Most content teams judge articles by page views, time on page and scroll depth. Those numbers show which articles people opened and how far down they got. They say nothing about whether the reader got what they came for.
Below is one way to find out: what to ask, where to ask it, how to read the answers without fooling yourself, and what to do with each kind of article once you know.
The short answer
Ask the readers, on the article itself. Put one Yes / No question at the end of every article, such as "Was this article helpful?", and add one follow-up that asks for the reason. Once an article has at least 10 answers, compare its share of Yes answers with similar articles. The ones at the bottom are the ones to fix first.
Why traffic numbers can't answer the question
Traffic numbers measure attention. Whether an article helped is a different question, and the same number can point both ways:
- A how-to with a high bounce rate may have answered the question in the first paragraph, so the reader left happy.
- An explainer with a long time on page may have been read twice because it was confusing.
- An article that ranks well in search may be getting the visits while out-of-date details send readers back to the results page.
You can't tell these cases apart from a dashboard. The reader can, and it takes them one tap to tell you.
Decide what "helpful" means for each kind of article
Before you ask anything, write down what success looks like for each section of your site. A recipe helps if it worked in the reader's kitchen. A news story helps if the reader understood what happened. Use that to word the question. Readers find it easier to answer a question about the thing they just did than a generic one.
| Kind of article | What helpful means | Ask |
|---|---|---|
| News and analysis | The reader understood what happened and why | Did this article explain what happened? |
| Guides and how-tos | The reader finished the task | Did this guide solve your problem? |
| Reviews and buying guides | The reader could make a decision | Did this help you decide? |
| Recipes and tutorials | It worked when the reader tried it | Did this recipe work for you? |
| Help and docs pages | The reader found the answer | Did this page answer your question? |
Ask one question at the end of every article
Put the question after the last paragraph, where the reader has just decided whether the article was worth it. Keep the survey to three questions at most, in this order:
-
A Yes / No question
It takes one tap, so it gets the most answers, and it gives you a single figure per article that you can compare.
-
A follow-up that asks for the reason
A multiple choice question such as "What would make this article better?" with four or five options: more detail, clearer steps, more up to date, more examples, something else. Every reader sees it, so word it so it also makes sense after a Yes.
-
An optional comment
A free-text box last, for readers who want to say more. Readers who have already tapped once are much more likely to write something.
Tip: Use the same first question on every article in a section. If each article asks something different, you can no longer compare them.
Example
Example: a cooking blog with 300 recipes
Say you run a cooking blog. You add "Did this recipe work for you?" to the end of every recipe, followed by "What went wrong?" with the options "Timing was off", "Couldn't find an ingredient", "Steps were unclear" and "Didn't taste right".
A month later most recipes have a clear majority of Yes answers. Your bread recipes don't: they get more No answers than any other group, and most of those readers picked "Timing was off". That is a short, specific job: bake them again, fix the times and check the answers the following month. Page views would never have pointed you there.
Read the answers fairly
A handful of answers can make an article look much better or much worse than it is. A few habits keep the comparison honest:
- Wait for at least 10 answers before you look at percentages. Three answers are not "67% unhappy".
- Compare like with like: guides with guides, news with news. A breaking story and an evergreen guide are read in different ways.
- Read the reasons and comments before you act on the score. The score tells you where to look, the comments tell you what to change.
- Use the same date range for every article you compare, and compare the same number of weeks before and after you change one.
Sort your articles into four groups
Once your articles have enough answers, put each one in a group by two things: how helpful readers found it, and how much traffic it gets. Each group has its own next step.
| Readers answered | Lots of traffic | Little traffic |
|---|---|---|
| Mostly Yes | Keep it. Link to it from related articles and use it as the model for new ones. | Readers like it but few find it. Improve the headline and link to it from your busier pages. |
| Mostly No | Fix it first. This is where the most readers are let down. | Update it if the topic matters, merge it into a better article, or remove it. |
Where Asklet fits
Asklet does the setup above with one snippet in your article template. It shows a short survey at the end of every article, keeps the path each answer came from, and shows results per article and per section. You can put articles into themes to compare whole topics, and it shows counts until an article has 10 answers, so early results don't mislead you.
It fits best when you have dozens or hundreds of articles and want to compare them. It is anonymous by design: readers don't sign in and you can't contact them, so it won't help if you need to know who said what.
Your next steps
-
Write down what "helpful" means for each section
One line per section, such as "the reader finished the task" for guides.
-
Add one Yes / No question to the end of every article
Word it for the section, and use the same wording on every article in that section.
-
Add one follow-up that asks for the reason
Four or five options plus "Something else", then an optional comment.
-
Wait until most articles have 10 answers
For a busy site that can be a week; for a smaller one, a month or more.
-
Sort articles into the four groups
Mostly Yes or mostly No, with lots of traffic or little.
-
Fix the busiest unhelpful article first
Change what the comments name, then compare the same number of weeks before and after.
Try these steps on your own site. Asklet adds the questions to every page with one snippet. Free plan, no card needed. Compare the plans.
Start freeCommon questions
How many answers does an article need before I trust the result?
At least 10 before you look at percentages. For a bigger decision, such as rewriting a whole section, wait for more, and look for the same reason coming up in several articles.
Will a question at the end of an article annoy readers?
It shouldn't, if it sits at the end. It appears after they've finished reading, it doesn't cover the article, and answering takes one tap. Readers who don't want to answer just scroll past it.
What is the best question to ask?
A Yes / No question worded for the kind of article, such as "Did this guide solve your problem?" for a how-to. It gets the most answers and gives you one figure you can compare between articles.
Do readers need to accept cookies to answer?
Not with Asklet. The widget sets no cookies and uses no tracking identifiers, and readers answer without signing in, so it doesn't need a consent banner.
Related
More from the blog
- Privacy and law EU cookie consent changes: where the Digital Omnibus stands in October 2026 9 min read
- Privacy and law Session replay and pixel lawsuits: what website owners should know 9 min read
- Privacy and law How to make a feedback widget accessible, and what the European Accessibility Act asks 9 min read