging technology is one key way that data can be collected more passively. - Reduce costs by leveraging existing resources to support data collection.
Rather than bringing in M&E staff or external enumerators, CAMFED relies on its volunteers — young women in the community — to act as its data collectors. Not only is this cost effective, CAMFED also feels it is valuable as volunteers have significant trust of the community and are able to immediately share back insights with the community through their data collection. - Assess collection methods & tools for equity.
The above recommendations must take into account an equity lens which often means striking a balance between scalable processes and nuanced approaches to address systemic inequities. For example, simplification shouldn’t mean there are no opportunities to capture client voice directly through more qualitative methods. Attention must also be paid to any power imbalances inherent in data collection processes — for example a teacher asking for assessment of their teaching from students. For additional tips on assessing collection methods and tools using an equity lens refer to the Lay the Foundation article here.
Read next:
