A household decides to switch, cancels the old provider on a Monday, orders the new one on a Tuesday, and spends nine days without broadband because the old line was released before the new one was activated. The saving was real. The gap was not planned for.
Switching is a sequence rather than an event, and the order in which the steps are taken determines whether there is a gap. Understanding the mechanics turns a nerve-racking fortnight into a routine admin task.
What is the correct order of steps?
Order the new service first, receive the activation date, and only then give notice to the old provider to end after that date. The overlap costs a few days of double billing and eliminates the gap entirely.
That order feels wasteful and it is the cheapest insurance available. A few days of duplicated charge is a fraction of the cost of a week without a connection.
Can two services run on the same line?
Not on the same physical pair at the same time for the same service, but the transition is managed by the network rather than by the household. In a proper switch, the new provider takes over the line on an agreed date and the old one ceases automatically.
That managed handover is what a one-touch switch is designed to do. Where it is not used, the household is responsible for the sequence and the timing.
What if the new provider needs a new line?
Then there is no takeover and the two services can coexist temporarily. That is the ideal case for avoiding downtime, because the new line is installed and working before the old one is cancelled.
It is also the case where the household may end up with two connections for a few weeks. That is a cost, and it is smaller than a gap.
How long does the whole process take?
Between two and six weeks. The order, the survey, the activation and the number transfer each take days, and they do not always run in parallel. A household with a fixed move date should start six weeks out.
Starting late is the most common cause of a gap. The providers are not slow individually; the sequence is simply longer than people expect.
Does the old provider need to be told?
Yes, and in writing, unless a one-touch process handles it. Telling them after the new service is active prevents the accidental early release that produces a gap.
Where a one-touch process is used, the new provider notifies the old one and the household should still confirm the cease date. The confirmation is the record that prevents a disputed final bill.
What happens to the phone number?
It can usually be ported, provided the order includes the request and the number is active. Porting adds a few days and occasionally fails for administrative reasons, which is another argument for starting early.
A number that is not ported is lost. That is a small disaster for a household that has had the same number for a decade, and it is entirely avoidable with one instruction at order.
What about the equipment?
The old router goes back, and the new one arrives. Where the technologies differ, the old equipment is useless to the new provider and must be returned within the deadline. That is a separate task from the service switch.
Packing the old equipment on the day the new service goes live is the habit that prevents a late-return charge. The two events belong together.
Can a switch happen mid-contract?
Yes, at the cost of the early termination fee. The new provider does not pay it, though some markets have rules about switching costs and some providers offer to cover part of it as a retention or acquisition incentive.
Where the fee is large, waiting for the term to end is usually the better decision. The comparison is the fee against the saving over the remaining months.
Does switching affect the credit file?
No, provided both accounts are settled. A closed account in good standing is a neutral record, and a switch is not a default. The exception is an unpaid final bill, which is a different matter entirely.
Households sometimes stay in a worse contract to avoid a hypothetical credit mark. Switching properly does not produce one.
What about a switch to a different technology?
Copper to fiber, or cable to fiber, involves new infrastructure and therefore a longer process and possibly a physical install. The old service continues until the new one works, which is the natural advantage of a technology change.
That advantage is worth using. A household moving to fiber can keep the old line running until the new one is tested, then cancel with confidence.
How should the old service be cancelled?
In writing, with the cease date stated, and with a request for confirmation. A phone cancellation is quicker and leaves no record, which is a poor trade when a final bill is disputed.
Keeping the confirmation until the final bill arrives and is checked closes the loop. The confirmation is the only document that proves what was agreed.
What if the new service is delayed?
Then the old one should be extended rather than ceased. That is only possible if the notice has not already been given, which is why the order of steps matters so much.
Where the old provider has already been told, asking to extend the cease date is sometimes possible and always worth attempting. It is easier before the date than after it.
Does a switch change the contract terms?
Yes, entirely. The new provider’s term, fees and conditions apply from the activation date, and nothing from the old contract carries over except the number and any equipment that was bought outright.
Reading the new contract as a new contract, rather than as a continuation of the old arrangement, is the discipline. The details that mattered last time will matter again.
How should the timing be planned?
Work backwards from the date the household needs the new service: activation date, minus the install, minus the order lead time, plus a margin. Then place the order and hold the notice until the date is confirmed.
A written plan with three dates on it is enough. Most gaps come from a plan with one date on it.
Is a gap ever acceptable?
Rarely, and only where the household can tolerate it. A week without broadband is a week of mobile data and missed calls, which is a real cost even if it is not a line item on any bill.
Where a gap is unavoidable, planning it is better than discovering it. Booking a mobile data allowance for the period is the usual mitigation.
What is the simplest way to switch safely?
Order the new service, wait for the activation date in writing, give notice to the old provider for the day after, keep both running for a few days, and return the equipment. Five steps, none of them difficult, in that order.
The sequence is the whole method. Households that switch in the other order are the ones who write about nine days without internet.
Does the new provider take over the old line?
Where the technology and the network allow, yes, and that is the fastest route. Where they do not, a new line is installed and the old one is cancelled afterwards, which is slower and safer.
Asking which route applies at the order stage determines the plan. The answer changes the timeline from weeks to days or the other way around.
What should be done with the old router during the overlap?
Keep it connected until the new service is confirmed working, then disconnect and pack it. A household that disconnects early loses the fallback exactly when it might be needed.
Leaving it connected for a few extra days costs nothing beyond the overlap billing. That is the cheapest insurance in the whole process.
Does the new provider need the old account number?
Usually, and it is the detail that most often stalls a switch. The number is on any bill, and a household that has gone paperless can find it in the old provider’s portal before cancelling access.
Taking a screenshot of the account page before the switch is a small precaution. Access to the old portal often ends on the day the service ceases, and the number is needed afterwards for the final bill.
What about a switch to a provider on the same network?
That is the simplest case: the line stays, the retailer changes, and the process is largely administrative. It can complete in days if the order is clean, and it needs no engineer.
Most switching advice applies to this case and it is worth recognising it. Where the network is the same, the physical risk is close to zero.
Switching is a handover rather than a leap, and the only thing that makes it feel like a leap is the order of the steps. If you switched tomorrow, which step would you take first?