Online shopping problems by age: surprise costs and the route to help
Eurostat’s 2025 estimates show a reversal in 15 EU countries: buyers aged 16–24 reported unexpected costs more often than complaint difficulties, while buyers aged 65–74 reported complaint difficulties more often. A closer look shows why a store should review both its price information and what happens when a customer needs help.
Eurostat’s 2025 household survey. Buyers aged 16–74 who purchased online in the previous three months. Analysis completed October 17, 2026.
Abstract
Online shopping problems do not have the same balance across age groups. In Eurostat’s 2025 EU estimates, 6.79% of recent buyers aged 16–24 reported final costs higher than indicated, compared with 5.59% reporting difficulty with complaints or redress. Among buyers aged 65–74, the corresponding figures were 2.48% and 5.72%. The ordering reversed between these ages in 15 of the 27 EU countries.[1]
A smaller share of older buyers reported at least one of the survey’s eight named problems, but complaint difficulties were more common within that group. These are descriptive survey estimates, not evidence that age causes the differences. For merchants, the useful question is whether the route to help receives as much attention as prices and delivery. This study provides a practical review worksheet, alongside the full age comparisons, country exceptions and calculation method.
Picture a small retailer preparing a product launch. The photographs are ready, the social posts are scheduled and the checkout works. Then someone asks a less familiar question: if a buyer receives the wrong item, can they find the next step without returning to a social post or searching through old messages?
That question sits beside several others. Is the delivery charge visible early enough? Does the promised arrival date match the order confirmation? Can a buyer find information about a guarantee? A successful test purchase does not answer every question about the experience around it.
We used Eurostat’s online shopping survey to examine which problems buyers report, and how their balance differs by age. The aim is to help a merchant decide what to inspect. The survey describes people’s experiences across sellers; it does not identify a particular store’s weaknesses or test which improvement would increase its sales.
A purchase has more than one place where information matters
- Price and delivery information
- The order and its confirmation
- Help when something goes wrong
Older buyers report surprise costs less often; complaints follow a different pattern
The clearest contrast involves two specific experiences: paying more than indicated and having difficulty with a complaint. Eurostat asks about both among people who bought through a website or app in the previous three months. “Complaint difficulty” here includes difficulty obtaining redress or receiving no satisfactory response after a complaint. It is broader than being unable to find a contact button.[2]
In the EU aggregate, higher-than-indicated costs were reported by 6.79% of buyers aged 16–24 and 2.48% of those aged 65–74. The percentage declines across each of the six published age bands. Complaint and redress difficulties follow a different pattern: 5.59% in the youngest group, 5.72% in the oldest, and fluctuating figures in between.[1]
Complaint difficulties exceed surprise costs in every EU age band from 25–34 onward. That is an ordering of published estimates, not a finding that something changes at a person’s 25th birthday. The age bands cannot locate such a turning point, and the survey does not follow the same people as they grow older.
For a merchant, price clarity and complaint handling deserve separate checks. A page that explains the full price may still leave a customer unsure what to do about a damaged order. Equally, an easy contact route does not resolve an unexpected charge. Combining both into one broad “customer experience” score can hide which task needs attention.
Costs and complaint difficulties across six age groups
Percentage of recent online buyers in each age group, EU aggregate, 2025 survey. Purchases took place in the three months before the interview.
| Age group | Final costs higher than indicated | Complaint or redress difficulties |
|---|---|---|
| Ages 16–24 | 6.79% | 5.59% |
| Ages 25–34 | 5.36% | 7.11% |
| Ages 35–44 | 4.60% | 6.20% |
| Ages 45–54 | 3.97% | 7.03% |
| Ages 55–64 | 3.03% | 5.69% |
| Ages 65–74 | 2.48% | 5.72% |
The reversal appears in 15 countries, with useful exceptions
The EU average could conceal different national patterns, so we compared the same two problems within each country. In 15 countries, costs were more common than complaint difficulties among buyers aged 16–24, while complaint difficulties were more common among those aged 65–74. All 27 countries had the four estimates needed for this comparison.[1][3]
The other 12 countries matter. Belgium, Czechia, Germany, Greece, the Netherlands and Portugal had complaint difficulties above costs in both age groups. Bulgaria, Ireland, Latvia, Poland and Romania had costs above complaint difficulties in both. Cyprus showed the opposite reversal, although its percentages for these problems were small.
These exceptions change how the result should be used. Fifteen countries sharing an ordering does not make the pattern universal, and a country count gives a small country the same weight as a large one. It also says nothing about which country has the best stores. Respondents may buy from domestic or foreign sellers.
The country table is a check on the headline, not a targeting rule. A retailer serving several markets can use it to choose questions for its own review, while checking the actual complaints and buying journeys it already has evidence about. The study does not justify assigning a problem to an individual customer because of their age or country.
