From a practical point of view, I've only ever seen Virgin Media cable and fibre optic to the premise achieve a high percentage of the maximum (e.g. 20 Mbps, 50 Mbps, and 500 Mbps). For measured bandwidths over copper phone cables, or fibre to the cabinet, 35 Mbps out of 50 Mbps is around the average I've seen. These number may be a lot better now especially with new hardware in cabinets and new cables, even if these are copper cables.
To achieve accurate benchmarking, it's standard practice to ensure network quiescence (e.g. not run a benchmark at the same time someone or the whole street is streaming from Netflix) so a measurement at between 0100 and 0400 is likely to ensure optimal numbers (realistically, just before bedtime or just when you wake up). This will give an estimate of the upper bound of what you can expect in practice and it could be disappointing or could indicate a bottleneck somewhere (perhaps in the provider's network) that could be addressed.
Usually the upload bandwidth can give a better hint of what is going on as it is less impacted by congestion. So if an upload bandwidth is able to hit a large percentage of the maximum, that's usually a good indication that things are reasonably ok (e.g. line quality is not abysmal). Loaded latency is also a good indicator of something bad, so if you see loaded latency of upwards of 600 ms you can just throw all the results away as a congestion issue. Unloaded latencies can be as low as 12 ms and for many people who are not downloading large files, low latency may be more impactful on the user experience (e.g. internet usage, gaming, and even responsiveness of streaming).
Edit: your percentage of the current link speed of 29593 kbps is not phenomenal but also not abysmal. 🤓
