Why Operational Design is the Secret Sauce of Personalization.
If you were to ask us about the organizations that inspired our early days of School of One, you’d hear about pioneering educational organizations, transformative technology companies, and… Whole Foods. Not because of their organizational values or their fresh produce or even their abundance of granola options, but rather for their check out line.
At a neighborhood Whole Foods store, customers get in one of four lines and patiently (for New Yorkers at least) wait until an automated female voice and corresponding screen instructs shoppers on which register we should go to. All day she carries on without missing a beat. Register 12. Register 4. Register 3. Slight changes in her intonation often make us forget that it’s a computer.
What we love about the checkout line is that it reflects a concept that we think about as operational design. Whole Foods didn’t simply set out checkout lines like all groceries do; they put real thought into exploring whether there was a better way of checking out shoppers. And it’s faster! After all, what’s more maddening than getting in what you think is the shortest checkout line, only to find out that the person in front of you needs one-too-many price checks or is digging through her purse for the latest coupon?
You might typically associate Whole Food with high quality food and might not think about the choreography happening behind the scenes to make it all work. Similarly, when it comes to personalized learning, you might think about technology first. But there’s a whole world of operational design that brings it all to life.
Take homework, for instance.
In a world where each student is receiving personalized learning each day—either from a teacher, online, or in a collaborative setting—how does each student receive homework assignments that are aligned to the particular skill that he or she is working on that day? How does that assignment get selected? Distributed? Returned? Graded? Entered into the grading system? Factored into a student’s overall grade? Homework is a good example of our willingness to try out an idea, observe how students and teachers respond, listen to feedback, and iterate based on what we learn.
Back in 2013, we put in over 500 hours into designing a homework solution—and that was the third time we had done it.
Our first attempt was to create homework customized to what each individual student was studying that day. This meant that schools might send out 60-100 unique homework assignments on any given day. Grading unique homework for each of their students created a logistical challenge that proved unsustainable.
To solve for the administrative burden that unique homework assignments had created, we next tried connecting homework to multi-day projects that students worked on with the same teacher. This meant teachers would be grading the same assignment for each student rather than dozens of unique assignments.
While this approach had solved the logistical challenge of grading unique assignments and meant homework was connected to students’ individual learning, we began to hear from teachers that it didn’t provide them with a consistent touch point with their students on the instruction they received each day. Teachers wanted homework to more closely reflect all that students learned, not just what they learned as a group.
Our third and most recent iteration of homework provides customized packets of homework assignments to students at the beginning of the week and to be collected at the end of the week. We’re still reviewing this solution and gathering feedback. There’s still more for us to do to improve homework in Teach to One, but we feel hopeful that we’re getting closer to a solution that both meets logistical challenges and supports student learning.
We’ve been maniacal about the details on homework design, as well as the operational designs of every other part of our model. The same is true when it comes to how grading works, for substitute teacher plans, and for traffic flow (to avoid dropped laptops). We’ve tested dozens of chairs to find those least likely to squeak, the best manipulatives for a given skill and lesson, and the best use of a school’s learning space. (We fully accept that we are edu-geeks!)
That’s not to say there’s only one way of doing each of these things. Each school provides its own twist. But it’s much easier for schools to customize thoughtful options that have already been developed than to come up with them from scratch.
If you are a district or school that is beginning to think hard about blended or personalized learning models, push hard on operational design. Otherwise you’ll be making your teachers’ already difficult job even harder.