Costs and complaint difficulties in all 27 EU countries
Percentage of recent online buyers in the stated age group, 2025 survey. Compare the two problem percentages within each age group; the table shows all 27 countries, not a ranking.
View the estimates for all 27 countries
| Country | Ages 16–24: costs higher than indicated | Ages 16–24: complaint difficulties | Ages 65–74: costs higher than indicated | Ages 65–74: complaint difficulties |
|---|---|---|---|---|
| Belgium | 6.73% | 7.01% | 0.81% | 2.29% |
| Bulgaria | 8.31% | 5.08% | 7.77% | 4.57% |
| Czechia | 3.27% | 8.62% | 0.48% | 8.57% |
| Denmark | 8.19% | 4.43% | 1.38% | 3.73% |
| Germany | 0.99% | 2.98% | 0.54% | 2.66% |
| Estonia | 3.71% | 1.57% | 1.03% | 2.67% |
| Ireland | 23.95% | 0.58% | 6.95% | 5.31% |
| Greece | 0.00% | 1.13% | 0.79% | 1.77% |
| Spain | 12.32% | 8.73% | 3.57% | 6.98% |
| France | 7.77% | 5.99% | 3.18% | 4.74% |
| Croatia | 0.94% | 0.46% | 1.23% | 8.53% |
| Italy | 12.43% | 9.36% | 6.63% | 9.81% |
| Cyprus | 0.18% | 0.26% | 1.39% | 1.02% |
| Latvia | 2.03% | 0.64% | 1.57% | 0.00% |
| Lithuania | 2.48% | 1.26% | 0.00% | 1.10% |
| Luxembourg | 13.19% | 7.15% | 5.62% | 8.90% |
| Hungary | 9.47% | 8.30% | 3.90% | 17.85% |
| Malta | 15.29% | 9.57% | 7.35% | 11.20% |
| Netherlands | 11.22% | 13.64% | 2.36% | 14.33% |
| Austria | 7.70% | 4.46% | 3.79% | 6.28% |
| Poland | 2.02% | 0.51% | 1.72% | 1.40% |
| Portugal | 0.36% | 1.02% | 0.68% | 3.42% |
| Romania | 5.04% | 2.67% | 5.38% | 0.82% |
| Slovenia | 1.80% | 1.51% | 2.78% | 10.34% |
| Slovakia | 3.08% | 1.66% | 0.54% | 2.49% |
| Finland | 6.29% | 4.37% | 1.28% | 6.61% |
| Sweden | 8.59% | 8.42% | 2.40% | 8.86% |
Lower overall reporting does not apply to every problem
Buyers aged 65–74 reported at least one of eight named problems less often than those aged 16–24: 29.17% versus 36.55% in the EU aggregate. The older group’s percentage was lower in 20 of the 27 countries.[1]
Delivery shows a clearer difference between the age groups. Slower-than-indicated delivery affected 13.13% of older recent buyers and 22.13% of younger recent buyers. Complaint difficulties, by contrast, were reported by roughly 5.6% to 5.7% of each group. Those similar point estimates do not establish statistical equality.
A person is counted once for each problem they report, however many purchases or incidents lie behind that answer. Someone making one purchase and someone making ten purchases both enter the buyer denominator once. The results therefore cannot tell us that older buyers have fewer problems per order, shop more carefully or are more satisfied.
This distinction is useful when reading a store’s own reports too. Counts of orders, affected buyers, support conversations and reported problems answer different questions. Before comparing two periods or customer groups, decide which one is being counted and keep that basis consistent.
Overall problem reporting and two specific experiences
Percentage of recent online buyers in the stated age group, EU aggregate, 2025 survey. The combined measure covers eight named problems and excludes Other-only answers and the separate safety/health question.
| Age group | At least one of eight named problems | Delivery slower than indicated | Complaint or redress difficulties |
|---|---|---|---|
| Ages 16–24 | 36.55% | 22.13% | 5.59% |
| Ages 65–74 | 29.17% | 13.13% | 5.72% |
Among buyers reporting a problem, the route to help becomes more prominent
We then asked a different question: among buyers reporting at least one of the survey’s eight named problems, how many reported each specific problem? This calculation divides the specific-problem percentage by the named-problem percentage for the same age group. It changes the denominator, not the underlying experiences.[3]
In the EU aggregate, complaint or redress difficulties were reported by 19.61% of older buyers in this named-problem group, compared with 15.29% of younger buyers. Difficulties finding guarantee or legal-rights information were also more common within the older group: 25.09% versus 22.79%. The older percentage was higher in 20 countries for complaints and 21 for rights information.
Slower delivery showed the opposite balance: 45.01% of older named-problem reporters and 60.55% of younger ones reported it. The older percentage was lower in 23 countries. These are percentages of people reporting a named problem, not shares of all incidents or minutes spent handling them. One person may report several categories, so they should not be added together.
