Bats on site. What happens now.

§1This is a separate legal regime

The first thing to understand is that protected species law sits alongside planning, not inside it. Planning permission does not authorise you to disturb a roost. A licence does.

That means two processes, two decision-makers, and two sets of tests — and it means a granted planning permission can leave you legally unable to start work.

It's also entirely separate from biodiversity net gain. The BNG exemptions that came in during summer 2026 changed nothing here. A site can be exempt from BNG and still have a bat roost that stops everything.

§2Season is the expensive constraint, not fees

People assume the problem is cost. It's usually the calendar.

A preliminary roost assessment can generally be done at any time of year. The activity and emergence surveys that follow — the ones that establish what species are using the building, how, and how often — are seasonally constrained, broadly to the warmer months, with the genuinely useful window narrower than that.

Miss the season and you don't pay more. You wait a year.

Which is why a positive preliminary assessment in early autumn is a materially worse outcome than the same result in March, and why an ecological walkover early in a project is worth far more than it costs. It's the cheapest piece of programme insurance available on a building with any age to it.

⏳ Confirm current survey windows with your ecologist. Guidance is periodically revised and species differ.

§3The three tests

Where works will affect a European protected species, a Natural England licence is needed, and three tests must all be satisfied:

One — purpose. The activity must be for preserving public health or safety, or for other imperative reasons of overriding public interest, which can include reasons of a social or economic nature.

Two — no satisfactory alternative. Not "no cheaper alternative" and not "no alternative we'd prefer". Whether there is a satisfactory way of achieving the objective with less impact.

Three — favourable conservation status. The action must not be detrimental to maintaining the population of the species at a favourable conservation status in its natural range.

Test three is the one that mitigation answers. Tests one and two are answered by the case you make for the development itself — which is a planning argument, not an ecological one.

§4Why the tests matter at the planning stage too

Here's the part that catches people, and it's the reason a planner should be involved rather than only an ecologist.

The local planning authority cannot simply grant permission and leave the licence to Natural England. In determining the application, the authority has to have regard to whether a licence is likely to be granted — which means the three tests are effectively in play at the planning stage as well.

So an application that treats the roost as a matter to be sorted out later is an application with an unanswered question in it. The evidence that satisfies Natural England and the evidence that satisfies the planning officer are largely the same evidence, and assembling it once, early, for both, is the efficient route.

§5What mitigation looks like in practice

It is more specific and more constraining than most people expect:

• Roost replacement — bat boxes or integrated bat bricks, in specified numbers, in specified positions, at specified heights and aspects.
• Lighting limits. This is the one that surprises clients. Commuting and foraging routes have to stay dark, which constrains external lighting design, security lighting, and sometimes glazing. It is a real design constraint, not a note on a drawing.
• Timing. Works to affected areas restricted to defined periods, avoiding the maternity and hibernation seasons.
• Supervision. Certain operations under ecological watching brief.
• Long-term securing. Usually by condition, sometimes by obligation — and it binds successors, which matters on a site you intend to sell.

§6What to do if a survey has just come back positive

Don't stop. The project is on a timetable, not over.

Establish the season position immediately. Whether you can survey this year or must wait decides your programme, and everything else follows from it.

Build the tests into the application. Purpose and alternatives are planning arguments and they need making properly, in the application, at the time.

Design the mitigation in, not on. Lighting in particular. Retrofitting a lighting strategy around a bat constraint after the scheme is fixed is how good schemes become bad ones.

Do not touch the building. Disturbing a roost without a licence is a criminal offence, and "we didn't know it was still occupied" is not a defence anyone should be relying on.