Morning Sarah

Difficult to answer "what is best" with any authority when it comes to risk assessments.

"What is best" depends on the circumstances and, often, organisational policies - where the policies may dictate what is done and thence is "best" but perhaps definitely not optimal, often because what happens is that people trying to push square pegs into round holes.

Why not think about this from a different direction?

A piece of paper headed "risk assessment" has never saved a life. If a brick is falling towards someone's head that piece of paper isn't going to make one iota of difference to the severity of injury.

But that piece of paper may say "a safety helment must be worn for this task" and if that mitigation is implemented and the person about to be struck is wearing a helmet then they may be unhurt or only sustain minor injury.

So, what actually matters, other than the legal nicety in the Code or often Codes of Codes Regulations that says "thou shalt do a suitable and sufficient risk assessment and [in most cases] thou shalt make a record of the significant findings", is what happens in terms of applying the appropriate set of mitigations that reflect a package of risks that apply to a task and the environment around that task.

So, you might have a broad risk assessment for an entire site and/or substantial part of the site which defines some common ground rules, then perhaps some generic or area risk assesments, then something very specific to what is happening for each task going on at any time.

But, what is important is doing what works for those at the scene.

....and whatever they say in their marketing spiel, no outsider can tell you what would work for your organisation. That takes an intimate understanding of how the organisation ticks and what it does and then, often learning from trial and error. Nothing wrong with making mistakes as long as one learns from them [preferably before something really bad happens]!