There is an important detail in the definition. Eurostat’s combined measure covers eight specified problems. It excludes people whose only answer was “Other”, and a separate question about safety or health problems. The calculation should not be described as the experience of every dissatisfied shopper. Within its stated group, it shows why a lower overall problem rate can coexist with a greater prominence of getting help.
The problem mix among buyers reporting a named problem
Percentage of buyers reporting at least one of eight named problems who reported the specified issue, EU aggregate, 2025 survey. This denominator differs from the other charts.
| Problem | Ages 16–24 | Ages 65–74 |
|---|---|---|
| Delivery slower than indicated | 60.55% | 45.01% |
| Complaint or redress difficulties | 15.29% | 19.61% |
| Difficulty finding guarantee or rights information | 22.79% | 25.09% |
A fictional lamp shop checks the journey from the post to the reply
Consider a fictional independent shop selling a desk lamp for €80, with €8 delivery to a particular destination. The owner is preparing social posts and asks a colleague to follow the purchase and support journey. This is a worked example of the review process, not a business observed in the survey.
The first check starts with the post. Its price is €80, but the colleague cannot find the delivery charge until checkout. The owner records the point at which the €88 total becomes visible and checks whether the post, product page and checkout give a consistent account. The decision is about information placement, not whether €8 is an acceptable delivery charge.
Next, the colleague compares the delivery promise with the order confirmation and the message a buyer would receive if dispatch were delayed. A clear promise needs a matching update when circumstances change. The review records which team member can correct the information and where that correction must appear.
For the third check, the colleague imagines that the lamp arrives with a cracked shade. They follow the published help route without sending a real complaint. The shop checks whether the buyer can identify the relevant order, describe the issue and understand the next step. It also checks that the person handling the message can access the information needed to respond.
The owner discovers that the product page links to help, while the order confirmation points only to the homepage. Updating the confirmation becomes a concrete task. That proposed improvement comes from the fictional review, not from applying a European percentage to this shop. The study supplies questions to ask; the shop’s own evidence decides what needs changing.
Nothing in the example requires a separate buying journey for each age group. The starting point is understandable information and a usable route to help for everyone, followed by testing with a range of customers. Any effect on sales or complaint resolution would need to be assessed separately.
A four-part store review worksheet
Review one complete buying and support journey
Follow the purchase
Compare the post, product page, checkout and confirmation. Record when the total and delivery expectation become clear.
Try the route to help
Check a wrong-item or damaged-item scenario and find the guarantee information. Record the next step a buyer can understand.
Assign and recheck
Give each demonstrated issue an owner. Review the changed page or message and confirm that the related information still agrees.
Use one copy of this worksheet for a specific product and destination. Record the page or message checked, what a customer can see, any discrepancy, the person responsible and the date for a follow-up. Work through the journey on a phone as well as a larger screen.
Price and charges. Start with the social post or advertisement, then follow its link to the product page and checkout. Write down the stated price, delivery charge and point where the total is clear. Note any condition that changes the amount. The useful output is a specific inconsistency or missing explanation, rather than a general instruction to “be transparent”.
Delivery and updates. Record the delivery expectation presented before purchase and compare it with the order confirmation. Identify the message used when an order is delayed and who can send or update it. Keep the customer’s promise distinct from internal processing milestones that they may not understand.
Problems with an order. Follow the help route from the product page, confirmation and delivery message. Check a wrong-item or damaged-item scenario. Record whether the buyer knows what information to supply, whether the message reaches the right person and what next step is explained. A visible contact link and a satisfactory response are separate things to review.
Guarantees and further help. Find the information a buyer would need about a guarantee and the available route when an initial reply does not resolve the issue. Check the wording against the business’s actual arrangements. Record missing information for the responsible person to resolve; this worksheet is an operational review, not a test of legal compliance.
Choose the next improvement from your own evidence
Start with a problem you can demonstrate in the journey, such as inconsistent delivery wording or an order email that makes help difficult to find. Give the change an owner and check the revised journey before treating the issue as resolved. The European figures help broaden the questions; they do not tell you which problem costs your business most.
Use existing customer messages and order records to understand what happened, keeping repeat conversations about one issue distinct from different affected buyers. A small collection of comments can reveal a confusing instruction without establishing how common the problem is. Conversely, a low complaint count does not establish that every customer found the process easy.
The practical takeaway is to examine both the purchase information and the response when a purchase goes wrong. Prices and delivery remain important. The age comparison adds a reason to look carefully at the less visible part of the experience: finding information, raising an issue and understanding what happens next.
