A company builds an FAQ page. Thirty eight questions, carefully written, accurate and kept reasonably current.
The same emails keep arriving. The team concludes that customers do not read, which is the wrong conclusion and an easy one to reach.
What actually happens is that a customer with one question opens a page of thirty eight, scans for four seconds, does not immediately see theirs, and sends the email. That is a rational decision about effort, and the page caused it.
Mistake 1: putting every answer in one place
Which optimises for the writer rather than the reader, since the reader has one question and is handed thirty eight.
Mistake 2: ordering by internal logic
Grouped by department or by product, which means nothing to somebody who does not know which department their question belongs to.
Mistake 3: writing the questions in company language
Visitors search in their own words, and a question phrased the way the business phrases it is invisible to somebody scanning for their own phrasing.
Mistake 4: burying it in the footer
The page exists to reduce contact and is placed where only people already looking for it will find it.
Mistake 5: keeping the answers away from the decision
A question about delivery answered on an FAQ page rather than on the product page is answered in the wrong place entirely.
Mistake 6: answering the questions nobody asks
Built from what the team thought people would want rather than from the actual contact record, which is the gap in website content that answers nobody's actual question.
Mistake 7: writing answers that hedge
Legal caution and vagueness, so a visitor reads the entry covering their question and still has to ask.
Mistake 8: letting it go stale
Prices, policies and processes change and the page does not, so it becomes actively misleading and generates corrections.
Mistake 9: using it as a substitute for contact
An FAQ presented as an obstacle before a contact route produces resentment rather than deflection.
Mistake 10: measuring page views
Which says nothing about whether anybody found an answer, and is why an ineffective page survives for years.
Mistake 11: never reading the questions that still arrive
The emails that keep coming are the precise measure of what the page fails to answer, and nobody categorises them.
Mistake 12: rebuilding it rather than replacing it
A better organised list of forty answers is still a list of forty answers, which is the format that caused the problem.
Mistake 13: assuming search will fix it
Site search helps people who know what to type, and most visitors with a vague concern do not.
Mistake 14: treating deflection as the only purpose
An FAQ built purely to stop emails is written defensively and reads that way. The same answers written to help somebody decide convert as well as they deflect, and the difference is entirely in the intent behind the writing.
Answer one question rather than displaying forty
An assistant takes a question in the visitor's own words and returns the one answer, on whichever page they were reading, and records the ones it could not answer so the underlying content improves. Build the starting set with the ai faq generator, read what arrives as in reading your chatbot conversations, and the deflection argument is set out in why your faq page is not reducing your emails. Start at try, see features, and the sector view sits under marketing agencies.