Process Makes Perfect


NOTE: This is a blog post I wrote for One of the perks of my job is that I get a deep view into the creative operations process for our clients. We are intimately woven into the fabric of some of the largest fashion brands in the worlds and are granted access to their inner workings. Likewise, we work with potential clients and prospects to understand how we can fit in their process.

Because of this access, we can learn, study, and define best-practices across verticals and improve our client’s processes. Production headaches are not unique and their causes are more common than not. This puts us in as much as a consulting role as it puts us in the software support role.

Well, it turns out that the most successful clients have a documented process. There are clearly defined roles, artifacts, deliverables, and hand-offs between participants. A well- structured process forces your teams to operate predictably and efficiently.

I’ve spent a lot of time reading client notes, talking to clients, and trying to synthesize best-practice templates for our core industries. The one I want to talk about is an e-Commerce workflow.

E-commerce is about scale

Creative operations is more about operations than it is about creative. E-commerce typifies this by the sheer size of their needs. Let’s talk about what it takes to get a photo of a new winter coat for Stuckey Co’s website. There are 5 colors of this coat, and 5 different sizes per coat of each coat.

Most clients want photos of: Front, Back, and sides 3 on-model photos 2 close-ups

Now, to get the final image, let’s assume that there are 15 shots of each. That means that the client will need to manage: (4 angles x 15 shots) = 60 images (3 on-model x 15 shots) = 45 images (2 close-ups x 15 shots) = 30 images

= 135 images * 5 colors = 675 images (or >1,300 if JPEG + RAW)

675 images - assuming that they are only shooting one size. If this is going into a product life cycle management (PLM) system, they may do this for EACH unique SKU.

This means nearly 7,000 photos could be created FOR ONE COAT. What happens when they sell a dozen different variations? What if they want distinct photos of each size?. It should be apparent by now that this is a scale problem that quickly becomes intimidating.

But wait, there is more. Retailers will typically add SKU, season, color, and style information to each image’s metadata; they have to to simply keep track of everything. And the images have to be sorted, rated, and culled down to the few approved images. TAnd then they get to be retouched, have the retouching work approved, and then added to the DAM, PLM, and sent to Scene7 or their CMS. And how are you shipping these around? If you’re using a hard drive, how do you know you have the latest versions?

That’s a lot to manage for a single coat.

Here is a diagram that I’ve been working on to try and abstract the typical eC-commerce ecommerce workflow workflow.

We find that many of the clients in the most pain have some sort of an ad-hoc process that has grown organically and uses consumer tools. This certainly works to a degree but it has a distinct pain: Ooutput is unpredictable, highly variable, and the team members are stressed.

The solution is straightforward - but it requires some effort

There are two parts: The process and the tools.

Part 1: When we engage clients, we like to start with a diagrammodel like the one above. We customize it based on client feedback and try to name who is responsible for what, how they get their work done, and who is next in line. Many clients haven’t done this exercise and we find that even just getting people to spell out how they work has value. Inefficiencies stand out. People disagree. Sub-optimal choices become apparent.

Part 2: Our software is the other half of the solution. We found that a huge amount of time is wasted when people have to track down decisions. For example - was an image approved? By whom? We address this head on by keeping a running log and filtering tools so a team member can see “all of the new approved images that have been marked up” and need to be retouched.

Second, we keep all of the assets in a single location. We can intelligently manage JPEG, CR2, XMP stacks of images so that all of your assets are intelligently consolidated on one platform. This means that if you have an iPhone, you have access to the version of the image that was updated 5 minutes ago in a different state.

Lastly, we manage all of the meta-data. We can compile photographers to upload key information during their work (SKU, Style, etc.) and then give you the tools to bulk-edit these. This metadata lives with the assets and is exportable when you want to ship the finished assets to your DAM or PLM system.

So what’s in it for me? I’m biased, I’d love everyone to use our platform. The platform is only part of the solution. Just the act of documenting your workflow, asking WHY you operate like you do, and being open to refining the process is a huge win for most teams.

If you’re starting a new season and looking at the prospect of producing tens-of-thousands of new assets, taking a little bit of time to reset and reflect on your process could save your sanity.

Photo credit: Flickr user Mark Sebastian (used under CC)

EverythingBrian Stuckey