Pulse: Using texting diaries to discern Philly's information needs

Kristine Villanueva and Andrew Haeg
A matrix table showing the kinds of information people said they needed, whether they were able to get the info, and how badly they needed it.

How do you listen to an entire city? It’s a question I have mentally pinned–a big-swing aspiration to making journalism that’s truly responsive to the needs of citizens.


Picking that question apart reveals layers of daunting logistical and technical challenges: How do you scale listening — typically an intimate, one-to-one act — to thousands, or even millions? How do you reach people you’re not already connected to? Even if you could, how would you make sense of what they were saying?

These are the questions Resolve and Relatable (a new collaborative consultancy formed by Public Input and GroundSource) have set out to find answers to with Pulse, a GNI-funded texting diary project to discern the information needs of Philly residents. Participants receive $25 for completing two series of text message prompts a day for three days–answering questions about information they’re seeking, why they’re seeking it, and whether they could find it, as well as questions about topics of interest to Resolve.

We’ve completed the first round of Pulse, in which 20 people drawn from a pool of 78 people initially screened (who represented 30 zip codes and 39 Philly neighborhoods) completed the three-day texting diary. Their answers revealed gaps between their need for important information (e.g. how to get help paying for rent or utilities, neighborhood-level concerns) and their ability to easily find it. They also were asked to share their personal stories, to put those needs in the context of their day-to-day life.

So how does listening to 20, or 100, or even 1,000 people like this make it possible to listen to an entire city? Our thesis is that by building Pulse panels that reflect the city’s demographics, and listening deeply enough, you can extrapolate information needs for a much larger population. We’re leaning here more into the tactics of design thinking or ethnography vs. statistically valid polling — deeper insights from a small, representative group can inform what a newsroom covers more broadly.

We will be tweaking and refining our approach to outreach, survey design, and simply administering each Pulse to make it easier to do over and over; but that’s only half of it. Each Pulse generates a great deal of data and personal stories, spread across multiple prompts, over three days.

Equally Informed Philly’s textline is anonymous so the first pulse was the first time the team has received in depth insight on the textline subscriber day-to-day information needs. Other than an end-of-year survey and data the team keeps for responses, we are generally limited to insights on broader information needs than needs for the day-to-day. Through these Pulses, we are able to tweak our overall text line strategy and compare this year’s and last year’s end-of-year survey results to the one the team plans to send in December. This way, we will try to understand the impact of the changes we plan to make while we experiment throughout the year.

In addition to seeing trends in needs, the team was surprised to see more participants from communities of interest for example, seniors and LGBTQ+ people. One respondent asked about how to move to a senior living facility or how to navigate certain technology. The team also received a response related to anti-Asian racism from someone who identifies as trans, which reflects some of the work Resolve Philly’s community engagement team is currently doing.

Another objective of Pulse is to automate organizing and visualizing responses, again with an eye to making it easier to discern participants’ needs. We prototyped this by plotting the answers in Airtable to two questions, based on the importance of the information they were seeking and whether they were able to find that information. The results for the day 1 morning prompt, pictured above, reveal a cluster of info needs related to housing, and utility assistance.

It also reveals areas where people are already well-informed, which may be instructive. Visualizations like this, and other analytical tools we’ll be building, will help newsrooms become open to gathering more feedback so they can ground coverage and product decisions in lived experience, instead of guesswork. This can help newsrooms better align their work with actual needs, becoming more responsive and more efficient.

This is the first of what will be several iterations of Pulse this year. Each one will provide the opportunity to refine each of the steps, including outreach, survey design, synthesis, and sharing the results in a visual, easily accessible format with the broader Philly media and social services ecosystem. The larger goal is to make administering Pulse-style projects easy enough for small newsrooms to pull off, and powerful enough to serve as a window into the needs of an entire community.