- The Conveyor
- Posts
- How Our Place Saved $1M a Year Cutting Packaging Waste
How Our Place Saved $1M a Year Cutting Packaging Waste
Deep dive on fixing a packaging system that couldn’t scale
Gowtham Ramachandran
April 21, 2025
Our Place built its reputation on sleek, thoughtfully designed cookware - highlighted by the fan-favorite Always Pan.
But as their product line expanded beyond 30 SKUs, the fulfillment setup that had worked beautifully when they were smaller became a massive liability.
Initially, Our Place designed bespoke packaging for each product, perfectly sized and branded. But growth brought complexity. Customers started ordering multiple items, and the team only had two standard master boxes for multi-item orders.
Soon, inefficiencies piled up:
Multi-item orders often shipped in oversized boxes filled with air.
Single orders frequently arrived split across multiple shipments.
Packaging inefficiencies were tracking toward a costly $2 million annual problem.
That’s when they called in Paccurate - a software platform that helps ecommerce brands pack smarter using cartonization APIs and simulation tools.
Together, they overhauled Our Place’s packaging logic.
The result: Oversized shipments dropped from 20% to under 1%, and they saved over $1 million a year on shipping and materials.
I sat down with Darlene Yu (Dir. of Fulfillment & Logistics at Our Place) and Kosta Cherezov (Dir. of Customer Success at Paccurate) to unpack how they pulled it off — and what you can apply to your own fulfillment challenges.
Here are the top lessons from that conversation 👇
When packaging strategy stops scaling
When Our Place started, they had just a few SKUs. Each product had its own perfectly fit box, designed to look great on arrival. That made sense when the catalog was small.
But once customers started buying multiple items - say, a pan, a pot, and a set of bowls - the system broke. They only had two standard master boxes to consolidate multi-item orders.
The result: Five individual boxes shipped to one customer. Or a massive box packed with smaller ones. Or worse, a giant box filled with nothing but air.
“Customers were either getting gigantic boxes with little boxes inside, or they were getting five individual boxes at their door.”
That wasn’t just an aesthetic problem - it was driving up shipping costs and customer complaints. It was also leading to more stolen packages and missed deliveries, especially for apartment dwellers.
Darlene and her team knew they had to solve this.
But they also knew: the issue wasn’t just the number of boxes. It was the logic behind how products got packed in the first place.
Why Data Matters Most in Packaging Optimization
Here’s something most teams miss when they try to fix packaging: it’s not a tooling problem. It’s a data problem.
“Your dims will never be perfect, but they need to be close. A quarter-inch off can drastically impact shipping.”
Before Our Place could make any improvements, they had to clean up product data across the board. Even small errors were causing big issues.
For example, a product that’s listed as 8 inches tall instead of 8.2 inches might not sound like a big deal - but when you’re optimizing hundreds of thousands of shipments, that gap compounds fast.
To address this, the Our Place team:
Conducted a detailed audit of all product dimensions, weights, and naming conventions.
Implemented internal triggers to quickly catch and correct data inaccuracies.
Defined clear business rules, including maximum weight limits (e.g., 40 lbs per box) and item compatibility rules.
Without this foundational step, no amount of software automation could have succeeded.
“We created triggers that check our data now. If something’s off, we catch it within a day. Before, it would be at least a week before we’d notice a product was shipping in the wrong box.”
Implementing Smarter Packing Logic
Our Place partnered with Paccurate, using two complementary solutions to overhaul their packing process:
Real-Time Cartonization: Integrated with their WMS (Fulfil), this tool selects the optimal box in real-time based on order contents, available boxes, and packing rules. "It instantly finds the best fit, no guessing," notes Kosta Cherezov, Director of Customer Success at Paccurate.
Packaging Simulations: The team also leveraged PacSimulate, running detailed analyses on historical order data to continuously optimize their packaging strategy. These simulations help determine:
Which boxes were consistently underperforming.
Optimal new box sizes to introduce.
How product redesigns or dimension changes would affect fulfillment costs and efficiency.
“The API takes in product dims and weights, plus a list of available boxes. It finds the best fit. But simulations let us model thousands of orders at once.”
These tools worked together. Real-time logic improved daily operations. Simulations helped guide longer-term changes, like investing in new bundle boxes or adjusting the box assortment.
For example, when Our Place resized the Always Pan, they immediately understood the impact on their fulfillment costs before the new product even shipped.
Training Your Warehouse Team: From Tetris to GIFs
Technology and data alone weren't enough. Warehouse execution remained critical.
Initially, packers struggled to adapt to new packing configurations.
So Darlene created simple visual guides - GIFs - to show packers exactly how each order should be packed.
“Some people are just bad at Tetris. And you don’t want your most expensive orders packed by someone who’s guessing.”
Now, when a packer scans an order, a visual guide shows them how to arrange the products.
