Booking systems will not sell a connection below a defined minimum, and that figure changes from airport to airport. It is a measured quantity rather than a general rule of thumb.
The figure is built from walking and processing
A minimum connecting time reflects how long it takes to leave an aircraft, move between the relevant gates, pass any required checks and reach the next boarding process.
Airports and airlines assess this for specific combinations, so a connection within one terminal has a different figure from one crossing to another building.
Where a train or bus links terminals, its frequency becomes part of the calculation, and a shuttle every fifteen minutes adds meaningfully to the figure.
International connections add steps
A connection that involves entering a country requires immigration and customs processing, collecting bags and checking them again, then clearing security a second time.
Each step has a queue whose length varies with the arrival bank, which is why these minimums are considerably longer than domestic ones at the same airport.
Connections that stay within a secure transit area avoid most of this and can be much shorter.
Baggage sets a separate limit
Bags must be moved between aircraft by the baggage system, and that transfer has its own timing, particularly where the two flights are handled in different parts of the field.
A passenger can sometimes make a connection that their bag cannot, which is why bags occasionally arrive on a later service.
Where two airlines do not have a baggage agreement, the transfer may not happen at all, and the passenger must collect and recheck.
The minimum is not a comfortable target
A minimum connecting time assumes an on-time arrival, an available gate and normal processing. It contains little allowance for anything going wrong.
Booking at exactly the minimum means a modest delay breaks the connection, and on a thin route the next available service may be the following day.
Protection depends on how it was booked
When two flights are sold on one ticket, the airline generally takes responsibility for reaccommodating a passenger who misses the second because of a late first flight.
Two separately purchased tickets are unrelated contracts, and the second airline has no obligation created by the first flight's delay.
The practical consequence is that self-connections need far larger buffers than the published minimum, because the risk sits with the traveler.