What data does freight cost software actually need?

Anyone wanting to digitally analyze freight costs usually asks the technical question first:

Which file do we need?

Excel? CSV? Invoice PDF? TMS export?

But the more important question is:

What do we want to do with the data?

Because a freight cost analysis doesn't need exactly the same information as an automated invoice audit or a freight tender.

That's why ShipCheck works with different types of data: invoice data, transport or shipment data, and tariff information. For the KPI dashboard, ShipCheck explicitly names invoices or shipment data in PDF, Excel, or CSV format.

For automated freight invoice auditing, however, invoices and tariff data are needed.

The key principle is therefore:

Not every function needs the same data.

Three data types, three different tasks

For freight cost management, the relevant information can be roughly divided into three groups:

• Invoice data: What was billed?
• Shipment and transport data: What was actually transported?
• Tariff data: What should this transport have cost according to the contract?

Depending on the use case, this data is combined differently.

This quickly makes one thing clear:

"What data does freight cost software need?" has no single universal answer.

The answer depends on the use case.

1. What data do you need for a freight cost analysis?

For the KPI dashboard, ShipCheck names as the required basis:

Invoices or shipment data in PDF, Excel, or CSV format.

This data is structured and then displayed as KPIs, trends, and cost analyses.

The website also describes how ShipCheck combines carrier invoices, TMS exports, and ERP data into a unified data basis.

Among other things, the following dimensions can be evaluated:

• Costs
• Lanes
• Modes of transport
• Carriers
• Individual shipments

ShipCheck also calculates metrics such as cost per kg, per shipment, per route, and per carrier. Which metrics are particularly relevant is covered in the article on freight cost KPIs.

This results in an important distinction

Not every company has to deliver the same maximum data set.

For example, if only the cost trend by carrier is to be examined, fewer analysis attributes are needed than for a drill-down by:

Mode of transport → lane → shipment → cost structure

The more dimensions that are to be analyzed, the more matching transport attributes must be present in the data basis. How granular such an evaluation can become is described in the article on freight cost analysis at shipment level.

The relevant question is therefore:

Which evaluation should be possible later?

2. Invoice data: What was actually billed?

Invoice data forms the financial actual side.

First, it answers:

What did the carrier bill?

ShipCheck can process carrier invoices for this. The website also mentions credit notes and EDI exports in addition to invoices.

The system extracts shipment data from this, identifies invoice line items, and structures the information in a shared data basis.

Invoice data is therefore important for two different tasks:

• Freight cost analysis: What costs were actually billed?
• Freight invoice auditing: Do these billed costs match the agreed tariff?

For the second question, invoice data alone is not enough.

3. Shipment data: What was actually transported?

Costs become particularly valuable analytically only once they can be linked to the underlying transports.

That's why ShipCheck works at shipment level and displays, among other things:

• Lanes,
• Recipient structures,
• Mode-of-transport information,
• Costs

at this level.

In automated invoice auditing, the engine also identifies, among other things:

• Carrier
• Route
• Weight
• Service level
• Additional services

to calculate what the respective shipment should have cost.

These details are therefore typical examples of information that can become important depending on the use case. The article Analyzing carrier invoices and shipment data shows how such data can be used in practice.

This does not mean, however, that every single record must necessarily contain each of these fields.

What matters is:

The data must contain enough context to perform the desired analysis or calculation.

4. What data do you need for freight invoice auditing?

The data logic is clearer for automated invoice auditing.

The ShipCheck website explicitly states:

"Invoices as well as tariff data are needed and digitized."

Because the audit compares:

Actual costs from the invoice with target costs from the agreed terms.

To do this, contracts or tariffs are first made calculable. ShipCheck processes for this purpose:

• Freight contracts
• Attachments
• Surcharge tables
• Tariff matrices

as PDF or Excel.

Among other things, the AI recognizes tariff logic, tiered pricing, diesel formulas, and special clauses, and makes the tariffs calculable. The article Digitizing freight tariffs shows how this works technically.

Invoice line items are then checked against these terms.

The minimal logic is therefore:

• Invoice: What was billed?
• Tariff: Which terms were agreed?
• Shipment: To which specific transport must these terms be applied?

Without tariff data, there's no solid target side.

Without invoice data, there's no actual side.

5. Why tariff data is different from transport data

This difference is often underestimated.

A shipment record can, for example, show:

• Route
• Weight
• Carrier
• Service
• Date

But it doesn't automatically say:

What should this shipment have been allowed to cost according to the contract?

For that, you need the agreed pricing logic. This can include, for example:

• Base rates
• Weight brackets
• Zones
• Surcharge tables
• Diesel formulas
• Special terms

ShipCheck digitizes such tariff structures so that a target cost calculation can be derived from the transport attributes.

This means two data types differ:

Transport data describes the shipment.

Tariff data describes how this shipment should be priced.

6. What data do you need for tendering?

With a freight tender, the task shifts again.

Here the question is not primarily:

What was billed so far?

But rather:

What would new carrier prices cost applied to our actual transport profile?

For this, ShipCheck needs historical transport or shipment data.

The reason is simple:

An offer can only be evaluated realistically if it's applied to the transport structure the company actually has.

Depending on the tender, this includes, for example, different:

• Lanes
• Shipments
• Weight classes
• Modes of transport
• Cost components

The tendering function works with real transport data at shipment level and can compare offers by route, mode of transport, and cost components – see Evaluating a freight tender.

