With the recent announcement of the UN’s Sustainable Development Goals for the next fifteen years, it’s clear that there will be more intense scrutiny on organizations to deliver work which yields strong, quantifiable results. This means it’s now more important than ever to make sure you’re using the right mobile data collection tool. However, there are so many tools out there that it can be overwhelming to decide on the right one. At TaroWorks, we speak to lots of organizations at the early stages of looking for tools, and want to share 5 major themes that come up consistently, so that you can find the most appropriate one that allows you to focus on making an impact.
1. What will the data be used for?
Field research: Generally one-time projects spanning over a few months, where you don’t need a heavy tool and a general data collection tool should be adequate
Impact monitoring & evaluation over time: If you need to track results over time for a specific entity, make sure the tool allows for that. If you sample and don’t need to track specific entities, then a general data collection tool can work.
Managing business operations: Monitoring ongoing transactions in order to make business decisions. This entails managing field operations and supply chain activities, and may require tools that go beyond simple data collection.
2. What kind of analysis are you doing?
Ad hoc reporting: If you’re doing one-off analysis and need quick results, you may not need anything heavier than an Excel spreadsheet!
Monitoring over time: If you have set performance indicators that will be monitored over time, your tool should be able to capture historical information, and generate those indicator figures.
Statistical analysis: If you’re using stats packages like SPSS or Stata, make sure that your data collection tool can export data easily, in a format that plays nicely with your stats package.
3. Where are you storing data?
Where you store your data is significant because it affects the analysis you can do, and who will have access to it. There are 3 factors that define the spectrum of choices: the degree of centralization you need, how flexible the system is to changes, and the amount of set-up time needed. At one end of the spectrum, there are spreadsheets and databases (i.e. Access, SQL), which are easy to spin up, are super flexible for making changes, but are very decentralized systems. At the other end of the spectrum are CRM (i.e. Salesforce, Zoho, MS Dynamics) and ERP (i.e. SAP, Oracle) systems, which are increasingly cloud-based and therefore centralized so all analysis is accessible throughout the organization, are more standardized, but require more upfront set-up.
4. What level of resources do you have?
There are 3 types of resources to consider. Depending on these, you may need to consider providers who offer service options like set-up, maintenance and training.
Staff skills: Do you have staff who can implement a tool? Cast away the impression you need a software developer. Previous knowledge of spreadsheets, databases, system architecture or mobile devices may be a good starting point. Genuine interest in implementing a system can get you the rest of the way there – some of our fastest implementers have been people who’ve never used databases!
Staff bandwidth: Consider that implementing a more centralized and standardized tool may require someone’s full capacity in the initial months, but will taper off over time. Meanwhile, a less standardized tool will take less set-up time, but may require a consistent level of attention over time.
Budget: Now let’s talk budget. Is budget for the solution coming from in-house or is it outsourced? Also, there are solutions which are free but they won’t necessarily meet all your needs. Other solutions require investment but come with more robust capabilities.
5. Where is your data coming from?
Now let’s consider the actual point of data capture, as this will really narrow down the tools you consider.
Direct channels: Getting information from beneficiaries or clients directly through an SMS or IVR system. A direct solution is scalable, discreet, and cost effective because it doesn’t require equipment. However, this relies on accurate self-reporting, and doesn’t allow for long or robust forms.
Intermediated channels: Getting information via a field agent. Intermediated systems can provide robust and deep data sets, doesn’t require beneficiaries have devices, and works well with programs who may already have staff in the field. However, they can be more costly because of equipment, staff, and training.
Written by Emily Tucker, CEO of TaroWorks. TaroWorks is a mobile data collection tool that seeks to make it as easy to manage across the last mile as it is across the first mile. Learn more at www.taroworks.org
Thanks for laying out some key considerations when choosing a mobile data collection tool. I agree with all the considerations outlined, though I would add a couple more that we at SurveyCTO think are really important (if not the most crucial) but are often overlooked in the development sector:
1. Data security: As more and more organizations are capturing more and more data electronically and storing it in the cloud, sharing it more widely, etc., they also need to consider whether the platform they are using is actually secure and will keep identifying information and other sensitive data encrypted and private. Questions to think about are: Is my data encrypted at all times (both in transit and at rest)? What kinds of firewall and security protections does my cloud server have (assuming one is using a cloud-based service)? Is the data hosted in a “zero knowledge” environment (where technicians behind the software cannot view user data even when accessing servers for maintenance, etc.)?
To give a small example people don’t often consider: any service that shows charts/visualizations/dashboards directly in the cloud server is not “zero knowledge”: for the visualizations to be produced the cloud server has to be able to read the data, which means the data stored in the cloud is completely readable to anyone who can access that cloud server (including staff from the company providing the service). So a safer option is to go with a tool that allows for encrypting data even at rest on the server and perform visualizations off the cloud.
2. Data quality: Whether the monitoring, analysis of indicators, etc. that is being done on one’s data is good or bad depends, first and foremost, on the reliability of the raw data itself. While moving from paper-based surveys to mobile data collection tools has improved data quality, electronic data collection alone does not guarantee good (or even real) data. The data collection process, especially for organizations doing field based surveys, is complicated and dirty, with lots of room for errors and fraud. Organizations often spend hundreds of thousands of dollars collecting data – which is a huge investment – and it makes sense to make sure that the data is accurate and usable. So it is important to select tools that make it easy to capture rich metadata on surveyors to detect fraud and make it easy to check data for error patterns, so that one can act on and fix any issues quickly. Good data = good decisions.
I think the amount of data that any data collection tool consumes is a serious consideration as well. For rural areas, data costs are a major factor.