This sped up onboarding, reduced packing errors, and saved packers from having to guess. It also ensured that the strategy in the software actually made it to the physical world.
Results: From 20% Oversized Boxes to Under 1%
The improvements were dramatic:
Oversized shipments dropped from 20% to below 1%.
Annual packaging and shipping savings surpassed $1 million.
More accurate shipments reduced customer complaints and improved satisfaction.
Better visibility into when, where, and why things were going wrong.
Perhaps more importantly, the fulfillment team had a tighter feedback loop. If something shipped in the wrong box, they could trace it back immediately - to bad data, the wrong rule, or a new SKU that hadn’t been configured yet.
They also had a system that could evolve. Every time they launched a new product, they re-ran PacSimulations to check if it would affect the box mix.
“Customers were buying Always Pan + Perfect Pot together. So we built a bundle box at the origin. That alone saved on packaging and simplified fulfillment.”
Key Lessons for Your Own Packaging Optimization
If you're tackling similar fulfillment challenges, here’s what to prioritize:
Data Accuracy First: Audit your product dimensions, weights, and names rigorously. Set up automated checks to maintain accuracy.
Don't Rely Solely on Your 3PL's Boxing Logic: Basic cartonization logic at most 3PLs doesn't consider product-specific rules, complex stacking, or real-world packing constraints. Retaining control gives you better outcomes and transparency.
Simulations are Crucial: Use historical order data to simulate packing scenarios regularly. This helps you anticipate costs, identify waste, and proactively optimize.
Visual SOPs are Game-Changers: Don't rely solely on text-based instructions. Implement simple, clear visuals to ensure your warehouse team executes effectively.
Continuous Optimization: Packaging efficiency isn't a one-time fix. Every new SKU, product redesign, or seasonal change should trigger a re-evaluation. Our Place runs simulations quarterly to stay ahead of changes.
Packaging Optimization is an Ongoing Journey
The biggest takeaway from Our Place’s experience: Packaging is never fully "solved" - it evolves with your product line and customer behaviors.
New SKUs change things. Product redesigns change things. Seasonality changes things. That’s why the team re-runs their simulations every quarter.
“Packaging isn’t a one-time project. Every time we launch a product or a bundle, we rerun the model to see if we need to update our box mix.”
If you're facing packaging inefficiencies, start by fixing your data, understanding your fulfillment constraints, and creating a flexible process. Your warehouse, budget, and customers will thank you.
Q&A with Darlene Yu & Kosta Cherezov
If you want to go deeper, here’s the full conversation I had with Darlene and Kosta.
We get into the weeds on everything from fixing bad DIMS, to working with 3PLs, training packers with GIFs, and how to build a packaging system that keeps improving over time.
Table of Contents
This conversation has been edited for length and clarity.
1. Background: How Our Place Used to Pack Orders
Darlene, what was the packaging process like at Our Place before using Paccurate?
Darlene Yu: Before Paccurate, we were shipping most of our products as SIOC—Ship in Own Container. Since we launched during COVID, our assortment was small, and this worked well. We had the Always Pan and a few complementary items, like the steamer basket.
Each item had a custom box, and we were extremely thoughtful about packaging—everything recyclable, no plastic, and no wasted space. We even have someone internally, who we call the "packaging queen," dedicated solely to packaging design.
At that point, all orders were shipped individually, and since it was a small assortment, we only needed a couple of box sizes.
2. The Challenge: Growing SKU Count and Inefficient Packaging
So what changed that led you to look for an optimization tool like Paccurate?
Darlene Yu: The challenge came when our assortment grew—from a few products to over 30. All of them were still designed as SIOC, meaning they were already bulky. But our average order was two to two-and-a-half items. We only had two outer boxes—a medium and a large. That meant customers were either receiving enormous boxes filled with smaller boxes or getting multiple shipments.
This was both inefficient and expensive. From a customer perspective, it's also a bad experience - especially for apartment dwellers, who risked missing or losing packages.
From a business side, we were literally shipping air. We needed a solution that could optimize packing across multiple items in an order.
3. Why Paccurate: Choosing a Cartonization Partner
If you already had tight control over packaging design, what made you reach out to Paccurate?
Darlene Yu: While we had excellent individual packaging, we didn’t have a solution for optimizing multiple items per order.
That’s where Paccurate came in.
We needed something dynamic that could analyze the best way to pack multiple items into a single or few boxes—and we needed that to be scalable.
Kosta, what stood out to you about Our Place's use case when you first started working with them?
Kosta Cherezov: They had relatively few SKUs - 30 to 40 - but they’re large and uniquely shaped. Plus, Our Place was in the middle of migrating their ERP to Fulfil, so we had to ensure a smooth integration across systems. Our role was to plug into that larger initiative and stay aligned as their fulfillment process evolved.
Volume utilization was a key focus - how do we use the smallest number of boxes and ship as little air as possible while maintaining a great unboxing experience?
