In-network caching is a central aspect of Information-Centric Networking (ICN). It enables the rapid distribution of content across the network, alleviating strain on content producers and reducing content delivery latencies. ICN has emerged as a promising candidate for use in the Internet of Things (IoT). However, IoT devices operate under severe constraints, most notably limited memory. This means that nodes cannot indiscriminately cache all content; instead, there is a need for a caching strategy that decides what content to cache. Furthermore, many applications in the IoT space are timesensitive; therefore, finding a caching strategy that minimises the latency between content request and delivery is desirable. In this paper, we evaluate a number of ICN caching strategies in regards to latency and hop count reduction using IoT devices in a physical testbed. We find that the topology of the network, and thus the routing algorithm used to generate forwarding information, has a significant impact on the performance of a given caching strategy. To the best of our knowledge, this is the first study that focuses on latency effects in ICN-IoT caching while using real IoT hardware, and the first to explicitly discuss the link between routing algorithm, network topology, and caching effects.