String — Operational Automation Engineering
For operations whose shipping rules have become specific to the business — beyond what any horizontal platform is built to express.
We build and maintain the shipping logic your operation actually runs on — cartonization, rate selection, batching, warehouse routing, inventory and temperature rules. You tell us how your warehouse operates. We implement it, maintain it, and improve it as you change.
A working review of your operation. No cost, no obligation.
Operations we build and maintain automation for
Not on day one. But reliably, as the operation matures.
Products change. Packaging changes. Carrier contracts get renegotiated. Temperature requirements appear. Inventory moves between locations. A channel is added, a warehouse is opened, a 3PL takes half the volume. Each change is reasonable on its own. Together they push your shipping logic somewhere no general-purpose tool was designed to follow.
This is not a defect in ShipStation. ShipStation serves a very large and varied merchant base, and it has to work for all of them. It cannot reasonably build the temperature threshold that governs your coolant spec, the batching sequence that only makes sense given your pick paths, or the joint cartonization-and-rating decision your rate card actually rewards. No horizontal platform can. That is the honest limit of generic automation, and it is where String begins.
At that point most teams are choosing between three options. Absorb it as permanent manual work, and pay for it in labour, errors and your best people's attention. Rip out the stack for a full warehouse management system: tens of thousands of dollars and months of retraining, to solve what is really a handful of decisions. Or hand fulfillment to a 3PL: the right answer when the problem is running a warehouse at all, an expensive one when it is a few dozen decisions inside a warehouse that already works.
String is the fourth, and the only one that leaves the operation where it is. We are a small engineering firm that does one thing — we build, deploy and maintain the operational automation that runs inside ShipStation accounts. There is no dashboard, no seat to license, nothing for your team to learn, and no rules for you to configure. You tell us how your warehouse operates. Our customers don't log into String. That is the point.
The software is not the product. What we actually sell is the accumulated knowledge of how these operations fail, and how they should be run. The automation is simply how that knowledge gets executed, on every order, without anyone having to remember it.
"There wasn't an existing interface that could handle that workflow. That's where String stepped in. They didn't just offer a workaround, they built a custom-coded solution from the ground up that is now effectively a turnkey integration."
These are the eight patterns we find most often in operations that have become operationally specific — multiple carriers, real packaging constraints, exceptions handled by judgement. Volume correlates with them, and most of these operations ship 6,000+ a month, but complexity is what actually produces them.
None of them show up as a line item. They show up as labour, exceptions, and a shipping cost per order that drifts upward while volume grows.
Carrier flat-rate programs only pay off if the order is packed into that carrier's specific box. If cartonization runs first and rate selection runs second, the cheaper option was already made impossible before rating ever happened. The two decisions have to be solved together.
The tell — Flat-rate volume is near zero, or suspiciously constant month over month.
Native rate comparison handles the clean middle of your order profile well. The money is in the exceptions — zone bands where a slower service arrives the same day, weight breaks that flip the winner, negotiated rates that only beat list past a threshold.
The tell — Two orders to the same zone, same weight, ship on different services with no reason.
Dimensional weight means an under-filled box is billed as if it were full. Box selection driven by what the packer reaches for, rather than by the actual cube of the items, quietly converts every oversized carton into a surcharge.
The tell — Your billed weight consistently exceeds your actual weight across a month.
When an order won't fit one box, the cost-optimal split depends on your packaging inventory and your rate card at the same time. A packer making that call under time pressure is making a pricing decision without the pricing.
The tell — Multi-package shipments cost more per unit than the same items shipped separately.
Real batching is a function of zone, cutoff, service level, pick path and priority simultaneously. Systems that batch on a single attribute push the remaining logic onto whoever builds batches each morning.
The tell — Someone spends the first hour of every shift assembling batches manually.
Whether an order needs an insulated box, more coolant, or a faster service is a function of the forecast at the destination on the delivery date — not a static rule set at the SKU level. Static rules are wrong in both directions: spoilage in summer, wasted dry ice in winter.
The tell — Your cold-chain packaging spec is identical in February and July.
Without a FIFO-aware hold, orders that stock can't fully cover still release, and get picked, and get short-shipped. The cost lands downstream as split shipments, re-picks and support tickets rather than in the fulfillment line item.
The tell — Split shipments and cancellations cluster right after a popular SKU sells through.
Every operation has an operator who just knows — which SKUs can't go ground, which addresses need a signature, which customer gets held for Monday. That knowledge is real and valuable, and it is also a single point of failure that takes vacations.
The tell — Error rates change measurably depending on who is working that day.
If four or more of these are familiar, there is money on the floor of your operation. The Assessment is how we find out how much.
Orders arrive in your ShipStation account exactly the way they do today. Nothing is rerouted and nothing passes through us first. String reads each order, makes the decisions, and writes the finished shipment back — before your warehouse ever sees it.
String — decision layer
Built & maintained by String
Reads
Decides
reads the order · writes back the shipment
Source
Shopify & other channels
Your system of record
ShipStation
Unchanged. Your team's only screen.
