Every category draws a different size of company
The middle of one category sits entirely below the bottom quarter of another. Rank the categories by the median size of the companies shopping in them and the range is not subtle: at one end are categories whose in-market buyers have thousands of employees; at the other, categories whose median buyer has fewer than 50.
Median employee count of the companies in market for each category. Highlighted rows are the four examined in detail below.
Medians can hide overlap, so the quartiles are worth stating: three quarters of the companies shopping for subscription management are smaller than a quarter of the companies shopping for a CDP. The distributions are substantially separated, and a benchmark drawn from one tells you very little about the other.
Inside a category, the leading product changes with company size
In all four categories examined, the most-rated product among smaller companies is a different product from the most-rated one among larger companies. This is the part that matters when you read a category page: the category is one label, but the product leading the ratings changes depending on which end of the company-size range you examine.
| Product | Avg rating | Which end of the size range it leads |
|---|---|---|
 Shopify |
8.1 | Leads among smaller ecommerce-platform buyers, by a wide margin. |
| 7.7 | Leads among smaller buyers in two categories at once — email platforms and CDPs. | |
| 7.6 | Well represented among smaller reviews buyers and almost absent above the line. | |
| 6.9 | Leads among larger reviews buyers — the one category here where the larger end carries the higher rating. | |
| 6.8 | Leads among larger CDP buyers, with no ratings at all from companies below the line. | |
| 6.8 | Leads among larger email-platform buyers. | |
| 6.7 | Leads among smaller reviews buyers. | |
| 6.6 | Leads among larger ecommerce-platform buyers. |
The separation can be close to total in the ratings data. Salesforce Data Cloud has no ratings at all from companies below the line; Okendo, which is well represented among smaller reviews buyers, is almost absent above it.
The counter-current: the obvious readings are both wrong
Two conclusions suggest themselves from the table above, and neither survives checking. The first is that smaller companies are harder to please: across the whole corpus they rate their software slightly higher than larger companies do, not lower. The second is that enterprise software is simply worse: within individual products the direction is mixed, with smaller companies rating some above their larger peers and some below. Both come from comparing groups that are not rating the same mix of products.
What that means for a rating
A category average is weighted by whoever is in market, and that is a different crowd in every category. Because the two size groups are often rating different mixes of products, an average blends experiences from materially different buyer populations — which makes it a reasonable summary of a category and a poor way to compare across categories.
It also cuts against the neatest version of the story. In three of the four categories here the product leading among smaller companies carries the higher published rating — but in reviews and ratings it is the other way round, with Bazaarvoice above Yotpo. Four categories is a thin base for a rule, and the direction already breaks in one.
Want this read against your own stack?
Get my read →What this means for your shortlist
Ask who is behind a benchmark before you let it narrow your options. Our read, not a practice buyers describe, is that the useful move is a filtering one. When a category page, an analyst grid or a peer recommendation tells you what most companies use, the question worth asking is which companies — because at the extremes of this data, the answer is a population you may share almost nothing with.
Two practical consequences follow. A product that leads the ratings at the other end of the size range is not automatically wrong for you, but its score may primarily reflect experiences from companies operating at a different scale. And a category average that looks disappointing may be carrying the weight of buyers unlike you — which is an argument for looking at the products inside the category rather than the number on top of it.
For the pressure pushing stacks in the other direction, see the great consolidation; for where shortlists come from before any of this applies, how software buyers actually find vendors.
Common questions
Does company size affect which software you buy?
In Alium's interviews it is one of the clearest dividing lines there is. Each category draws a characteristic size of buyer, and the range across categories is wide: the median buyer in market for subscription management has tens of employees, while the median buyer in market for a CDP or an ecommerce platform has thousands. The distributions are substantially separated — three quarters of subscription-management buyers are smaller than a quarter of CDP buyers. Within each of the four categories examined, the most-rated product among smaller companies differs from the most-rated product among larger companies.
Why do software ratings differ between small and large companies?
Part of the difference is compositional: smaller and larger companies are often rating different mixes of products. In email platforms, for example, ratings from companies under about 500 employees are dominated by Klaviyo, while ratings from larger companies are dominated by Salesforce Marketing Cloud — two products with different published scores. The tempting conclusion, that smaller companies are systematically harder or easier to please, does not survive the check: within individual products the direction is mixed, and across the whole corpus smaller companies rate their software slightly higher rather than lower. So a category average blends experiences from materially different buyer populations.
Is a category's average rating useful?
It is useful for the category and less useful as a comparison between categories. An average is weighted by whoever happens to be in market, and since categories attract very different sizes of company, two averages are rarely describing the same kind of buyer. Our read, not a rule buyers state, is to treat a category average as a starting point and then ask which products inside it are actually rated by companies at your scale — because in several categories the most-rated product at one end of the size range is barely present at the other.
Do enterprise tools rate worse than tools for smaller companies?
Not reliably, and it is worth resisting the pattern. In three of the four categories examined here the product that leads among smaller companies carries the higher published rating, but in reviews and ratings the opposite holds: Bazaarvoice, which leads among larger companies, is rated above Yotpo, which leads among smaller ones. Four categories is a small base for a rule, and the direction already breaks in one of them.
Get this research made for your stack
A category average blends companies that look nothing like each other. Do a 15-minute interview and get the read for your size and vertical — what peers actually run, and what they found out afterwards.
Get my personalized read — 15-min interviewNo password · your interview is anonymized before it ever informs a page like this one.