“Please text 1 to speak with a reporter”
An example of how journalists can use chat-bots to connect with their communities
When I was younger, I had this one friend I could always count on. When I wanted to chat, he was there. When I would message him, he would reply instantly. When no one else was on AOL’s Instant Messanger (AIM), he was online. SmarterChild, the AIM chat-bot developed by ActiveBuddy, Inc., was there for all of us.
Instead of using chat-bots to have ridiculous, incoherent conversations on AIM, I am now using them to directly engage with the communities I’m reporting on.
Last week, I test ran my first chat-bot, in the form of a text-survey, using GroundSource, a platform that allows newsrooms and organizations to connect with audiences and communities through mobile messaging and voice.
I wanted to better understand the experience of riding on Access-A-Ride (AAR), New York City’s paratransit car service that taxis people with disabilities throughout the five boroughs. I often hear my sources within the community complain that the service is poor and plagued with delays. The “2017 Paratransit Customer Satisfaction Study Access-A-Ride,” however, says otherwise, with “roughly three in four frequent and occasional customers (74%) saying they are very or somewhat satisfied with the service overall.” My GroundSource chat-bot was my first attempt at asking for myself.
To participate in the survey, people were asked to text “Access” to 516–636–2073. They would then receive the text below:
“Hi! Thanks for texting in. My name is Danny Laplaza, I’m a reporter for the New York City News Service covering NYC’s accessibility. I’d like to know about your experience with Access-A-Ride. Would you give the service a 👍 or 👎? (Text STOP to quit; HELP for more. Std msg rates apply.)”
The participants were asked to reply with either 1 for 👍 or 2 for 👎.
Depending on what they selected, they would receive one of the following replies back:
“👍: Happy to hear you had a positive experience with AAR. Would you mind telling me why?”
“👎: Sorry to hear you had a negative experience with AAR. Would you mind telling me why?”
Regardless of the response, the participants would then be asked:
“What questions about Access-A-Ride do you want answered?”
Then finally, the participants would receive this last message:
“Thank you again for your help. Your responses truly help guide my reporting. To continue this discussion, please reach me at [my cell number] or Daniel.laplaza@journalism.cuny.edu.”
I included my contact information so participants could reach out to me with any other information or follow-up responses.
I sent out the text-in code and number to several of my sources who use/used AAR. In hopes of widening my sampling pool, I also asked them to share the information with anyone they know who used AAR.
In the span of a week, I received only four responses. All said they had a negative experience with AAR. Two explained why:
“Access a ride is unreliable They can show up anywhere between hour to 2 hours late to pick you up at your address but you can’t be 5 minutes late to get on the ride when they arrive the dispatchers they have at the office do not know good customer service quality and they talk to patrons any kind of way they like they’re not very welcoming or helpful”
“I had a negative experience because one night I was going out with my friend and we was getting ready back home . However my friend and I had to wait in the cold for AAR for 30 minutes and I had to go to the bathroom. So AAR should not be taking that long to pick me and my friend.”
Although I often hear about time-management related complaints from the AAR customers I interview, it’s always helpful for me to read/hear how this issue directly affects AAR riders. These responses also provide quotes for future stories, along with sources I can call back to for additional information.
Three of the participants provided additional questions they wanted answered:
“Number one is who is mapping out the trips for the drive to get dispatched?”
“Why do we have to wait outside for AAR ??”
“My questions are why is the window so small to schedule transportation?”
These questions give me an example of what issues riders want solved. With GroundSource, I can connect to these question-askers and even include them in. Or I can reach out to them using GroundSource to answer their question once I’ve done the necessary reporting.
I did run into some issues using the GroundSource online interface. There were several times where my questions and prompts would not save. It was frustrating having to rewrite my survey each time it happened.
Overall, I thought GroundSource was excellent. It allows me to directly survey community members that are difficult to reach because of how inaccessible the city is. A source even told me they enjoyed how they could discuss these issues in a private, text message conversation.
The next time I create a survey, I plan on distributing it to a much wider audience. Sending the text-in information to online community groups, social media channels and through an extensive word of mouth campaign will hopefully generate a variety of responses and opinions that exemplify the AAR experience.