Execution
Your warehouse
Orders are never rerouted. If String were switched off tomorrow, your operation would keep running — it would simply go back to deciding by hand.
1
No code for your team to write, no keys to manage, no new dashboard. Connection takes minutes; the work that follows is ours.
2
We map your SKUs, packaging, carrier agreements and the exceptions your team currently handles by hand — then build the logic that makes those decisions the same way every time.
3
By the time an order appears on the floor, the box, the service, the carrier and the batch have already been decided. Nobody has to think about it.
4
New SKUs, new packaging, a renegotiated rate card, a rate increase, a new warehouse. The logic is maintained by us — it does not decay into another thing you own.
String is a poor fit for most companies, and a very good fit for a specific kind of operation. It is worth knowing which one you are before you commission an Assessment.
Roughly 6,000+ orders a month
Not because volume is the qualifier, but because it is where operational specificity usually shows up.
A dedicated fulfillment team
People whose actual job is getting orders out the door, rather than a founder doing it between other work.
Procedures that rely on tribal knowledge
The rules that live in one operator's head and change depending on who is working that day.
Complexity beyond standard automation
Temperature thresholds, cartonization, warehouse routing, ERP data, inventory-aware decisions, carrier-specific rules.
Shipping mistakes carry real financial weight
A wrong box, a wrong service or a short shipment costs enough that consistency is worth engineering for.
A few hundred orders a month
At that volume the arithmetic rarely works in your favour, and we would tell you so on the call.
Standard automation already covers it
If ShipStation's native rules express your logic today, use them. They are good at what they are good at.
You want software you configure yourself
There is nothing to log into. If you want to own and tune the rules directly, a self-service tool is the better fit.
You are choosing primarily on price
We are not the cheapest way to buy shipping software. We are a way to buy operational expertise.
If you recognise yourself on the right, that is a genuinely useful answer and it costs you nothing further. Most operations on that side of the page are better served by ShipStation's native automation — here is what it does well and how to get the most out of it. Stay on native automation for as long as it expresses your rules. Come back when it stops.
05 — The Assessment
A working review of your fulfillment operation, performed by the engineers who would build the automation. Not a demo. Not a deck. A document you could hand to your operations manager and act on tomorrow.
Every String engagement starts here. We do not scope work, quote a model, or write a line of automation for anyone we haven't assessed first.
You keep the report either way. Three of those five answers can end with us recommending you don't hire String, and we would rather reach that conclusion now than eighteen months into an engagement.
The deliverable
Operational Assessment report
Contents
Modelled against your own order profile and rate card — not an industry benchmark. We tell you the figure and we show you the arithmetic behind it.
Every recurring decision your team currently makes by hand, ranked by what it costs you to keep making it that way.
Where your current process breaks first as volume grows, and roughly what volume it breaks at.
Single points of failure, undocumented dependencies, and the decisions that currently rely on one particular person being at work that day.
Quantified in hours per week, attributed to where it actually sits — the morning batch build, the exception queue, the packer's judgement call.
Specifically which of your operational rules need logic beyond native automation, and why, with references throughout so you can verify the reasoning.
A sequenced plan, ordered by payback rather than by difficulty, with an estimate of what each stage is worth.
Which engagement model your operation actually belongs in, and the reasoning — including when the honest answer is Foundation, or nothing at all.
Every assessment also improves our understanding of how fulfillment operations actually behave. That's why the Operational Assessment report gets better over time.
How every String relationship begins
Steps 4–5 only if you decide to proceed
Step 1
A short conversation with an engineer, and an export of your recent order and rate data. That is the whole of your involvement.
Step 2
The work happens on our side. We model what a correctly automated version of your operation would have cost over the same period, against your own rate card.
Step 3
Delivered within five business days. We go through every finding with you, including the ones you can fix without us.
Step 4
Based on what the data showed, not on what we would prefer to sell. Sometimes that recommendation is Foundation. Sometimes it is nothing at all.
Step 5
If you proceed, the engineers who assessed your operation are the ones who build it. Nothing is handed off to an implementation team that wasn't in the room.
The report is yours whether you work with String or not.
You are free to take every recommendation in it and implement them yourself, with your own team or another vendor. That is a legitimate outcome and we will tell you so during the walkthrough. We would rather be the firm that told you the truth about your operation than the one that sold you software you didn't need.
Every fulfillment operation eventually accumulates rules that are specific to it. Each review explains the operational problem, how it was approached, and what measurably changed.
All four are anonymized to protect customer confidentiality. Every figure and quote is taken from the original engagement records. The operational lessons are the important part.
Cartonization
4 min
A packing process that depended on the judgement of whoever reviewed orders that day. The judgement was good. The dependency was the problem.
Key result
$9.33 → $8.49 average label
Read the review →
Rate Selection
4 min
The team knew their rules were mis-allocating orders when they built them. They built them anyway, because the alternative did not scale.
Key result
13.7% lower label cost, 60% faster to fulfil
Read the review →
Rate Selection
4 min
A carrier relationship that solved a real reliability problem, then kept being applied to orders it was no longer the right answer for.
