How to Use Knowledge Base Search Analytics

Search Analytics shows what visitors type into the SoftwareRoad Knowledge Base search box.

It helps you understand what people are trying to find, which searches are successful and where your documentation may be missing important information.

Instead of guessing which articles visitors need, you can use real search activity to guide your next updates.

Open Search Analytics

To view the search report:

1. Sign in to your WordPress dashboard.
2. Open Knowledge Base > Analytics.
3. Select the Search Analytics section.
4. Review the recorded search terms.
5. Look for repeated searches and terms returning no results.

Search information will only appear after visitors use the SoftwareRoad Knowledge Base search box.

What Search Analytics can show

Depending on the current plugin version, the report may include:

• Search term
• Number of times searched
• Number of results returned
• Date last searched

For example:

Search term	Searches	Results
product key	42	3
change logo	19	2
licence not recognised	14	0
article colours	8	4

This gives you a quick view of what visitors want and whether your current articles are being found.

Repeated search terms

When the same wording is searched more than once, SoftwareRoad Knowledge Base can group it into one report entry.

For example:

product key

may show:

42 searches

rather than appearing as 42 separate rows.

This makes it easier to identify popular subjects.

Search wording may vary

Visitors do not always use the same language as your documentation.

For example, different visitors may search for:

• Product key
• Licence key
• Activation key
• Serial number
• Activation code

These may all refer to the same subject.

Review similar terms together rather than treating each one as a completely separate topic.

Searches with results

A search returning results means the knowledge base found one or more matching articles.

This is a useful start, but it does not guarantee that the results were helpful.

Run the search yourself and check:

• Whether the best article appears
• Whether the article title is clear
• Whether the excerpt matches the search
• Whether unrelated articles appear
• Whether the most useful result appears near the top

A search can technically return results while still being difficult for visitors to use.

Searches with no results

No-result searches are particularly useful because they reveal gaps.

A search may return no results because:

• No article covers the subject
• The article uses different wording
• The article title is too vague
• The article is marked as unlisted
• The article is still a draft
• The search term contains a spelling variation
• The visitor searched for a feature that does not exist

Repeated no-result searches should be reviewed first.

Investigate a no-result search

When you find a repeated search returning no results:

1. Run the search yourself.
2. Decide what the visitor probably meant.
3. Check whether an existing article covers the subject.
4. Review the article title, excerpt and wording.
5. Add the visitor’s wording naturally if it is relevant.
6. Create a new article if the subject is genuinely missing.
7. Test the search again.

Do not create a new article for every unusual one-off search.

Look for repeated or clearly useful topics.

Improve an existing article

Sometimes the right article already exists but search cannot connect it to the visitor’s wording.

For example, you may have an article called:

Where to Find Your Purchase Email and Product Key

but visitors repeatedly search for:

activation code

You could add a sentence such as:

Your product key may also be referred to as an activation key or activation code.

This helps search recognise the alternative wording without changing the main terminology used by the product.

Improve article titles

Article titles have a strong effect on how easily visitors recognise search results.

Compare:

Licence Information

with:

Where to Find Your Purchase Email and Product Key

The second title is clearer and more likely to match what a visitor needs.

Good titles normally explain:

• The task
• The setting
• The problem
• The expected result

Avoid titles that are too broad.

Improve article excerpts

The excerpt appears with the article in several knowledge base layouts and may also help explain why a result is relevant.

For example:

Learn where to find the purchase email and product key needed to activate SoftwareRoad Knowledge Base.

A good excerpt gives visitors enough information to choose the right result.

Keep it short and specific.

Use visitor wording naturally

Search Analytics can reveal the language customers actually use.

For example, your documentation may say:

reading progress bar

while visitors search for:

scroll bar

You could explain the alternative wording naturally:

The reading progress bar, sometimes described as an article scroll indicator, shows how far a visitor has moved through the page.

Do not add long lists of keyword variations purely to influence search.

Use wording only where it genuinely helps understanding.

Correct common spelling variations

Visitors may search using spelling mistakes.

You do not need to add every misspelling to your article.

However, a common variation may be worth considering when it reflects genuine confusion.

For example:

• licence
• license

SoftwareRoad Knowledge Base documentation may use UK spelling, while some visitors use US spelling.

Including both naturally in suitable places can improve search without making the article awkward.

