The phyla classification problem
The highest level of classification within the animal kingdom are the 35 or so phyla, each of which is identified by a distinctive body plan which we can use to categorise all the known animals on the planet.
It is true that one of the features of the Cambrian and pre-Cambrian fossils is that they don't always seem to fit into the 35 or modern phya and classifying them can be difficult (Ref 2.) The question is why?
Stephen Gould explains that the concept of a cone of diversity which steadily widens over time is not actually very appropriate when looking at the history of life. What actually seems to happen is that there is a widening diversity of species until the planet is inevitably hit by an extinction event such as a dramatic temperature change or meteor strike. The creatures that survive the extinction then go on to repopulate the planet and the missing branches remain only as fossil imprints.
Thus most of the branches of life seen on the planet today have been pruned multiple times in the ancient past, so we see the very tips of the very few branches that managed to make it unscathed through the last 3 billion or so years. Unsurprisingly, our modern classification system is based on what survived rather than what could have been, so when we find fossils of creatures from the pruned branches it's not always obvious where they fit.
Another factor which will cause phyla to become more and more distinct over time is that when a beneficial mutation happens all the other features of the lucky recipient population will be given a boost at the same time.
Imagine three species of shellfish which have red, green and blue shells respectively, living in a sea where the temperature was dropping due to an incoming ice age. If the red shellfish gained a useful mutation which helped its metabolism cope with lower temperatures and as a result outcompeted the other two species to the point of their extinction then all the remaining shellfish would have red shells, even though the shell colour had little or nothing to do with their survival. Over time more and more of these features will build up to make each phyla more and more distinct from the others.
So when we try to classify creatures near the root of the tree of life then it is not surprising that they don't have all the distinct features which separate the modern phyla. If you go back far enough to species such as the halkieria then you end up trying to classify creatures which appear to have spawned multiple modern phyla i.e. the mollusc, the annelids and the brachiopods so how do you classify it?
Phylogenetic Hypotheses of the Relationships of Arthropods to Precambrian and Cambrian Problematic Fossil Taxa, Systematic Biology,
Vol. 45, No. 2, (Jun., 1996): 190-222