How we analyzed the survey
We used Eurostat’s isoc_ec_iprb21 dataset for survey year 2025, retrieved October 17, 2026. The frozen extract was last updated by Eurostat on April 17, 2026. We selected PC_IND_BUY3, the percentage of people who purchased online in the previous three months, for the EU aggregate, all 27 member countries and six age groups from 16–24 to 65–74.[1]
The survey concerns private-purpose purchases through websites or apps. Its standard population is people aged 16–74 in private households. These results do not describe nonbuyers or people aged 75 and older. Reference periods are relative to the interview, and collection dates vary nationally; they are not a common calendar-year purchasing window.[2][4]
For the reversal, we compared I_BCPR1, final costs higher than indicated, with I_BCR1, complaint or redress difficulties. A country counts when the cost percentage exceeds the complaint percentage among 16–24-year-olds, while the complaint percentage exceeds the cost percentage among 65–74-year-olds. Each country contributes one observation to the count, regardless of population. We show all six age bands to make the endpoint comparison transparent.
For the conditional percentages, we divided each named-problem percentage by I_BARR2Y for the same country and age group, then multiplied by 100. I_BARR2Y is explicitly the union of E9a–h, excluding the Other answer and the separate safety/health question. All 486 required values across the 27 countries, two endpoint age groups and nine measures were present and carried no source flags.[2][3]
We allowed each published input to vary by 0.005 percentage points to test sensitivity to two-decimal rounding. All 15 reversal classifications remained. This checks rounding only. The aggregate extract does not provide the joint responses and survey-design information needed to calculate appropriate confidence intervals for these comparisons. We make no significance, causal or precise age-threshold claim.
National reports describe differences in collection and questionnaire implementation. Hungary notes possible limited effects from question ordering; its 2025 annex lists no E9 deviation. Ireland’s linked Eurostat report is older, so we checked the CSO’s 2025 background notes. The cross-country comparison remains descriptive. It does not adjust for purchase frequency, products, sellers or other differences between buyers.[4][5]
Earlier research already examines online shopping problems and demographic differences. Our contribution is a reproducible comparison of the 2025 age-specific ordering and problem mix across the EU, with the exceptions shown. We considered several problem categories before selecting this comparison, so it should be read as exploratory description. This article compares age groups within one survey year, not change over time.[6][7]
- Download the age and country comparisons (CSV)
- Download the analysis package (ZIP; English instructions)
The CSV includes published percentages and calculated conditional percentages for six age groups in the EU aggregate and all 27 countries, with source flags. Conditional percentages refer to buyers reporting at least one of eight named problems; that field is blank for the combined measure itself. Numeric fields use a decimal point.
Sources and notes
- Eurostat, Problems encountered by individuals when buying or ordering goods or services over the internet (isoc_ec_iprb21), survey year 2025. Unit PC_IND_BUY3; six age groups and EU member countries. Retrieved October 17, 2026; source update April 17, 2026. Calculated results use the published percentages, not a new respondent sample.
- Eurostat, 2025 model questionnaire, printed page 18, questions E9 and E10. E9 applies to people buying online in the previous three months and permits multiple answers. The accompanying aggregated-variable definitions, printed page 21, define I_BARR2Y as E9a–h. Other-only answers and the separate safety/health measure are outside that union. Both source documents are preserved in the analysis package.
- Influence IQ calculations: within each country, compare higher-than-indicated costs and complaint/redress difficulties at ages 16–24, then at ages 65–74. For the conditional mix, divide each problem percentage by I_BARR2Y for the same group and multiply by 100. Responses can overlap. Source-rounding checks allow ±0.005 percentage points per input; they do not measure sampling uncertainty.
- Eurostat, ICT usage in households and by individuals metadata, especially population, reference period, sampling error and geographical comparability. National reports linked there were checked for survey period and questionnaire deviations. The separate CIRCABC country-notes links were unavailable during this review. Published percentages may still have sampling and non-sampling errors even when no source flag is attached.
- Ireland CSO, Household Digital Consumer Behaviour 2025, Background Notes: 2025 model questionnaire, private-household coverage and three-month reference period. Its national publication covers people aged 16 and over; the article uses Eurostat’s 16–74 estimates instead. Hungary’s 2025 national report and linked questionnaire annex are retained in the analysis package.
- Kráľ and Fedorko (2024), Problems with Online Shopping in Selected Countries of the European Union. The public abstract describes comparisons of Germany, France and Spain using earlier Eurostat surveys. Our study addresses the stated 2025 age comparisons rather than claiming that analysis of shopping problems is new.
- Ivanova (2026), Examining Associations Between Socio-Demographic Characteristics and Online Shopping Risk Determinants of Consumers in Bulgaria. Demographic comparisons using official aggregate data have precedents; the analysis here is a descriptive comparison across all 27 EU countries, without significance tests.
- Eurostat permits reuse of its statistical data with attribution. Calculations, interpretation and translations are by Influence IQ. Eurostat is not responsible for these adaptations.
Calculations, interpretation and translations are by Influence IQ. Eurostat is not responsible for these adaptations.