So that's essentially one of the drivers for it, understanding where there are cultural issues and then putting wheels in motion to address. Even if the underlying cultural issues may be slow-moving, some of the action points were pretty straightforward - managers sitting with teams is something that can be changed overnight. There were some low-hanging fruit, even if some things are a slower burn. TBH the company culture was not bad in my opinion, had a good industry reputation, and it was ranked in the top10 best companies for work for in the UK three years running (top 5 when I left). You can take Glassdoor with a pinch of salt but it's not something that's easy to benchmark. The irony was, it was the best place I'd worked for culture and at the same time was the one doing the most to try and improve things. The places with weak culture weren't doing anything about it.
The formal process was annual and was used to identify changes YoY i.e. to understand direction of travel in key areas, and benchmark across business units, region, industry etc. It was checkpointed less formally (i.e. no empirical data) more frequently. To be honest, due to the inertia of analysing results, devising a plan, implementing a plan and then having enough window to see the impact of the changes, I don't think you could do it more frequently than 6 monthly. Being annual gave consistency in terms of alignment to pay reviews, bonuses, other cyclical annual factors (e.g. due to the nature of their work some departments might have particularly hectic periods) etc.
Regarding the positive growth, that can be true if you consider it a scale rather than an absolute. So e.g. if something is flagged as terrible in survey N and becomes mediocre in survey N+1, that's a positive change, but could still be called out in the next survey, because there is more to do. Our surveys didn't always show positive growth, there was a big drop off in my business unit one year which was probably what spurred the biggest call to action.