News

How the Financial Times streamlined its image workflow

As a leading global news organisation, the Financial Times gets more than 10,000 images each day.

by Elizabeth Shilpa elizabeth.shilpa@wan-ifra.org | July 23, 2020

Hundreds of photos are published daily by the organisation, which has an archive of some 1.5 million pictures. This meant that the company needed to have a robust image workflow system.

Their initial system had some hiccups, including a major limitation in that it was desktop-based and not many people had access to it. Added to that was the difficulty in understanding where the pictures were coming from as the FT had an open FTP.

There were many instances where pictures could not be used because they didn’t come with a certain amount of metadata, which left the company unclear about the kind of rights it had over the image. This prompted the FT to switch to FotoWare, an image workflow system that runs via browser and cloud, which was developed at Afterposten in Norway.

The implementation of FotoWare meant that more people could log in through the browser allowing the company to configure and manage complex metadata, which showed them the rights they had over pictures.

Implementation

“During implementation we put in all the different parts of FotoWare – Foto Web, which is the front end, Index Manager, which is the part doing the indexing in the background, Colour Factory, which allows you to do the processing of pictures and Connect, which allows you to bring the pictures from SFTP or elsewhere and put them into the system,” said Andrew Watkins, senior developer in the CMS team at the FT, to participants at WAN-IFRA’s recent World Media Leaders eSummit.

At the FT, while FotoWare was mainly being used for production, the team still needed to know where the images came from, what rights they had, and then be able to integrate that with the FT’s online and print CMSs.

With large agencies such as Reuters, AP, Bloomberg or API, finding out where the pictures came from was easier. They sent it via dedicated software and also put in information that gave the team information about where they were coming from and what they could be used for. But that was not the case with the images that came in from smaller agencies and individual photographers.

“Those pics were more complicated because there wasn’t much information and we didn’t necessarily know what rights we had or even if we had a contract with them. So we didn’t use them because we didn’t know whether or not we were allowed to use them. We spent a long time to find out as much as we could about where these pictures were coming from. Then we contacted probably 40 of the largest users and left the long tail of other people off the list,” – Andrew Watkins

“A new SFTP was put in place which was online with username and password access so that people could not send pictures if we didn’t know who they were,” he added. That was the first step indeed to identify the picture sources and to match them up with contracts and rights.


For more takeaways and analysis from the World Media Leaders eSummit, download our easy-to-digest slide deck report (free for WAN-IFRA Members and available for purchase for non-members)


Adding rights

Next the pictures were sent to the backend into Connect and Colour Factory where all the processing of metadata gets done. Once the pictures were coming in, the important thing was adding the rights on to the images.

“We had about 50 different contracts, but they were all different and it was very very difficult to know what rights we had,” Watkins said. It was quite difficult to find commonalities and rules for processing, but the team tried to look at what was in common in these contracts and also tried to work out what are the rights that the FT was interested in, for instance, if the image cost money, if it could be used more than once, if it had a time limit and so on. Then sources with similar rights were grouped together. The team set up feeds and dedicated processing channels according to the rules that were decided on.

FT-specific usage rights were defined next. “We went through the list of agencies and matched them up with those rights and categorised them into groups like ‘stock agency single use’,” Watkins said.

Given that thousands of pictures were coming in every day all this had to be automated as rights of each photo could not be worked out manually. The FT also wanted to put its entire image archive through this process.

So the team worked on the backend of FotoWare and set up its own specific elements that allowed them to map the things in their spreadsheets onto processing channels in FotoWare. Some pre-processing of metadata had to be done because all agencies don’t use metadata in the same way, necessitating some standardisation.

Making rights visible

Once the metadata was attached to the photos, the team worked on how to make the usage rights visible. The pictures were divided into different sets of archives, filters were introduced and usage rights of photos were displayed through icons – for example payable ones carried a pound sign. Every picture also had a more detailed list of rights if the users wished to get down to that.

Currently, photos that are to be used in print can be sent to Méthode, the print CMS, from FotoWare. The digital CMS has an integration that allows people who are inside the CMS to see the same pictures that they would see if they were inside FotoWare.

“That is the basis of putting together a digital story and a print story using a picture from FotoWare, all done in a browser. We are very happy that we were able to be in this situation before we were all suddenly working from home,” Watkins said.

To rewatch sessions from the World Media Leaders eSummit, please fill out this form.