The Washington Post doubling size of its Visual Forensics team
This is the first expansion since the team was launched in the summer of 2020. The current team comprises four video reporters, one video graphics editor and an executive producer. The new roles open up six positions for two video reporters, a motion graphics reporter, an investigative researcher, a senior video producer and a designer.
Boosting video reporting
The aim behind this expansion is to strengthen The Post’s capacity to explore and experiment with video reporting to power both short and long term investigative stories told in a myriad of ways, across mediums. The team uses open-source, widely available materials, social media videos and photos, Google Maps, public databases and weather reports, or high-quality satellite images offered through paid subscriptions.
“This move will help the future of investigative storytelling. So much of what happens in the world is captured in video and photos, and we are expanding our capacity to acquire and provide context for what people are seeing,” Kat Downs-Mulder, Chief Product Officer and Deputy Managing Editor at The Post, told WAN-IFRA.
“With an expanded team, we will be able to work on multiple stories simultaneously and increase our ability to respond to breaking news.”
Video production cycle
The Post’s Visual Forensics department, lead by Senior Producer Nadine Ajaka, captures and provides context around the firehose of visuals available on the ground, on social media, and the internet, in general.
A typical video production cycle starts with geolocation, which involves the forensic journalists dissecting scenes pixel-by-pixel, identifying landmarks, silhouettes, and cross-referencing images using free tools such as Google Earth and satellite subscription services. The team might also compare multiple videos of the same incident to unlock more information.
The next step is establishing the time of the recording, which can prove to be an uphill task and is not always possible. Often, reporters might be able to access metadata – a digital fingerprint of sorts that can reveal the place and time something was filmed – but that is again, not always readily available or reliable.
The team focuses on two types of stories:
- Quick-turn-around visual analysis that lends context to breaking news
- Long form investigations that uncover previously unpublished video and break news
The Visual Forensics vertical designs its stories to work seamlessly across platforms, since its audience is predominantly smartphone users.
“We also try to bring this work to the broadest audience, made possible through summaries on Instagram and publishing on YouTube, where longer-format videos do very well,” Downs-Mulder said.
The team has a rigorous verification process that involves multiple steps, from geolocating video using forensic tools to matching landmarks to checking the video hasn’t been edited, manipulated or tampered with in any way.
Whenever possible, reporters scope the place where the story is breaking, knocking on doors to talk to people. The team has also created resources that can help people become better judges of the visuals they see.
Technology is an increasingly important component of the department’s investigations, used to authenticate and unpack the contents of the video in question.
“In our stories about the attack on the U.S. Capitol, we used facial recognition software to count the number of people in the videos and de-duplicate faces. We hope to expand the use of these tools beyond faces to identify specific insignias, gear, patches and articles of clothing from video,” said Downs-Mulder.
With the war in Ukraine, the department has been working round the clock to authenticate and provide context to videos that contribute to their reportage. The team has been aided by people across the newsroom, supplementing their efforts, and has catalogued and authenticated more than 700 videos so far.
Combating misinformation is a constant battle for the team since there cannot be a margin for error. “With a story like Ukraine, the sheer volume of incoming video is a significant challenge to vet,” said Downs-Mulder.
The Visual Forensics reportage is among the company’s highest trafficked and typically displays a strong conversion performance when compared with a standard text article as far as the subscription funnel and audience engagement are concerned.
The team’s aggressive response to breaking news is a significant audience-driver. Three of The Post’s top 25 converting stories in 2021 were from Visual Forensics.
To achieve this, the team has undergone extensive training in open-source techniques and other investigative reporting with experts in the newsroom, and globally. Now, as the team expands, it is looking at bringing some of the world’s top open-source reporters onboard.
Some of the most successful projects from the team include an examination of George Floyd’s final minutes, the crackdown of Lafayette Square before Trump’s photo op, and how federal agents swept up Portland protestors based on inaccurate or insufficient information.