Measurement
Part of Check packaging marketing measurement before you report results
How to design a packaging marketing benchmark method
Agree the metric, peer set and period, record the method, then report a median and range on a fixed cycle — a repeatable packaging marketing benchmark method.
What to take away
Packaging marketing benchmark research is the work of comparing your own packaging and unboxing metrics with a defined peer group, using a method you can repeat and explain.
- Agree the metric, the peer group and the period before you collect anything.
- Write down your sample, your exclusions and your source in the same document as the results.
- Report a median and a range instead of a single headline number.
- Treat another company's figure as context, not as a target you must match.
- Re-run the exercise on a fixed cycle and log what changed.
Define the metric before you ask for data
Most benchmark requests arrive before anyone has agreed what is being measured. Average packaging cost can mean material cost per unit, total fulfilment cost per order, or cost per SKU per month. Those are different numbers and they move for different reasons.
Three meanings of packaging cost
Material cost per unit
- What it counts
- Packaging materials only
- Unit
- Pence per parcel
- Moves when
- Material prices change
Total fulfilment cost per order
- What it counts
- All fulfilment costs
- Unit
- Pence per order
- Moves when
- Shipping volumes change
Cost per SKU per month
- What it counts
- Packaging by product line
- Unit
- Pence per SKU
- Moves when
- Range size changes
Pick one primary metric and one supporting metric. Write the definition in a sentence a colleague could apply without asking you a question. Then state the unit: pence per parcel, grams per unit, or percentage of orders.
For example, a team shipping 4,000 orders a month might track packaging cost per order and the damaged-in-transit rate. Those two metrics together tell a more useful story than either alone.
Marketing metrics follow the same rule. An unboxing benchmark might define the share of deliveries that generate customer content, review sentiment on unboxing posts, or return rate and repeat purchase rate by pack format. Each one needs a written definition, a unit and a peer group that measures it the same way.
Choose a peer set you can actually measure
A benchmark is only as good as its comparison group. Direct competitors are the obvious pick, but you may not be able to measure them. Adjacent categories, trade body members or your own past quarters often work better.
Trade associations are often the quickest route to a peer set. The Food and Drink Federation represents UK food and drink manufacturers, and its members are the kind of comparable businesses most packaging benchmarks draw on when direct competitors will not share figures.
If your peer set depends on retailer sales data or customer records, that is personal data, and UK GDPR governs how you hold and share it. The Information Commissioner's Office (ICO) publishes the guidance that applies, and its rules on lawful sharing bind a benchmark dataset just as they bind a customer list. The data and privacy standards guidance from BSI sets out the controls, such as those in ISO/IEC 27001, that make a shared dataset defensible.
Attribution choices shape what you can compare. If one brand counts a sale in the week of delivery and another waits a month, the same metric produces different answers. Aligning those windows first makes the numbers comparable, and our guide to attribution methods sets out the common approaches.
Record the method so the numbers survive scrutiny
Every benchmark report needs a method note: who supplied the data, when it was collected, how many records, and what was left out. Without it, a number is a rumour with a decimal point.
Copy this structure for the method note:
- Metric name and a one-sentence definition.
- Unitpence per parcel, grams per unit, or percentage of orders.
- Peer groupevery member named, with the reason for including it.
- Collection windowstart date and end date.
- Sample size and exclusions.
- Data source and who supplied it.
- Resultmedian, range and sample size.
- Owner and date of the next refresh.
Your own figures are not the only comparable data in the UK. Large producers report the packaging they place on the market to the Environment Agency. That reporting ran through the National Packaging Waste Database under the 2007 regulations and now sits in the packaging Extended Producer Responsibility scheme (pEPR) under the 2024 regulations, and WRAP publishes UK packaging and recycling data. Those returns give comparable material figures, even though none of them tells you cost per parcel.
Packaging metrics also sit inside a wider economy. The ONS publishes gross domestic product figures that set the backdrop for consumer spending, which helps explain a fall in order volumes your packaging did not cause.
Your material metrics also sit inside regulation. The Packaging (Essential Requirements) Regulations 2015 limit heavy metals and require minimisation, so grams per unit has a legal floor as well as a commercial one.
Set a reporting rhythm you can keep
Quarterly is usually enough for packaging benchmarks. Monthly reporting invites noise, and annual reporting is too slow to change a decision. Fix the date, the owner and the audience.
Report the median, the range and the sample size together. If your peer group has five companies in it, say so on the slide, because a range from five data points deserves a caveat.
Reporting only matters if it changes a decision. The measurement and reporting guide shows how benchmark output feeds into planning and budget conversations. A benchmark that no one acts on is a spreadsheet, not research.
A benchmark research checklist
Benchmark research checklist
- One primary metric with a defined unit.
- A named peer group with a reason for each member.
- A start and an end date for collection.
- Sample size and exclusions recorded.
- Median and range, not only an average.
- A named owner for the next refresh.
Common questions
The questions below cover the choices that come up most often when a team runs its first benchmark.
How many companies do I need for a benchmark?
There is no fixed number, but three to five peers with comparable models is more useful than twenty loosely related firms. State the count in your report so readers can judge the range.
Should I use an average or a median?
Use the median. Packaging costs and returns rates include outliers, and an average pulled upwards by one unusual order will mislead a planning meeting.
How often should I refresh packaging marketing benchmark research?
Quarterly suits most teams. Refresh sooner if you change supplier, switch material or alter pack sizes, because those changes break comparability with earlier periods.
