Most of the attention on AI in tendering goes to the bid document. But the bid is private - the evaluators see it, and at best a debrief letter comes back. The clarification window is the exception: on eTenders, every question a bidder asks and every answer the authority gives is published to all registered parties. It is the one place the behaviour of the whole bidding market is visible in writing, at scale - and it has changed.
We read the full clarification record of 468 Irish public tenders from 2023 to 2026 - 8,052 question-and-answer pairs, coded against a written codebook. The full study is published separately; the results below are what a bidder should do about it.
What changed
The question got sharper, not more frequent. The median tender attracts about the same number of questions as in 2023. What rose is a particular kind of question: one that cites the clause, states the bidder’s own reading, and asks the authority to confirm or change a rule of the competition - from 28% of all questions in 2023 to 34% in 2026, and from 20% to 36% in software and IT tenders. Naive questions a bid professional would not need to ask stayed flat at 4-5% throughout. The window is not filling with novices; it is filling with questions that read as if a lawyer drafted them. Questions carrying visible AI artefacts went from roughly one in a thousand to one in twenty-five - and what those artefacts look like is a post of its own.
Buyers hardened rather than improved. Substantive answers drifted from 84% to 80% of responses; “refer to the RFT” deflections rose from 6% to 10%. The sharpest movement is concession: where a bidder asked for a requirement to be changed, the authority conceded in 42% of cases in 2023 and 19% in 2026. Hard clarification cut-offs, refusals of late questions, and “the Authority is satisfied that the documents are clear” are now standard responses to requests that would once have produced a corrigendum.
Volume moved to the tails. One tender in three now attracts twenty or more bidder questions, against one in five in 2023, and the largest run past a hundred. A minority of competitions have become documented pre-award negotiations of the rules - conducted on the record, readable by every competitor.
What this means for your clarification strategy
Precision still pays - but it no longer distinguishes you. A question that cites the clause and states a reading is answered substantively more often than a vague one. That is worth having. But when a third of the field can generate the same precision at near-zero cost, the question itself is no longer a signal of a serious bidder.
The concessions that remain go to the specific and proportionate. At a 19% concession rate, “please lower the threshold” is close to dead. What still moves authorities is a request that names the clause, quantifies the mismatch - a turnover floor several multiples of the contract estimate - and proposes a specific, proportionate alternative. Ask for one thing, precisely, with a reason grounded in the competition’s own numbers.
Ration the window. A hundred-question submission on a mid-sized tender is now more likely to meet a hardened authority than an amended one - and every question you ask is published to the field, revealing your reading of the pack and sometimes your bid strategy. The advantage has moved from being able to write the question to knowing which question is worth asking. Three questions that matter beat thirty that fill a template.
Remember who reads the register. Competitors mine your questions; authorities remember high-volume askers; and your own residue - template batches, pipe-delimited headers - is on permanent public display. Treat every clarification as a published document, because it is one.
The takeaway
The clarification window has become a crowded, hardening channel where precision is table stakes and volume is counterproductive. Spend your questions where the answer changes your bid decision or your price - and put the effort you save into the bid itself, where evidence still breaks ceilings.
Know what is worth asking - and what your bid must answer for itself. BidReview audits your finished response against the RFT before submission, including the ambiguities worth a clarification and the ones your answer should simply resolve. Run the free scorecard →
