The problem with 'trust us'
In most digital systems, you trust the operator to report results honestly. For a vote, that is a weak foundation: the operator could make a mistake, be compromised, or be accused of bias even when honest. Verifiability replaces trust with proof.
What end-to-end verifiability means
End-to-end verifiability (often shortened to E2E-V) gives two guarantees. 'Recorded as cast' means each voter can confirm their ballot was captured correctly. 'Counted as recorded' means anyone can verify that the final tally is the correct sum of all recorded ballots.
Crucially, these checks reveal nothing about how any individual voted. Verifiability and secrecy are not in tension - a well-designed system delivers both.
How voters and observers check a result
Each voter receives a way to confirm their ballot is included in the final set, typically through a tracker that does not reveal their choice. Independent observers can download the encrypted ballots and the proofs and re-run the verification themselves.
Because the data and the method are public, no single party, including the vendor, has to be trusted for the result to be believed.
Why it matters for legitimacy
A result that can be independently verified is far harder to dispute. When a vote is close or politically charged, verifiability is what turns 'we counted and you have to believe us' into 'here is the proof, check it'.
International guidance on e-voting, such as the Council of Europe's recommendations, points organizations toward exactly this kind of transparency and evidence.