This can be largely due to what you've included in your reference database. For example, if you lack outgroup taxa. In a nutshell, never make a database that only includes your taxa of interest.
For example, let's say you are interested in fungi. But your reference data only include fungi. Many of your sequences might classify as fungi, even to the phylum level, or not at all. But you'll likely find that many of these sequences are not fungi at all.
Or, there is a lack of representatives in your reference data. Depending on how you curate the reference database, you can consistently call one species over another. There are many factors at play when it comes to curating and using reference databases. Often biases in popular research organisms can be a problem, etc... See this post for more background.