Create a new article when needed

A new article is usually appropriate when:

• The subject is not covered anywhere
• The search appears repeatedly
• The topic needs a full set of instructions
• Adding it to an existing article would make that guide too broad
• It is a common support question
• Visitors are opening unrelated articles instead

For example, repeated searches for:

move licence to new website

may justify a separate migration guide rather than adding a large section to the general activation article.

Do not create duplicate articles

Before creating a new guide, check whether the subject already exists under another title.

Duplicate articles can:

• Split search results
• Confuse visitors
• Repeat instructions
• Become inconsistent over time
• Compete for the same search wording

Improve or rename the existing article when possible.

Compare searches with article views

Search data becomes more useful when compared with article views.

For example:

• A term is searched frequently
• The related article has very few views

This may indicate that the article is not appearing clearly in results.

Check:

• The title
• The excerpt
• The category
• The article status
• Whether it is unlisted
• Whether search can match the wording

Compare searches with article feedback

A search may return the correct article, but the article may still receive poor feedback.

For example:

• Search term: licence not recognised
• Results: 1
• Article feedback: mostly No

This suggests that visitors are finding the article, but it may not fully solve the problem.

Review the guide for:

• Missing steps
• Outdated instructions
• Unclear wording
• Incorrect menu names
• Missing error details
• A weak troubleshooting section

Popular searches with several good results

A commonly used search returning useful results may be a good candidate for the homepage Popular Searches section.

For example:

• Installation
• Product key
• Create category
• Search Analytics
• Licence error

Only add terms that consistently lead to helpful results.

Test each one before placing it on the homepage.

Searches returning too many results

A broad search may return several loosely related articles.

For example:

settings

could match a large part of the knowledge base.

You can improve this by:

• Using more specific article titles
• Writing clearer excerpts
• Splitting overly broad articles
• Removing unrelated wording
• Encouraging more focused popular search terms

A visitor searching manually may still use a broad term, but your article structure should make the best result easy to identify.

Searches returning the wrong articles

If an unrelated guide appears, check why the search phrase is present in that article.

It may appear in:

• The article title
• The excerpt
• A heading
• The article content

Remove or rewrite unrelated wording where appropriate.

Do not remove useful content simply to alter search results.

The article should remain clear for the reader.

Check whether the article is unlisted

Unlisted articles do not normally appear in knowledge base search.

If a useful guide is missing:

1. Open Knowledge Base > All Articles.
2. Edit the article.
3. Check whether Unlisted Article is enabled.
4. Disable it if the article should be publicly discoverable.
5. Select Update.
6. Test search again.

Keep an article unlisted when it should only be available through its direct URL.

Check the article status

Search normally uses published knowledge base articles.

A guide will not appear publicly if it is:

• Draft
• Pending Review
• Private
• In the Trash

Publish the article when it is ready for visitors.

Search Analytics and compact search

Searches from both the main knowledge base search and compact search areas can contribute to the report.

This means visitors may search from:

• The knowledge base homepage
• Category pages
• Article pages

The report focuses on the search wording rather than where the search box appeared.

Search minimum length

SoftwareRoad Knowledge Base requires a search term to contain at least two characters before running a search.

Very short input may not produce results or be recorded.

Useful search phrases are normally more specific than a single letter.

Search result limit

The live search area displays a limited number of matching results so the dropdown remains manageable.

SoftwareRoad Knowledge Base can show up to six live results.

If more articles match, visitors may need to use more specific wording.

Clear article titles help the most relevant options stand out.

Review search terms regularly

A practical review schedule could be:

• Weekly after launching a new product or feature
• Monthly for an established knowledge base
• After a major plugin update
• After publishing several new articles
• After customers report difficulty finding help

Search behaviour changes as the product and documentation grow.

Look for seasonal or update-related searches

Search activity may change after:

• A new plugin release
• A licence system change
• A redesigned settings page
• A new feature launch
• A WordPress update
• A temporary technical issue

A sudden increase in one search term may indicate that visitors are experiencing the same problem.

Review related support messages and plugin changes before creating new content.

Search terms can reveal unclear feature names

Visitors may search for a feature using a different name from the one shown in the plugin.

For example, they may search for:

article menu

when the plugin calls it:

Table of Contents

You can improve the article by explaining the feature in more familiar language while keeping the correct official name.

For example:

The Table of Contents creates a clickable article menu from your headings.

Do not copy personal search information into articles

If a visitor enters personal or account-specific information into search, do not copy it into documentation.

