An issue that frequently comes up is that more frontloading is needed for zoned sites.
I would frame the issue like that. Is any more frontloading required for a zoned site, where only the principle of development and number of units is in a local plan compared to avoiding costs on an appeal on an outline application refused despite being allocated in a conventional local plan.
Less – probably.
In order to be sure that the number of units in a draft allocation is ‘deliverable’ the LPA and the site promoter need to put in a minimum amount of work. If there are habitats issues they need to be sure conditions on SANG etc. can be delivered. They need to know how much of the site may be undevelopable because of protected species etc. They need to know the access to the site can deliver the number of units.
On a large site in outline only it is normal not to reserve the main site accesses in order to test and prove the site capacity. The highways authority should be giving advice on this anyway.
If you have allocated a site where the sequential flooding test applies you don’t need to do it again.
A minimum of habitats work needs to be done anyway as part of the AA of a development plan.
So some additional work is needed but no more than good practice in determining if a site is deliverable, and probably less in cost if an LPA refuses an allocated site and is awarded costs against them on an appeal.
Of course you could really front load costs and work with LPA funded masterplanning, but that is an optional extra, masterplans and design codes can and should be required by LPAs for large allocated sites prior to outline approvals come forward.
The concern I fear about frontloading comes from poorly resourced authorities who are unable to do the work on many small and medium size sites to demonstrate they are deliverable, and so are forced to allocate a large buffer of sites in the expectation that many of them will not be.