Can phones replace access badges?
Where the access system was built for it, and there only. The readers decide rather than the phones, so the question belongs to your integrator rather than to Apple or Google. And a flat battery becomes a door somebody cannot open, which plastic never did.
The readers decide
A phone can behave as a card. Whether your door accepts what it presents is a property of the reader and the system behind it.
A system designed for phone credentials will have readers that expect them, an enrolment process, and a way to issue and revoke.
A system installed years ago will not, and no phone capability changes that. Retrofitting usually means replacing readers, which is the real cost of the project.
So the first question is not what phones can do. It is what your readers verify and whether the manufacturer supports phone credentials at all.
What each platform allows
Android exposes card emulation to applications directly, which makes phone credentials possible for a system that implements them.
iOS governs which credentials may be emulated through arrangements with the operator rather than leaving it open to applications, which is why some systems support iPhones and superficially similar ones do not.
That asymmetry decides feasibility for a mixed workforce. A solution that works on one platform and not the other is not a badge system, it is a pilot.
Ask the supplier which platforms they support, in writing, and what happens to staff whose phone is not on the list.
The failure mode plastic does not have
A badge with a flat battery is a person outside a door. Plastic has no battery and no software, and it fails only by being lost or damaged.
Phones are also updated, replaced, reset and repaired, each of which can end a credential at a moment nobody planned.
So a phone based system needs a documented answer for someone standing at a door with a dead phone, and that answer usually involves a physical card kept somewhere.
Which means the plastic does not go away. It becomes the fallback, in smaller numbers.
Running both
Most successful deployments run both rather than replacing one with the other.
Phones for daily convenience among staff who opt in, cards for visitors, contractors, staff without a supported phone, and anyone whose phone has failed.
That doubles the credential types to manage, which is a real administrative cost, and it removes the single point of failure.
Decide the fallback before the rollout rather than after the first person is locked out, because the first person locked out is how a phone based system loses its supporters.
Frequently asked questions
Can I copy my badge into my phone?
Not by installing an application. Most access systems check something the badge proves in hardware, and phone credentials are issued rather than copied.
Why do some buildings support phones?
Because their system was designed for it and their readers expect it. It is a property of the installation rather than of the phone.
What if someone's phone is flat?
They are outside the door. Any serious deployment keeps physical cards as a fallback and documents how to get one.
Is it cheaper than cards?
Rarely, once readers, enrolment and the card fallback are counted. The argument is convenience rather than cost.