Italian Retail Marketplaces Spend €180,552 on Instagram With No CPM or Format Records

Italian retail marketplaces spent €180,552 on Instagram sales advertising in the 90 days ending September 17. Yet the cohort's most consequential metric may be the one that is absent: not one of its 723 ads has a recorded CPM, and every ad is classified as an unknown format.[1][3]

That combination makes it difficult to assess where money went, which creative formats were used, or whether the operation was buying media efficiently. The spend is observable. The format-level learning is not.

A fragmented account structure

The cohort includes 723 sales-objective ads, 71 ad accounts, and just seven brands.[1] That works out to roughly 10 ad accounts per brand.

The data does not establish why this structure exists. Multiple accounts can reflect separate sellers, regional teams, business units, agencies, legacy account setups, or other operational choices. But it does indicate a distributed buying environment, one in which consistent naming, creative taxonomy, and reporting governance are especially important.

The total spend is meaningful, but it is concentrated unevenly across ads. Over the 90-day period, the distribution was as follows:[2]

Half of all ads spent less than €13.31 during the full measurement window, while one quarter spent less than €0.70. At the other end of the distribution, overall cohort spend reached €180,552.08.[1]

These figures point to a long tail of very low-spend ads alongside a smaller group of ads that accounts for much of the budget. If every one of the 723 ads had spent exactly the 75th-percentile amount, total spend would have been roughly €68,850, well below the actual total. Some ads therefore carried far more spend than the typical ad.

A long tail is not inherently inefficient. It can be deliberate when teams are testing creative variants, localized offers, catalog combinations, or audience segments. But low-spend testing needs clear rules: what counts as a test, what performance signal advances it, and when an ad should be stopped. Without those rules, high ad volume can make reporting noisier without producing comparable learning.

Reach data raises a measurement question

The median ad recorded 26,957 in total EU reach, while 704 ads had a reach-per-day measure available.[3] However, the 25th percentile, median, and 75th percentile for reach per day were all zero.[3]

That does not necessarily mean the ads reached no one on a typical day. It does mean the available daily-reach field is not providing a useful basis for interpreting delivery cadence across most of the cohort. The contrast between nonzero total reach and zero-valued daily reach percentiles should be treated as a data-quality or metric-definition question before it is treated as a media-performance conclusion.

There is a related arithmetic anomaly. Dividing the median spend of €13.31 by the median total EU reach of 26,957 produces an implied figure of roughly €0.49 per thousand people reached. This is not a CPM. Reach is not impressions, and the ratio of two medians is not the median of individual ad-level ratios.

Still, it is notably below the available Italian Instagram CPM benchmarks: €2.48 for feed carousels, €3.36 for Reels, and €5.08 for Stories.[5][6] The difference could arise from several factors, including incompatible definitions, different time periods, frequency, organic or non-paid reach in the reach field, or data-processing differences. The cohort data cannot resolve the issue because CPM is missing for every ad.

The format and CPM blind spot

Every ad in the cohort is labeled unknown in the format field. The data also reports zero ads with a CPM value.[3][4]

This is more than a reporting inconvenience. Format is a core dimension of paid-social analysis. Without it, operators cannot compare the economics of Reels, Stories, carousels, and other placements within the cohort. They cannot identify whether higher spending concentrated in a specific creative type. They also cannot tie cost and delivery patterns to the content formats being produced.

Italian Instagram benchmarks show why that distinction matters. Stories have the highest CPM among the relevant benchmark rows at €5.08, compared with €3.36 for Reels and €2.48 for feed carousels.[5][6]

The engagement-value coefficients also vary by format. In the benchmark data, a share is valued at €0.80 for Stories and €1.20 for Reels. Reels include a €0.50 save value, while Stories have no save value listed.[5][6] These coefficients should not be interpreted as universal outcomes for every campaign, but they reinforce the point that formats carry different cost and engagement profiles.

If format is unknown, a team cannot determine whether it is paying a premium for Stories, benefiting from lower-cost carousel inventory, or mixing placements in a way that changes performance. A blended total conceals those decisions.

Organic activity is also uneven

The organic Instagram cohort covers 13 brands and 1,941 posts during the same 90-day window.[7] Median posting volume was five posts per brand, while median engagement rate was 1.22%. Median views per post were 268, and median interactions per post were three.[7]

The gap between median and 75th-percentile results is substantial: views rise from 268 at the median to 7,965 at the 75th percentile, while interactions rise from three to eight.[7] That pattern suggests that organic outcomes are unevenly distributed, with a minority of posts generating much stronger results than the typical post.

Organic performance alone will not explain paid-media efficiency, particularly because the paid and organic cohorts differ in brand coverage and unit of analysis. But sparse posting and variable post performance make clean paid-media measurement more important, not less.

Three practical actions

1. Repair format taxonomy first. Require a usable format label for every new ad, then backfill historical records where possible. Even partial coverage is materially more useful than an all-unknown field.

2. Restore CPM reporting and validate definitions. Confirm whether impressions, reach, spend, dates, and currency are consistently populated. Reconcile the reach-per-day calculation before using it to judge delivery behavior.

3. Define the role of low-spend ads. Separate deliberate tests from inactive or abandoned ads. Establish a minimum data threshold, a promotion rule, and a stop rule so that the long tail produces actionable evidence.

The central finding is straightforward: Italian retail marketplaces are operating Instagram sales campaigns at significant total spend, across a highly distributed account structure, without the format and CPM data needed for meaningful format-level optimization. Fixing that measurement gap should come before drawing strong conclusions about creative efficiency or scaling budget.