For decades, spreadsheets have been one of the most useful tools in packaging.
They are flexible, familiar and almost universally available.
Packaging engineers use them to manage specifications. Procurement teams compare suppliers. Sustainability teams calculate material usage. Regulatory teams track requirements. Finance teams model costs. Project teams evaluate alternatives.
And for many of these tasks, spreadsheets remain perfectly appropriate.
The problem is not Excel.
The problem begins when a spreadsheet designed to support a calculation gradually becomes the place where an organization tries to understand an entire packaging portfolio:
Hundreds of SKUs become thousands;
One market becomes twenty;
Material specifications evolve;
Suppliers change;
Regulations multiply;
Evidence expires;
Different teams create different versions of the same dataset.
And eventually, a tool that is excellent at manipulating information starts being asked to manage something fundamentally different:
complexity, relationships and decisions.
That is where the limits begin to appear.
Spreadsheets work extremely well when the question is contained
There are many packaging decisions that companies should continue making in spreadsheets:
Calculating the grammage of a material;
Comparing the cost of three packaging alternatives;
Estimating pallet utilization;
Analyzing test results;
Building an initial business case;
Running an engineering calculation.
Excel is exceptionally effective when the dataset is relatively contained and the relationships between variables are understood.
But consider a different question:
“Which packaging components should we redesign first to reduce our exposure to future European requirements?”
Answering that question may require understanding:
Which products are sold in Europe;
Which packaging systems they use;
Which components those systems contain;
What materials those components contain;
Their weights;
Their suppliers;
Their recycled content;
Their recyclability;
The evidence supporting those characteristics;
Which components are shared between SKUs;
And which regulatory requirements may apply.
Then the organization may want to understand the cost of changing them.
And their carbon impact.
And whether an alternative supplier exists.
And whether the same redesign makes sense in the United States.
That is no longer simply a spreadsheet calculation.
It is a packaging knowledge problem.
The first warning sign: multiple versions of the truth
One of the earliest symptoms appears innocently.
A spreadsheet is emailed to another team.
Someone makes a copy.
A supplier sends an updated file.
A colleague adds three columns.
Another business unit adapts the template.
Someone saves:
Packaging_Data_Final.xlsx
Then:
Packaging_Data_Final_v2.xlsx
Eventually:
Packaging_Data_Final_v2_UPDATED.xlsx
The problem is almost a cliché.
But its consequences are not.
When packaging decisions become strategically important, companies need to know not simply what a value is, but whether it is the current governed value.
If one file says a component weighs 12.5 grams and another says 13.1 grams, the organization needs more than a formula.
It needs to understand:
Where did each value come from?
When was it provided?
Which specification version does it describe?
Was one measured and the other declared by a supplier?
Has either been verified?
Which value is currently approved?
Should the difference be investigated?
A spreadsheet can store both numbers.
It cannot, by itself, create the governance required to determine what those numbers mean.
The second warning sign: relationships become more important than rows
Spreadsheets naturally encourage us to think in tables:
One row represents one SKU;
Another spreadsheet contains suppliers;
Another contains packaging components;
Another contains regulatory information;
Another contains sustainability data.
But packaging itself is not a table.
It is a network of relationships.
A product can have multiple SKUs.
A SKU can use a packaging system.
That system can contain several packaging levels.
Each level can contain several components.
A component can contain multiple materials.
The same component may appear across hundreds of SKUs.
One supplier may provide several components.
A material may be subject to different requirements in different markets.
Evidence may support one specific characteristic of one component for a defined period.
Once those relationships become important, flattening everything into one enormous spreadsheet creates a choice between two bad outcomes.
Either information is duplicated repeatedly across rows.
Or critical relationships exist only implicitly in the minds of the people maintaining the file.
Neither scales particularly well.
Companies should stop using spreadsheets to reconcile conflicting packaging facts
This is one of the clearest boundaries.
Imagine three files.
An engineering specification says:
Component weight: 18.4 g
A supplier declaration says:
Component weight: 17.9 g
A sustainability dataset says:
Component weight: 18.0 g
Which one is correct?
The answer should not automatically be:
the value in the latest spreadsheet.
There may be perfectly legitimate reasons for the difference.
The files may describe different revisions.
One value may include an accessory.
Another may represent a nominal specification while another represents an actual measurement.
The supplier may have changed the component.
Or one value may simply be wrong.