Examples include:

• Email addresses
• Complete product keys
• Names
• Payment references
• Account numbers

Search Analytics is intended to identify documentation topics, not collect customer records.

Handle any sensitive entry according to your website privacy and data management procedures.

Review unusual search terms carefully

A strange search may be:

• A spelling mistake
• A pasted error message
• A customer-specific question
• Automated traffic
• An unrelated search
• A one-off event

Do not restructure the knowledge base around one unusual entry.

Prioritise repeated terms and clear patterns.

Keep a list of article ideas

Search Analytics can become a useful documentation planning tool.

Keep a simple list containing:

• Search term
• Number of searches
• Current result
• Planned action
• Article to update or create

For example:

Search term	Action
move licence	Create migration article
serial number	Add wording to product key guide
change article colour	Improve design article title
search empty	Create search troubleshooting guide

This makes it easier to turn analytics into practical work.

Check whether your changes worked

After updating or creating an article:

1. Run the original search term.
2. Confirm that useful results appear.
3. Check the result title and excerpt.
4. Open the article.
5. Make sure it answers the search clearly.
6. Review future Search Analytics entries.

The goal is not only to produce a result, but to produce the right result.

New articles may take time to build views

Once a new article begins appearing in search, it may take time to collect views and feedback.

Give visitors time to use it before deciding whether the article is successful.

You can support discovery with:

• Internal links
• Automatic Article Links
• Category placement
• Recommended Articles
• Popular Searches
• Related Articles

Search Analytics is empty

If no terms appear, check that:

• Search Analytics is enabled
• The main knowledge base search is visible
• Compact search is enabled where expected
• Visitors have used the search box
• Your SoftwareRoad Knowledge Base licence is active
• Searches contain at least two characters
• The website cache has been cleared

Run one public test search while signed out.

Searches are not being recorded

If search works but the analytics remain empty:

1. Confirm Search Analytics is enabled.
2. Test in a private browser window.
3. Search for a unique test phrase.
4. Refresh the Analytics page.
5. Clear the WordPress cache.
6. Check for JavaScript errors.
7. Confirm the licence is active.

Avoid creating many test searches because they will affect the real report.

Results count looks wrong

The result total may change when:

• Articles are published
• Articles become unlisted
• Titles or content are updated
• Categories change
• Articles are deleted
• The website cache is refreshed

Run the search publicly and compare the current results.

Older Analytics entries may reflect the number of results available when the search was last recorded.

Similar searches appear separately

Terms with small wording differences may appear as separate rows.

For example:

• product key
• product keys
• find product key

Review them as a group when planning documentation.

You do not need to force every variation into one article title.

Old search terms remain

Search Analytics can retain historical terms even after you improve the documentation.

This allows you to see what visitors searched for in the past.

Focus on recent activity and whether the same no-result searches continue after your changes.

Clearing search data

Some plugin versions may include a control for clearing Search Analytics.

Before removing the data:

• Check whether you need historical trends
• Note important search terms
• Confirm that test data is the only information being removed
• Make sure a website backup exists
• Understand that deleted data may not be recoverable

Do not clear useful search data simply because some terms look unusual.

Search Analytics and privacy

Search terms are stored to help improve documentation.

Visitors should not be encouraged to enter names, complete product keys, passwords or other private information.

Your privacy policy should explain that knowledge base search terms and result counts may be recorded when Search Analytics is enabled.

Only trusted WordPress users should have access to the Analytics dashboard.

A simple monthly search review

Each month, you could:

1. Review the most searched terms.
2. Check repeated no-result searches.
3. Group similar wording together.
4. Improve article titles and excerpts.
5. Update articles using visitor terminology.
6. Create one or two genuinely missing guides.
7. Test the affected searches.
8. Update the homepage Popular Searches where useful.

Regular small improvements can make the knowledge base easier to use over time.

Search Analytics checklist

Before finishing your review, check that:

• Repeated searches have been identified
• No-result searches have been investigated
• Similar search terms have been grouped mentally
• Existing articles have been checked before creating duplicates
• Titles and excerpts use clear wording
• Visitor terminology has been added naturally
• Unlisted and draft articles have been checked
• Search results have been tested publicly
• Article views and feedback have been considered
• Personal or sensitive search entries are handled appropriately
• Useful terms have been considered for Popular Searches
• Changes are reviewed over time

Search Analytics can now help you improve how visitors find articles and reveal the documentation subjects that matter most to them.