Anatomy of an Efficient Process – How We Pack Orders

I mentioned in my last post that I was going to come back and delve a little deeper into our new packing area.

Pure Adapt Shipping Area

The process outlined below is a combination of George’s prior e-commerce experience, my industrial engineering experience, Greg’s overall business experience, and the work that Mike has done in studying how our competition and other e-commerce sites ship orders (specifically this great newegg video), along with a whole lot of trial and error.  It allows us to process orders faster, with less mistakes, and with less variation.   As any process should be, it’s been designed explicitly with future employees in mind.  Not only is it efficient for us, but it will be very easy to learn and the built-in checks and balances make it very hard to mess up.

The general “flow” of packages follows the red arrows in the diagram below, flowing from the puller’s table to a packing area, and eventually out the garage door into a backed-up FedEx truck. (I should note that prior to this, we just had a few packing tables without much organization or flow.  It worked, but as we’ve grown the need for a well structured system became readily apparent).

Pure Adapt Shipping Area

The system is built to work with anywhere from 2-5 people.  I’m going to describe a Monday morning when all four of us are working.  That’s when we’re the busiest and therefore that’s when we see the biggest impact of having a system in place like this.

  • 2 pullers – these people are responsible for physically pulling the orders from the shelves and then for picking a box-size for the order (now automated with our new box-size system).  For small orders, they just make the box and place the items directly in the box.  For larger orders they use one of the bins and put the box underneath the bins.  Once they’ve pulled an order, they double check it and place it on the packing table next to the #1 in the image below, sliding any existing orders down the table in the direction of the arrow.
  • 1 packer/inspector – this person stands behind the tables, near the #2 in the photo above.  If the box isn’t already made, they first make the box.  If the order was in a bin, they put the bin back.  They then take the order, double check that it was pulled correctly, and then arrange it in the box in the safest manner.  They tighten tops, wipe off dust, and do anything else to ensure that the order is ready to ship.  They separate the invoice from the shipping label, place our promotional cards (located in the black organizer) in the invoice, fold the invoice up and place it in the box, and finally pass the order to the left, again following the arrows.
  • 1 shipper – this person also stands behind the tables, near the #3 in the photo above.  They fill the box to the brim with packing peanuts, seal the box, and push it forward onto the double tables.  When FedEx shows up, they back their truck right up to the door and it’s only a walk of a few feet to put the boxes on the truck.

The peanut dispenser is what really tied the whole thing together for us.  Previously we used all sorts of packing material, from bubble wrap to re-used material from our vendors and pretty much anything else you can could imagine.  Not only did our customers receive packages with different boxes and packing material each time they ordered, but more importantly the quality of the packing job was left up to the discretion of that individual packer.  Now we simply have a rule that every single order gets these strong anti-static biodegradable packing peanuts filled to the top, no exceptions.  Every order is presumably packed just as well as an order shipped out by someone else.

We keep a shared spreadsheet with any shipping mistakes or damaged items.  Since implementing this a few months ago, we’ve dropped off to almost zero, which I take as a testament to how good of a system we built.  The real test will be when we plop a few employees in.  If McDonald’s can train any zit-faced 15 year old in the world to make the  exact same Big Mac, we think we can infuse any competent employee into this system without a drop off in results.  Only time will tell, but for now we couldn’t be happier with what we’ve built.

7 comments on Anatomy of an Efficient Process – How We Pack Orders

  1. I was impressed by your “box-size system”. The algo you designed is simple and seems efficient. I can see how it could save you guys tons of time.

  2. Adam McFarland says:

    Thanks Phil. It’s actually been interesting to track the accuracy. We made a few adjustments yesterday because it was consistently picking ~25% of the boxes a little bit too small. Obviously our formula doesn’t take into account how each item is going to be placed into the package. Total volume would assume you had a clay ball of that total volume and could mold it however you wish, which isn’t the case with products of finite dimensions. After quite a bit of banter back and forth, we decided the simplest way would to just add a percentage to the volume so that many close orders would get kicked up a box size. We started with 10% today and had 100% accuracy, although orders were a bit slow since everyone is waiting for Black Friday sales. We’ll see next week for sure, but we are working in the right direction.

  3. Gordon says:

    Roughly what percentage of orders are wrong on the first go-through? (is it worth it to have someone checking?)

  4. Adam McFarland says:

    Hi Gordon –

    Thanks for the comment. Not exactly sure what you’re asking? We have the puller double check the order before the second person “officially” checks it. Are you asking what percentage are found to be wrong at that point? I’d say about 1%, although we aren’t tracking that currently.

    We obviously do track how many get shipped out wrong, and that number is very low. I checked the spreadsheet and we’ve had 20 in six months, which is far far less than 1% of orders. Assuming that number stays essentially zero we probably won’t start tracking anything more than what we currently do.

    Happy Thanksgiving!

  5. Adam McFarland says:

    Thanks for the link Rob 🙂

  6. […] the most frequently sold brands closest to our pulling table (this post from last year has a good overview of how we pull/pack […]

Comments are closed for this post.