The appropriate response is therefore not always to choose a number.
Sometimes it is to create a controlled conflict that requires resolution.
That distinction matters enormously.
Because when companies force uncertain information into apparently clean spreadsheets, they can create something more dangerous than missing data:
false certainty.
Companies should stop using spreadsheets as evidence systems
A spreadsheet cell might say:
Recycled content: 35%
But what supports that number?
A supplier declaration?
A certification?
A laboratory report?
An email?
An estimate?
Which supplier?
Which component?
Which version?
When was the evidence issued?
Does it expire?
Does it support the entire component or only one material within it?
The number and the evidence supporting the number are not the same thing.
As regulatory scrutiny and sustainability substantiation increase, this distinction becomes increasingly important.
Companies need to be able to move from:
value
to:
source
to:
evidence
to:
version
to:
decision.
A hyperlink in an Excel cell may help someone locate a document.
It does not create an evidence architecture.
Companies should stop using spreadsheets to determine whether their data is “complete”
A spreadsheet can easily identify empty cells.
That does not necessarily tell a company whether its packaging data is sufficient.
Because packaging data completeness is contextual.
Suppose a company does not know the exact dimensions of a secondary carton.
For one analysis, that missing information may be irrelevant.
For another — such as logistics optimization — it may be critical.
Likewise, a missing recycled-content value might not prevent an engineering performance assessment but could prevent a regulatory or sustainability analysis.
This means the meaningful question is not:
“Is our packaging database complete?”
It is:
“Do we have sufficient trusted information to make this specific decision?”
That is a fundamentally different concept.
A packaging dataset can be incomplete and still perfectly usable for one decision.
It can also appear highly complete while missing the few pieces of information required for another.
Traditional spreadsheet completeness metrics rarely capture that distinction.
Companies should stop manually translating international packaging data
Global packaging portfolios create another layer of risk:
European and American datasets frequently use different conventions;
Millimetres and inches;
Grams and ounces;
Kilograms and pounds;
Grams per square metre and pounds per thousand square feet;
Decimal commas and decimal points;
And even dates can create ambiguity: 04/07/2026 could mean 4 July 2026 in one dataset and April 7, 2026 in another.
These issues appear trivial because each conversion is simple:
At portfolio scale, they are not;
A manually converted value can become detached from its original representation;
An ambiguous date can be interpreted incorrectly;
A formula can be copied incorrectly;
A converted value can be rounded and then reused as though it were the original measurement;
And once the result is copied into another spreadsheet, its history may disappear entirely.
International packaging data therefore needs more than conversion.
It needs normalization with provenance.
The original value should remain visible.
The normalized value should be deterministic.
Ambiguity should be identified rather than guessed.
And the display value can then be adapted to the working conventions of the user.
Companies should stop managing regulatory complexity as an endless collection of columns
Packaging regulation creates an understandable temptation.
Add another column.
PPWR recyclable?
Then:
Recycled content
Then:
EPR category
Then another market.
Then another requirement.
Then another deadline.
Soon the spreadsheet contains dozens or hundreds of regulatory attributes.
But regulation does not exist as a static collection of yes/no fields.
Requirements depend on context:
Market;
Material;
Packaging level;
Format;
Use;
Timing.
Sometimes company characteristics.
And regulations evolve.
The more jurisdictions a company operates in, the more problematic a flat regulatory spreadsheet becomes.
This is particularly visible across Europe and the United States.
The same packaging architecture may interact with a European regulatory framework and several different US state requirements.
The solution is not simply to add more columns.
It is to connect packaging knowledge with regulatory knowledge.
Companies should stop using spreadsheets to model portfolio-wide change manually
This may ultimately be the biggest limitation.
Imagine changing one packaging component:
Which SKUs are affected?
Which factories?
Which suppliers?
Which markets?
Which specifications?
Which tests need to be repeated?
Which artworks might change?
Which regulatory assessments need to be updated?
What inventory could become obsolete?
What happens to cost?
What happens to sustainability performance?
A spreadsheet can certainly calculate some of these consequences.
But first someone needs to know every relationship that should be included in the calculation.
At portfolio scale, that becomes increasingly difficult.
This is where structured packaging knowledge creates a fundamentally different capability.
If relationships between products, packaging systems, components, materials, suppliers, markets and evidence are explicit, the organization can begin to calculate the impact of change through the portfolio.
