An ultimately great experience with the Theme Garden, and scaling customer service
Last week I submitted my theme to the Tumblr Theme Garden. The process was very frustrating because I received no feedback in the Dashboard or via e-mail, and had to visit the Theme Garden and discover a new “Manage your themes” link to learn that my theme was rejected for failure to meet Tumblr’s guidelines.
No specific guidelines were called out — I was simply presented with a generic list of Tumblr’s guidelines and was expected to figure out what I needed to do better. One option was to try to more closely adhere to the guidelines, resubmit and cross my fingers… and check the Theme Garden management page once a day or every few days to see if I’d been accepted or rejected again… and then repeat that process over and over again. Another option was to give up. A third option was to complain.
I chose to complain. I vented my frustration and Peter Vidani, Tumblr’s lead designer, heard my complaint. Within a few hours I had an e-mail from Peter explaining why my theme was rejected and offering advice on how to fix it.
I took Peter’s advice, resubmitted my theme, and it was approved within minutes.
So I’d like to thank Peter for his help. Peter was incredibly helpful and very nice. Peter turned a very frustrating experience into a delightful experience by taking time out of his day to offer personal service. His design work is excellent, and the fact that he took the time to offer this support says a lot about Tumblr as a company — all of it good.
But there’s a flip side, too. Companies like Tumblr (and companies like blip) need to figure out how to scale customer service as our user bases scale. Personal intervention like Peter offered me is awesome and leads to an excellent experience (we try do it at blip all the time for this reason) but it doesn’t scale well.
At blip we have a similar problem. For example, show producers have to apply for distribution to some of our partners (such as TiVo, Verizon FiOS, Sony TVs and even YouTube). Our content team has an amazingly long list of shows to approve and disapprove, just as I’m sure Tumblr has a very long list of themes to approve and disapprove. As a result of the incredibly long waiting list that we have we tend to have a similar problem as Tumblr seems to have — we just approve or reject shows for specific distribution outlets. If we reject people they get an e-mail with a link to a generic list of things that they can do better, and they’re offered an opportunity to reapply for distribution. This experience is better than Tumblr’s in that we send an e-mail, but it’s still not good enough.
One way we’re trying to improve the application experience at blip is to shrink the queue. We’re doing this by making it harder to apply for specific distribution outlets… instead of simply listing a bunch of prerequisites for distribution we’re programmatically checking for them. You aren’t allowed to apply for distribution to Sony televisions, for example, unless you have a certain minimum number of episodes under your belt, your episodes are available in HD and they’re in the right format for Sony TVs. This approach should start to shrink the queue while offering more meaningful feedback and a better experience for producers — they’re more likely to be approved when they apply.
Once we shrink the queue we’ll be able to start offering more constructive feedback to producers. We’ll try to do this in a way that allows our content team to move quickly by using canned rejection messages (hopefully more specific than a list of requirements) while at the same time allowing them to write longer and more specifically useful feedback if they want to or feel that they need to. Hopefully this will help. I imagine that Tumblr could probably adopt a similar approach if it’s important to the company (which it may not be).
Scaling customer service is really, really hard. We have an internal rule that every e-mail sent to support AT blip is returned within 24 hours. That’s hard. But it’s really important. At blip we’re really lucky that our producer base is relatively small — we have a very high ratio of producers to viewers. I can’t imagine facing the challenge Tumblr faces — offering an excellent service experience at scale to so many people. But it’s one of the most important things that we can do as our services grow up. I’d love to hear if anyone has ideas for how to do this well. We all face the same challenge, and we’re all trying to deal with it.
