Google Started Rewarding Helpful Content

What was announced

Google described the change last week and the rollout begins today, expected to take a couple of weeks.

The stated target is content made primarily for search engines rather than for people.

It is automated and continuous rather than a one-off adjustment, and it applies to English-language searches initially.

Google published a set of questions to help site owners judge their own content, which is a more useful document than the announcement itself.

Why this one is different

Two features distinguish it from most updates, and both matter for how you should respond.

It is described as a sitewide signal, meaning pages judged unhelpful can affect how a whole site performs rather than only themselves.

That is unusual. Most ranking factors are assessed page by page, and a weak page simply does not rank.

The second is that it is continuous, so a site that improves should recover over time rather than waiting for a specific event.

Together those change the calculation on pages you have been keeping because they do no harm, since the position now is that they might.

What it is aimed at

• Content summarising others without adding anything.
• Writing about topics outside what you do.
• Automated production across many subjects.
• Answering questions that have no answer.
• Pages built to a word count rather than to a question.
• Covering trending subjects for the traffic.

The second is worth reading carefully by anybody who has been advised to publish broadly. Writing about subjects adjacent to your trade, because they attract searches, is precisely the pattern described.

What it means for a small business

Less than the coverage will suggest, for most.

A business publishing about its own work, from its own experience, is doing the thing this update is meant to reward, and needs to change nothing.

The businesses affected are those that were persuaded to publish volume: town pages, topic pages outside their trade, and material produced to a schedule with nothing behind it.

If a supplier has been producing your content across many subjects, this is worth a conversation this week.

And if you have pages you would be embarrassed to have somebody land on, that instinct has now acquired a mechanical consequence.

The people-first framing

Worth noting because it is how the guidance is written and it explains the shape of the questions.

The distinction being drawn is about who the content was made for rather than about any measurable property of it.

Which means there is no checklist that makes a page pass: no word count, no structure, and no density of anything.

A short page written to answer a customer's question is on the right side of the line, and a long, well-structured, thoroughly researched page written because the term gets searches is not.

That is uncomfortable for anybody who has been optimising to a specification, and it is a considerably better description of what has always worked.

What not to do

Since the reaction to an update is frequently worse than the update.

Do not rewrite everything at once, which makes any effect impossible to attribute and risks pages that were working.

Do not delete pages in a panic before checking whether they have traffic, external links, or conversions.

Do not buy a service promising to fix your site for this, which will be offered within days by people who know no more than you do.

And do not judge anything by next week, since the rollout takes a fortnight and any assessment before it completes is measuring noise.

A worked example

A trades business had eleven service pages written from experience and about sixty pages produced by a supplier: town variants and general articles on subjects adjacent to the trade.

The eleven produced almost all of the enquiries. The sixty produced traffic and nothing else.

The owner had been reluctant to remove them on the basis that they might be doing something.

This update changes that reasoning, since the sixty may now be affecting the eleven.

The plan agreed was to check each of the sixty for links and conversions, merge the handful with value into the service pages, remove the rest, and record the figures first.

That was work he had been putting off for a year and now had a reason to do.

The self-assessment questions

Worth going through properly, since they are more concrete than the announcement.

They ask whether you have an existing audience who would find the content useful, whether it demonstrates firsthand expertise, whether the site has a primary purpose, and whether a reader leaves feeling they learned enough.

They also ask whether you are writing on subjects simply because they seem to attract searches, and whether the content promises an answer it does not deliver.

Read them against your own pages rather than in the abstract, page by page, and note the ones that fail.

Most small business sites pass on the service pages and fail on whatever was added for volume, which tells you exactly where to work.

Removing weak pages may help the rest

The practical consequence of a sitewide signal, and it is stated in the guidance rather than inferred.

Where content is judged unhelpful, removing it may improve how the remaining pages perform.

That is a different proposition from the usual position, where a weak page simply sits there ranking for nothing and costing you only the maintenance.

It also means the pruning that was previously a tidiness exercise now has a stated mechanism behind it.

Check each candidate for links and conversions first, as always, and merge anything with value into a page that is staying rather than deleting it outright.

The counter-case

Updates are consistently over-interpreted.

Every one of these is followed by confident explanations of what it means, most of which turn out to be wrong, and by services sold on the strength of them.

The effect on any individual small site is unpredictable and frequently nothing, and a site that never published filler has nothing to respond to.

The underlying advice is also not new. Writing from experience about what you actually do has been the right approach throughout, and this makes it slightly more mechanically rewarded.

Go through the questions, deal with the pages you already knew were weak, and otherwise carry on.

This fortnight

1. Record your figures before anything changes.
2. Read the self-assessment questions.
3. Apply them page by page.
4. List the pages that fail.
5. Check those for links and conversions.
6. Ask your supplier what they have been producing.
7. Wait for the rollout before judging anything.

Step one is the one nobody does and the only thing that will let you tell, in a month, whether anything actually happened to your site.

The previous announcement worth planning around is covered in Google set a date to switch off Universal Analytics.

Frequently asked questions

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.