4. Implementation: Data, 3PLs, and Workflows
What kind of data does Paccurate need before you can get started?
Kosta Cherezov: At its most basic, we need:
Item dimensions (LxWxH), weight, and quantity.
A list of available boxes or containers.
Then we can add rules: operational constraints, stacking preferences, carrier rates (for cost optimization), and more. The richer the rules, the better we can simulate real-life packing decisions.
What about 3PLs? How much do you need to know about them?
Kosta Cherezov: It’s essential. Each 3PL may have different box assortments or SOPs. We aim to mirror what packers are doing on the floor. So we want to replicate their operational constraints exactly in our API. Whether it’s direct shipping or through a 3PL, we need to understand how fulfillment happens on the ground.
Darlene, how did communication with your 3PLs go during this implementation?
Darlene Yu: It was definitely a learning process. Our first 3PL lacked the technical maturity, so we had to send manual shipping instructions - a pain. With ShipBob, it’s much better. They even have their own boxing algorithm, but we still rely on Paccurate for 99.9% of our fulfillment, especially for large, bulky items.
ShipBob’s system reads our instructions and double-checks with its own algorithm. If they suggest a smaller box that’s not in our assortment, we use theirs. So Paccurate acts as our insurance policy - and we’re able to catch errors and improve our internal dataset in near real-time.
Did you have to do any manual work to help 3PLs understand the new packaging process?
Darlene Yu: Oh yeah, it wasn’t just plug and play. I had to create 30+ GIFs showing how to pack specific orders - like visual Tetris. Packers are incentivized by speed, so if a layout isn’t intuitive, it slows them down. The visuals helped a ton. Now it’s much smoother.
5. The Tools: Paccurate API & PacSimulate
Kosta, can you walk me through the difference between Paccurate API and PacSimulate?
Kosta Cherezov: Paccurate API is our real-time packing engine. It takes a list of items and available boxes, and returns the best way to pack the order. It’s lightning fast—about 2 milliseconds response time. Perfect for active order fulfillment.
PacSimulate, on the other hand, is a bulk analysis tool. You can load thousands of past orders and test different scenarios—like what if we used a different box set? What if we shrunk the Always Pan dimensions by an inch? What if we changed our carrier mix?
It’s used to design strategy, whereas the API powers daily operations.
6. ROI: Measuring Cost and Space Savings
Let’s talk ROI. What were the savings like?
Darlene Yu: PacSimulate was life-changing. When we first ran it, we realized that 20% of our parcels were shipping in unnecessarily large boxes. With over a million shipments a year, that meant nearly 200,000 boxes were bigger than they needed to be.
Using incorrect carton sizes at just one of our warehouses for one week could result in a $10,000 loss. If this happened across all of our warehouses for a year, losses could exceed $2 million,
After running regular simulations and adjusting our box assortment based on the data, we brought that number down to under 1%. We now run Pac Simulate at least annually - sometimes quarterly, especially after new product launches. The boxes we use today are completely different from what we started with.
Do you track sustainability metrics too?
Darlene Yu: Not in hard numbers yet. But intuitively, using smaller boxes means less air shipped, fewer trucks on the road, and lower emissions. We’ve also started designing “bundle boxes” at origin, so if someone buys, say, an Always Pan + Perfect Pot, it ships in one optimized box straight from the factory. That cuts down on packaging waste and transit volume.
Kosta Cherezov: And it’s not just about space savings - it’s also about giving brands more control. A lot of 3PLs have basic boxing algorithms that treat all products the same. With PacSimulate, you’re using real data to make smarter decisions about packaging and cost - and we’re seeing more brands realize how powerful that can be.
7. Sustainability and Bundling Strategies
Can you expand on that bundling strategy?
Darlene Yu: Yes! Our data from PacSimulate showed us the most common product combinations. That insight led us to create custom “bundle” boxes that are pre-packed at the factory. Not only does that improve efficiency, it enhances the unboxing experience for the customer and cuts down on transit waste.
We also use PacSimulate to test new box assortments in advance—before launching new product lines. And because of the cost and sustainability trade-offs, we cap our in-use box sizes at five.
8. Biggest Lessons and Takeaways
What lessons would you share with other teams thinking about implementing Paccurate?
Darlene Yu: Don’t assume it’s turnkey. The tech is ready out of the box, but the data work is yours. Dimensions, weights, even product names—if they’re wrong, the whole system breaks. It took me three full weeks to clean and validate our initial data set. But now it’s on autopilot.
Also, we’ve learned to set business rules like weight limits (we cap at 40 lbs per box) and dimensional limits (to ensure customers can actually carry the box). These constraints can be encoded into Paccurate so that packing stays human-friendly.
Kosta Cherezov: Our best implementations come from partnerships—where the client is invested in accurate data, clear rules, and collaboration. Darlene’s team did an amazing job. And we continue to work together, especially as their product catalog evolves.
Reply