Key result
$15.57 → $13.59 average label
Read the review →
Exception Monitoring
4 min
Rules made fulfilment faster and, at the same time, removed the human check that used to notice when an order never actually shipped.
Key result
$2.86 saved per shipment, delays caught before customers called
Read the review →
Also in the knowledge base
Frozen product · Multi-channel · Mixed in-house and 3PL
Migrating from a legacy platform to Shopify with multiple sales channels, split fulfillment and shipping rules dictated by a frozen product. We built the logic that made every one of those requirements hold at launch.
"What I thought would be a difficult ask was no big deal for the String team to implement."
Dan Gilland — Director of Ecommerce
Supplements · Manual batch construction
The team was building ShipStation batches by hand each morning — grouping by product, then choosing carriers. We built rate-aware carrier selection and auto-batching by product and ship date.
"Each morning our orders are already organized by product and ship date, which makes printing and processing incredibly streamlined."
Gains In Bulk
BigCommerce → ShipStation → Extensiv
No existing interface connected their stack. We built the integration from the ground up — shipping data, multi-package logic and real-time order synchronisation — as a turnkey path where none existed.
"What used to be a complex, manual process is now fully automated."
Operations lead
The pattern across all four is the same: the operation was already competent. What it lacked was a way to make its best decision every single time.
A meal-replacement brand's packing expert was excellent — and unavailable on holiday. A nutrition brand's rules were reasonable — and built on estimates. A beverage company's carrier partnership was genuinely good — and applied indiscriminately. An apparel brand's automation was faster — and quietly losing orders. Competence isn't the constraint at your volume. Consistency is.
The Assessment tells you which of these patterns is running in your operation right now, and what it is costing you.
What you are hiring
Every operation we take on adds to something no single warehouse can build on its own: a working knowledge of how these decisions fail. Cold chain, heavy freight, apparel, supplements, frozen product, cross-border — each one taught us something the next operation never had to discover for itself. That accumulated experience is what you are actually hiring. The software is simply how it reaches your orders, correctly, every time, without depending on anyone remembering it.
These are not feature tiers. Each level describes how much of your fulfillment operation String takes responsibility for — from two specific decisions, up to the whole operation. Moving up the ladder does not unlock features. It transfers ownership.
There is no per-shipment pricing and no volume penalty. Growing does not cost you more, which means we are never paid more for making your operation less efficient.
Implementation included
Foundation covers standard operational automation and nothing beyond it — cartonization and rate selection, correct on every order. No business-specific logic, inventory decisions, warehouse routing or ERP integration. If your operation needs those, we will identify it during the Assessment and say so.
Cartonization · Rate selection · Standard automation only
Where most customers belong
Implementation from $2,500
String owns the automation for one warehouse, end to end. Whatever operational logic your floor needs, we build it — there is no per-rule pricing and no scope negotiation each time something changes.
Implementation from $5,000
String owns multiple operational domains at once — the point where shipping decisions stop being separable from inventory, temperature and how work moves through the building.
Scoped to the organization
String effectively owns fulfillment automation for the organization. Multi-warehouse, cross-border, multiple systems of record, and a standing engineering relationship rather than a project.
Most companies aren't. That is exactly what the Assessment is for. We review your operation, tell you which engagement makes sense — if any — and explain the reasoning.
It exists to recommend the right engagement, not to move everyone up the ladder. Sometimes the answer is Foundation. Sometimes it is nothing at all.
For the standard cases, yes — and it does them well. That is exactly what a horizontal platform should be good at, and we are not going to pretend otherwise.
String exists for the rules that are specific to your business: the temperature threshold, the fragility rule, the batching sequence that only makes sense given your pick paths, the joint cartonization-and-rating decision. Those are not gaps in ShipStation's product. They are the point at which a general tool stops being able to encode a particular operation.
What ShipStation automation does, and where custom logic startsBecause a dashboard would be another system for your team to learn, log into, and eventually stop maintaining. Every operations manager we've worked with already has too many tabs open.
Your team keeps working in ShipStation, which they already know. The automation runs behind it. When something needs to change, you tell a person and it changes — you don't go configure it yourself at 6pm.
Sometimes you should, and we will say so. If your problem is genuinely warehouse execution — labour management, slotting, directed picking across a large building — that is a WMS problem and String is the wrong tool.
But most operations that get quoted a WMS are really a few dozen decisions away from fixed. Spending six figures and six months of retraining to encode those decisions is an expensive way to buy logic.
Your ShipStation account is yours and stays exactly as it is. Nothing is rerouted through us, so there is no migration to unwind and no data to extract — the automation simply stops making the decisions, and your team makes them by hand again, as they did before.
We think that's the right way to hold a customer: because leaving would cost you the expertise, not because leaving would cost you your data.
Start here
A short conversation, an export of your order data, and an Operational Assessment report you keep — along with an honest answer about whether String can meaningfully improve your fulfillment operation. If it can't, we'll tell you. That answer requires a detailed review by us and could save you a year of building the wrong solution.