That moves packaging management from static reporting toward decision intelligence.
The solution is not to eliminate Excel
There is an important distinction here.
Companies should not stop using spreadsheets.
They should stop asking spreadsheets to perform jobs for which they were never designed.
Excel remains an extraordinary interface for working with data.
In fact, for many packaging organizations, spreadsheets may remain one of the most practical ways to collect and exchange information for years:
Suppliers understand them;
Engineers understand them;
Procurement understands them.
They are flexible and accessible.
The opportunity is therefore not necessarily to replace Excel.
It is to change its role.
The spreadsheet can remain an input and an output without remaining the system of intelligence.
A supplier can still return an Excel template.
An engineer can still analyze results in Excel.
A customer can still export a portfolio.
But behind those interactions should increasingly sit a structured layer capable of understanding what the data represents and how it connects.
The transition is from files to governed knowledge
That transition can be summarized simply.
A traditional workflow often looks like this:
Files → Spreadsheets → Analysis → Presentation → Decision
Much of the knowledge created during the analysis disappears when the project ends.
A more intelligent architecture looks different:
Files → Structured packaging knowledge → Analysis → Decision → Reusable knowledge
Information from spreadsheets, specifications, supplier documents and systems becomes part of a connected representation of the packaging portfolio:
Values retain their sources;
Conflicts remain visible until resolved;
Evidence remains attached;
Relationships are preserved;
Changes are versioned;
And future analyses can reuse what the organization has already learned.
This does not make the spreadsheet obsolete.
It makes the knowledge behind it persistent.
AI makes this distinction even more important
The arrival of generative AI creates a strong temptation to skip this foundation.
Connect an AI model to a folder full of spreadsheets and documents.
Ask questions.
Generate answers.
It can look remarkably powerful.
But there is a fundamental problem.
If two spreadsheets contain conflicting values, which one should the AI trust?
If a supplier declaration has expired, does the AI know?
If 04/07/2026 appears in a file, does it know whether the dataset uses European or American date conventions?
If a component has been superseded, can the model distinguish the historical specification from the current one?
If a required value is missing, will the system acknowledge the gap — or produce a plausible answer anyway?
AI can make fragmented information easier to interrogate.
It does not automatically make that information trustworthy.
That is why companies should resist the idea that AI can simply be placed on top of existing packaging spreadsheets and transform them into intelligence.
AI can accelerate packaging decisions only after the organization establishes what information it can trust.
The real question is not “Should we replace Excel?”
That is the wrong debate.
The better question is:
Which packaging decisions have become too complex to depend on spreadsheets alone?
The threshold will differ between companies.
But there are clear signals:
When multiple sources describe the same packaging facts;
When provenance matters;
When evidence needs to be maintained;
When the same components are shared across large portfolios;
When decisions span multiple countries;
When regulatory requirements interact;
When changes need to be traced through hundreds of SKUs;
When several teams need to rely on the same governed information;
And when AI is expected to support important decisions.
At that point, the organization no longer has simply a spreadsheet problem.
It has a packaging knowledge architecture problem.
From spreadsheet management to packaging intelligence
The spreadsheet has been one of the most important tools in modern packaging management.
It will remain one.
But packaging complexity is changing the role it should play.
The future is unlikely to be a world without Excel.
It is much more likely to be a world in which spreadsheets become one interface among many for interacting with a deeper, structured packaging knowledge layer.
A layer capable of knowing:
What a value means;
Where it came from;
Whether it can be trusted;
What it is connected to;
And ultimately which decisions it can safely support.
That is the transition from managing packaging data to building packaging intelligence.
And for organizations managing thousands of products across suppliers, materials and regulatory environments, it may become increasingly difficult to avoid.
The AUDREN Perspective
At AUDREN, we do not believe companies need to abandon the tools their packaging teams already use.
The opposite approach is often more practical.
Start with the packaging data you already have.
Spreadsheets, specifications and supplier files can remain valuable inputs.
The challenge is to transform that fragmented information into structured, traceable and decision-ready packaging knowledge.
That means preserving the original source, normalizing information without losing provenance, identifying conflicts rather than hiding them, connecting evidence to facts, understanding relationships across the portfolio and determining whether the available information is sufficient for a particular decision.
Only then does AI become truly useful.
The future of packaging will not be built by replacing every spreadsheet.
It will be built by knowing which decisions should no longer depend on one.