Habitat classification schemes provide tools for harmonized mapping, monitoring and assessment of habitats across regions. They also offer ways to simplify large biodiversity datasets to reveal main environmental and biological characteristics of a region, which might be sufficient level of detail for example in regional planning processes. Since 2013, HELCOM underwater biotope and habitat classification system (HELCOM HUB) has provided a framework to classify the marine underwater nature of the Baltic Sea, but so far, its functionality in describing the variation in Baltic Sea biological communities has only rarely been tested. We tested the functionality of HELCOM HUB in describing variation in rocky shore communities on a large scale, across the Finnish marine area. We found the classification tool very useful in simplifying complex biodiversity data and in creating quantitative presentations on community variation. The results show how the proportional occurrences of different rocky shore communities change in relation to each other along the environmental gradients of the northern Baltic Sea: along the coast in different salinity regimes, from sheltered archipelagos to the open sea and when going from shallow sublittoral to deeper waters. Although the importance of regional habitat classification schemes is recognized, we found also some weaknesses in HELCOM HUB. The red algal communities that are generally recognized as key components of northern Baltic Sea rocky shores were clearly “lost in classification”, although they were shown to be both common and to occur in relatively high coverages, especially in the southwestern part of the study area. This was mainly due to division of red algae into many “sub-groups” at levels 5 and 6 in the classification (e.g. perennial foliose red algae and perennial non-filamentous corticated red algae) that led to their “fragmentation” within the classification. This further resulted in low coverages of red algae within the sub-groups, and finally, in their classification to other classes such as blue mussels or other algal groups. The result highlights the need to consider the restrictions of any classification system when classified data is used in management contexts. When taxa are “lost in classification”, we might not just ignore species or communities, but also key ecosystem functions.