[compare]
The real alternative is a spreadsheet.
Not a rival platform. The honest competitor is the system you already run: a workbook, a reply-all thread, and a morning sweep of three carrier portals. Here is where it wins, and where it breaks.
[soc 2 type ii · 14-day trial · no card]
[01 / credit where due]
What is the manual system good at?
A comparison that pretends the spreadsheet is useless is lying to you. It is not useless. It is the tool that got you here.
- $0 and already through procurement. No seats, no security review, no renewal.
- Any shape you want, today. A new column is a schema change that takes ten seconds.
- Zero training. Everyone already speaks it, because they built it.
- At low volume it honestly works. One owner, two carriers, a morning sweep: covered.
That system was the right call at 200 shipments a month. The question is what it costs at 12,000.
[02 / at volume]
84h
[median exception age at discovery · weekly-review system]
[median age on a watched feed]
[illustrative data]
Where does it break at volume?
-
Detection waits for the calendar
An exception born Tuesday night surfaces at the Friday review, a median of 84 h old. The expedite that would have saved it closed on Wednesday.
-
Coverage follows memory
The morning sweep checks the carriers that burned you last quarter. The new lane fails silently until a customer calls, and the customer always calls first.
-
Ownership is a volunteer
Reply-all finds whoever is least busy. At volume nobody is least busy, so threads stall with everyone cc'd and no one due.
-
The whole system is one person
It works because one person runs it, and that is also the ceiling: about 1,000 shipments a month, and zero coverage the week they are out.
[the spreadsheet was never the problem. the volume was.]
[03 / line by line]
How do the two systems compare?
Eight rows, honestly scored. The spreadsheet wins two of them.
| Capability | Throughline | Spreadsheets + email |
|---|---|---|
| Software cost | $1,500 to $4,500 per month | $0, on licenses you already own |
| Time cost | About 6 min per exception, all of it working the fix | About 22 min per exception, most of it finding it |
| Time-to-detect | 38 min median, feed-paced | 84 h median, review-paced |
| Coverage | Every shipment on every connected feed | The lanes someone remembers to check |
| Ownership | Named owner and deadline, attached by rule | Whoever answers the thread first |
| Audit trail | One thread per shipment, exportable | Scattered across inboxes and tabs |
| Flexibility | Rules run on lane, carrier, tier, and value. That is the whole grammar | Any column, any formula, this afternoon |
| Key-person risk | Runs through PTO, turnover, and 2 a.m. | The sheet's author is the system |
[illustrative comparison · demo product · time figures from the roi model defaults]
[the short version]
Keep the spreadsheet. Give it fewer emergencies.[04 / the field]
What about the other alternatives?
Four more ways teams solve this, ranked by how often they are the real answer. Each one is good at something. Take the credit seriously and the limits get easier to see.
-
[02 / the tms module]
The exception module you already own
[good at]
It is paid for, it lives inside your system of record, and the vendor will help you switch it on. If yours is on and tuned, keep it.
[where it loses]
It sees what your TMS sees: your own milestones, not the carrier's feed. Turning it on is a services project, and its generic thresholds page everyone until someone turns them back off. That is usually why it was off.
-
[03 / visibility platforms]
The project44 / FourKites class
[good at]
Genuinely excellent at where freight is: global, multimodal, twenty times our breadth. If you need a network view, buy one.
[where it loses]
It answers a different question. A map of every shipment still leaves someone exporting the broken ones to a spreadsheet to work them: no owner, no deadline, no close. If you already run one, Throughline reads its feed as an input.
[real vendors named as market context and integrations · illustrative · no partnership implied]
-
[04 / another coordinator]
Hiring another person
[good at]
People handle the cases no rule anticipates, and the first dedicated coordinator is transformative. You will still want the humans.
[where it loses]
Roughly $68,000 a year fully loaded, awake eight hours in twenty-four, and the triage logic lives in their head. The question is not people or software; it is what the people spend the hours on.
-
[05 / doing nothing]
Finding out at the Friday review
[good at]
Free, until it is not. At low volume, with forgiving customers, finding out on Friday can be survivable.
[where it loses]
The cost still posts: expedites, freight credits, and the account that quietly re-sources. It just posts later, unattributed, on someone else's budget line. The calculator models it; bring your own numbers.
[05 / our side of the ledger]
What does switching actually cost?
A fair comparison prices both sides. This is ours, stated plainly.
- 2 to 3 weeks to the first live exception. Week one connects feeds and backfills 90 days; week two tunes tolerances against them.
- Carrier credentials or an EDI 214/990 feed. No TMS project, no ERP dependency.
- $1,500 a month floor. Public pricing, annual terms, a 30-day out inside the first 90 days.
- Your spreadsheet survives. Everything exports to CSV or Parquet, any time, including on the way out.
- The security review pack and subprocessor list live on the trust page.
And if you are under about 1,000 shipments a month: stay on the sheet. It is the right tool, and we would rather say so here than on a call.
[next step]
Run your own numbers first.
Every formula in the calculator is on the page. Bring the output to the demo and argue with it there.