This makes the data set the calculation basis for the new tariffs – not just a description of the past.

7. Historical data and forecast data serve different purposes

For tendering, historical shipment data is the starting point.

It shows:

Which transports actually took place?

But this doesn't necessarily match the future transport profile exactly.

If volumes, lanes, or business volume are expected to change, forecast data can also be used.

The two data types should therefore be considered separately:

• Historical data: real starting profile
• Forecast: expected future change

ShipCheck can apply forecasts to tender results, thereby calculating future transport volumes with new tender prices.

A forecast is therefore not a substitute source for historical shipment data.

It complements it.

8. What's more important: lots of data or the right data?

More data isn't automatically better data.

A data set with hundreds of fields doesn't help much if central costs or transport attributes are missing or can't be cleanly assigned.

The ShipCheck website describes exactly this problem with fragmented logistics data:

Invoices come in different formats, TMS exports aren't standardized, and manual Excel evaluations produce an incomplete picture. ShipCheck structures data from different sources for consistent evaluations.

For analysis quality, three questions are therefore more important than the sheer number of columns:

• Is the relevant information available?
• Can it be assigned to the correct shipment or cost item?
• Is it comparable across different sources?

9. What data is typically missing for a desired evaluation?

A simple reverse logic helps here.

Don't start with:

"What data do we have?"

But instead with:

"What question do we want to answer?"

Example:

• Question: Which carriers cause the highest costs? Then costs must be assignable to the respective carrier.
• Question: Which lanes are becoming more expensive? Then the analysis needs lane or route and cost data.
• Question: How is cost per kg developing? Then cost and weight data are needed.
• Question: Which invoice was billed incorrectly? Then you need the invoice, relevant shipment information, and the contractual tariff logic.
• Question: Which provider would be economically better in a new tender? Then historical shipment data and the new carrier prices are needed.

The basic principle is:

Every metric needs the data fields it's calculated from.

10. Does everything have to be in a single file?

No.

ShipCheck is specifically designed to combine different data sources. The website names:

• Carrier invoices
• TMS exports
• ERP data

and describes their combination into a unified data basis.

Elsewhere, this information is described as order data, transport data, and invoice data.

This is relevant because in practice a single source rarely contains everything:

• Costs may be in the invoice.
• Transport details in the TMS.
• Additional order information in the ERP.
• The tariff logic, in turn, in the contract or a tariff matrix.

The real value emerges when this information can be evaluated together. The article Does freight cost software need an ERP or TMS integration? clarifies whether an interface is strictly necessary for this.

11. Which file formats can ShipCheck process?

For the KPI dashboard, ShipCheck names:

• PDF
• Excel
• CSV

for invoice or shipment data.

For the audit function, tariff documents can be uploaded as PDF or Excel. For invoices, the audit page names:

• Carrier invoices
• Credit notes
• EDI exports

This means the data source isn't limited to a single standardized format.

What matters after that is how the content is structured.

12. Data quality determines how deep the analysis can go

An important principle applies:

The software can only analyze dimensions that can be derived from the available data.

• If, for example, no lane is available or can be determined, no reliable lane analysis can be produced.
• If no weight is available, no meaningful "cost per kg" KPI can be calculated from it.
• If tariff terms are missing, a complete contractual target cost audit cannot be carried out.

That's why data quality isn't an abstract IT topic. It directly determines:

Which questions the freight cost analysis can answer.

A practical data check before you start

Before a freight cost analysis, four questions can be used to check whether the data basis is suitable.

1. Which use case should be solved?

Analysis, reporting, audit, or tendering?

2. Which metrics or calculations does this use case need?

• Cost per shipment
• Cost per kg
• Cost by carrier
• Cost by lane
• Target-actual variance

3. What information is needed for that?

Is the relevant transport, cost, and, where applicable, tariff information available?

4. Can this information be linked together?

Can invoice, shipment, carrier, and tariff be assigned to the same transport as needed?

This makes it possible to assess data quality much more concretely than with the general question: "Is our data good?"

How ShipCheck creates a common basis from different data

The ShipCheck approach can be simplified as follows:

• 1. Import data: Depending on the use case, invoice, shipment, or tariff data is processed.
• 2. Structure data: ShipCheck combines information from different sources into a unified data basis.
• 3. Analyze transport costs: Costs can be evaluated by shipment, route, and carrier, among other things.
• 4. For audits, apply tariff logic: Invoice line items are checked against the agreed terms.
• 5. For tendering, recalculate historical transports: New carrier prices are calculated against the real transport structure.

This means the required data basis always depends on the question that needs to be answered.

Conclusion: Not as much data as possible – but the right data

Freight cost software doesn't need a universal data set for every function.

• For an analysis, invoice or shipment data alone can already be enough.
• For a tariff-based invoice audit, the agreed tariff data is additionally needed.
• For tendering, historical shipment data is decisive because new offers are calculated against the real transport structure.

The most important distinction is therefore:

• Invoice: What was billed?
• Shipment: What was transported?
• Tariff: What should it have cost?
• Forecast: What do we expect in the future?

The better this data fits the respective use case, the more precisely ShipCheck can analyze costs, check discrepancies, or calculate tender offers.

So the decisive question isn't: "Do we have all conceivable logistics data?"

But rather: "Do we have the data that is necessary for our specific freight cost question?"

A look at the subscription plans shows what scope is possible for getting started.