Systems Thinking for Online Sellers

Everything in your business is connected. Everything that happens has an effect on at least one other thing, positively or negatively.

For example, if you invest in fast order processing then your customers will get their orders more quickly. They’ll be pleased that they bought from you, and more likely to buy from you again. They’ll also be more likely to give you good seller feedback, which helps convince others to buy from you.

That’s the essence of Systems Thinking – understanding how your business works as a whole. In this post I’ll explain how to use Systems Thinking in your business to help uncover problems and find new opportunities.

A balancing loop

An online retailer, who I’ll call Seller X, has a problem. It’s holding their business back. They think through how they operate and draw up a diagram of their ecommerce fulfillment system:

Breaking it down, this says for Seller X:

1. When dispatch speed increases, so does customer satisfaction – buyers like fast delivery.
2. When customer service speed increases, so does customer satisfaction – buyers like being answered quickly.
3. When customer satisfaction increases, so does order volume – happy buyers buy again.
4. When order volume increases, dispatch speed and customer service speed decrease.

Point four exposes a very simple fact for Seller X – when they receive a lot of orders they get overwhelmed and can’t keep up! Dispatch and customer service speed suffer, and as a result they have more dissatisfied customers.

As a marketplace seller trading through eBay and Amazon, poor customer satisfaction rates can be disastrous. Seller X could fall in the search results and even be banned from trading if they get a lot of customer complaints, ship slowly, or are slow responding to emails. This limits their growth, because for them growth itself is a double edged sword: it leads to higher revenue but poorer customer experience.

Now Seller X feels locked in their situation. They would like to grow, but the negative effect of higher order volumes makes them hold back. They don’t compete as hard as they could, but keep their customers happy and the business afloat. In Systems Thinking this is called a balancing loop – the system finds a level which keeps it stable and it can’t change from that.

The perils of efficiency

So what’s the problem for Seller X? It’s tempting to question their efficiency, but the funny thing is that efficiency is something of an obsession for them. They listened to advice, and streamlined each part of the shipping process right from the start.

Here are the main steps in their process:

1. Every hour a supervisor prints a pick list of products for all the new orders.
2. The list is handed to a “picker” who collects all the products on the list.
3. At the packing area a “packer” packs the orders into boxes.
4. A porter takes the boxes to the dispatch area.
5. Carriers arrive twice a day to take the boxes away for delivery.

This process ticks several boxes for efficiency: workers specialize on one task, they aren’t constantly running between areas, and they process multiple orders at the same time. It sounds OK, but here’s what happens to the system:

I’ll break it down again:

1. When order volume increases, so does the pick list size – it’s printed every hour.
2. When the pick list size increases, packing speed decreases.
3. When packing speed decreases, so does the number of packages shipped – they just aren’t ready when the carriers come to collect.
4. When the number of packages shipped decreases, so does order volume – buyers don’t like slow delivery, as we saw in the first diagram.

It’s clear now that long pick lists are a problem, but why? Well, the pickers have more products to find, so more time passes between their drop-offs at the packing area. As a result, the flow of products to the packers becomes “lumpy” – they get nothing for a while then a large batch.

So the packing area swings between quiet times waiting for products, and being overwhelmed by huge drop-offs. Their speed suffers, carrier collections are undersized or missed completely, and they work overtime to clear the backlog. But it’s too late – the last collection has already been made, and a very large one is needed to catch up the next morning. And all this from an “efficient” system!

Don’t blame the people, change the system

Who’s at fault at Seller X? Traditional views of efficiency might blame the pickers – they should pick faster or bring the products to the packers more often. But Systems Thinking recognizes that most of an organization’s performance is due to the system, not the people (perhaps as much as 95%).

If pick lists are printed every hour and pickers are told to finish the list before dropping-off, then that’s what they’ll do. Even if they are told not to, they are steered that way just by having a list – you don’t go shopping and return home three times partway through your shopping list. Whether it’s intended or not, the system is governing their behavior.

When a system exists, even if it’s not perfect, it feels uncomfortable to start changing it. What should you change? How do you know that the change will be for the better? The simple answer is that you don’t, and you might not even be the right person to figure it out. The pickers, packers, and other workers are much closer to what happens on the ground. They are better placed to think up ideas, and consider what the subtle effects might be. Let them loose on the problem, and be creative with their ideas.

Here are some of the possibilities to address the pick-list problem at Seller X:

• Hire temporary staff for busy periods.
• Print the pick list more frequently than every hour.
• Print the pick list after ten new orders have come in.
• Print the pick list when twenty products need to be picked, whether that’s for one order or twenty.
• Get rid of paper pick lists and give the pickers smartphones that tell them what to pick and when the packing area needs a drop-off.
• Get rid of pickers and have the same person see each order all the way through the process.

There’s no way to know exactly what will work, so prioritize the ideas and try them out. Given a clear understanding of what you are trying to achieve, you can even let the workers change the system themselves. Either way, make sure you can measure how the system is performing so you can see the effect of your tests.

A reinforcing loop

We’ve seen how System Thinking can help draw out problems, but it can also help identify opportunities for growth. Imagine if the first diagram looked just a little different:

In this scenario, order volume increases but dispatch speed and customer service speed do not suffer – they actually increase too. That’s called a reinforcing loop, and in a reinforcing loop, working on any part of the system causes the whole business to grow along with it. In this example, faster dispatch speed increases customer satisfaction, which increases order volume, which increases dispatch speed – and the whole cycle repeats.

So if you want your business to grow, look for a way to create a reinforcing loop. In the scenario above, you might use the revenue from higher order volumes to continuously invest in dispatch and customer service processes, people and technology. That would add another step to the loop (investment) and help turn the relationships from negative to positive.

There’s another example in Brad Stone’s book about Amazon, The Everything Store. In the book, Stone describes a “flywheel, or self-reinforcing loop… which they believed powered their business.” It sounds counter-intuitive, but Amazon’s reinforcing loop means that lowering prices grows their business. It looks like this:

Breaking it down for the final time:

1. When prices are lowered, more customers visit the site.
2. When more customers visit the site, order volume increases.
3. When order volume increases, fixed fulfillment costs per order decrease.
4. Also when more customers visit the site, more third-party sellers are attracted to the site.
5. When more third-party sellers are attracted to the site, fixed IT costs per order decrease.
6. When fixed costs decrease the savings can be reinvested in lower prices.

A reinforcing loop may not be sustainable forever. But if you can create one, feed it for as long as possible and watch your business grow and grow.

In closing

In this post I’ve tried to make System Thinking accessible – and useful – for online sellers of all sizes. What do you think? Has it helped you reflect on your business, or completely missed the mark? Is there more you need to know? Whatever you think, I’d love to hear from you in the comments below!