The short answer
Ask the readers. Clicks and commissions show what readers did with your links, not whether the guide made their choice easier. Add a one-tap question, "Did this guide help you choose a gift?", then "Who are you shopping for?" and "How did you find this guide?". Read the answers per guide: a guide with many clicks and few "Yes" answers may be sending readers away to keep looking.
Gift guides take weeks to put together, and they are hard to judge. The affiliate dashboard ranks them by clicks and commissions, so the guide with the most clicks looks like the best one.
This post explains what those numbers miss, which three questions fill the gap, and how to read the answers so next year's guides are better than this year's.
What clicks and commissions can't tell you
Affiliate reports count what happened after a reader clicked a link. They leave out most of what decides whether a guide worked:
- Readers who found a gift idea in the guide, then bought it in a shop or somewhere you have no link.
- Readers who left because nothing in the guide suited the person they were buying for.
- Readers who clicked five links because no single pick convinced them. Five clicks look better in the report than one confident click.
- Where readers came from, when the visit arrives without a referrer and analytics files it as "direct".
Keep the affiliate numbers. They are the revenue. Asking readers tells you why a guide earned what it did, and what to change in the next one.
First question: did this guide help you choose a gift?
Make the first question a Yes / No. It takes one tap, so it gets the most answers, and it measures the one thing a gift guide is for. Ask it on every guide in the same words, so you can compare guides with each other.
Where you put it matters on a long guide. An inline survey after the last pick only reaches readers who scroll to the end, and readers who found their gift halfway down may stop there. A side tab stays in reach while they scroll, so they can answer as soon as they have decided.
Tip: Readers who answer "No" are the most useful ones. A short optional comment after it, such as "What were you hoping to find?", tells you which gift or budget the guide was missing.
Second question: who are you shopping for?
Ask it as multiple choice, with more than one answer allowed, because many readers use one guide for several people. Keep the options plain:
- Partner
- Parent
- Child
- Friend
- Colleague
- Myself
Across a few hundred answers this shows who each guide actually serves. A "gifts for him" guide used mostly by people buying for their dad is a hint for next year: a guide for dads may do better than another one for partners.
Third question: how did you find this guide?
More readers now start their gift search in an AI assistant. In a survey The Harris Poll ran for Forter, the fraud prevention company, in September 2026, 67% of 2,017 US adults said they planned to use AI for holiday shopping. A reader who follows a link from an assistant often arrives without a referrer, so analytics counts the visit as direct.
Ask the reader instead, with options in their words:
- Search engine (Google, Bing...)
- AI assistant (ChatGPT, Gemini, Perplexity...)
- Social media
- Newsletter or email
- Someone sent me the link
- I don't remember
The answers are self-reported and rough, so use them to see the split and how it changes, not to count visits. If most readers of a guide come from an AI assistant, they may arrive with a shortlist already, so lead with clear, specific picks.
Compare gift guide feedback across guides
Site-wide totals hide the difference between guides. Compare them side by side:
| What you see | What it suggests |
|---|---|
| Many clicks, high share of "Yes" | The guide works. Keep its format and update the picks next year. |
| Many clicks, low share of "Yes" | Readers are hunting. The picks may be too broad, or miss a budget. |
| Few clicks, high share of "Yes" | Readers choose here and buy elsewhere. Check whether your links cover the shops they use. |
| Few clicks, low share of "Yes" | Merge it into a stronger guide, or rethink who it is for. |
Wait for a few dozen answers on a guide before you read anything into it. A guide with eight answers can swing either way.
Example: a home and kitchen blog with five gift guides
Example
Say you run a home and kitchen blog. In November you publish five gift guides, from "Gifts for coffee lovers" to "Kitchen gifts under 30". You add the three questions to every page under /gift-guides/, with a side tab on each guide.
By mid-December, "Gifts for coffee lovers" has the most affiliate clicks, but only about half its readers answer "Yes". Their comments ask for picks under 20. "Kitchen gifts under 30" has fewer clicks and a much higher share of "Yes", and most of its readers say they are shopping for a parent.
Before the last delivery dates, you add three coffee gifts under 20 to the coffee guide. In your notes for next November you write down a guide for parents, and a budget filter at the top of every guide.
When to ask, and when to stop
-
Start on the day each guide goes live
The first weeks of answers are the ones you can still act on this year.
-
Stop after the last delivery date
Pause the survey on the guides when readers can no longer get a gift in time. Late answers mostly describe last-minute buyers.
-
Ask again next November with the same words
Same questions, same options, so you can compare this year with last year guide by guide.
Where Asklet fits
Asklet shows a short survey on your pages from one snippet, as an inline survey or a side tab. A path pattern such as /gift-guides/** puts the same survey on every guide, and results show how often each answer was picked per website, per page and per path, for the dates you choose. Multiple choice questions take 2 to 12 options, and you can let readers pick more than one. The Free plan allows 3 questions per survey, enough for all three questions above.
It sets no cookies and uses no tracking identifiers. It doesn't see your affiliate links, clicks or sales, so you put its answers next to your affiliate report yourself. Results are per question, so Asklet can't split the "Yes" answers by who readers are shopping for.
Your next steps
-
List your gift guides and their affiliate numbers
Clicks and commissions per guide are your starting point.
-
Add "Did this guide help you choose a gift?" to every guide
Yes / No, in the same words on each one.
-
Add "Who are you shopping for?" and "How did you find this guide?"
Multiple choice, with plain options and "I don't remember".
-
Use a side tab on long guides
Readers who find their gift halfway down can still answer.
-
Compare guides by clicks and share of "Yes"
Fix the guides with many clicks and few "Yes" answers first.
-
Pause after the last delivery date and ask again next November
Same questions, so this year becomes next year's benchmark.
Try these steps on your own site. Asklet adds the questions to every page with one snippet. Compare the plans.
Start freeCommon questions
How do you measure a gift guide? Use two sources together. Your affiliate network reports clicks, conversions and commissions per guide. A short question on the guide, "Did this guide help you choose a gift?", shows whether readers found what they needed, including readers who bought somewhere you have no link.
Where should the question go on a long gift guide? In a side tab, so it stays in reach while readers scroll, or inline after the last pick if most readers read to the end. Readers who find their gift halfway down may stop there, so an inline question at the end can miss them.
Should I ask readers about their budget? Yes, if your guides are organised by price. Ask it as multiple choice with ranges such as "Under 20" or "20 to 50", never as an exact amount, and keep it to the second or third question.
Do I need a cookie banner to ask these questions? Not for Asklet. The widget sets no cookies and uses no tracking identifiers. Its only browser storage is a short "already answered" note so the same reader isn't asked again on that page. Your affiliate links and analytics may still need consent, depending on what they set.