Google rewrites two out of three descriptions. Is snippet work still worth it?

Google often does not use the meta description you wrote. At first glance that sounds like you could save yourself the work.

You can't.

Because the snippet is one of the few parts of a search result you can influence yourself, without changing your position. And when a page already ranks well, this is exactly where something may still be left on the table.

So the more important question is not:

"How do I write a perfect meta description?"

But:

"Is a weak click-through rate actually caused by the snippet?"

A low CTR is not yet a task. It only becomes one once position, search intent and the amount of data all hold up.

Not every weak CTR is a snippet problem

A page at position 13 with a CTR of 0.8 per cent probably does not need a better snippet. It needs more visibility.

It looks different for a page at position 3 that gets noticeably fewer clicks than you would expect at that position.

So before you change title and description, three things should be in place:

• Enough impressions: With very few impressions, the CTR is too random to draw anything from it.
• A good position: The question of click-through rate only becomes worthwhile from a reasonable ranking position upwards.
• A visible gap: 6.8 per cent instead of an expected 7.2 per cent is not a clear problem.

Other causes come into play too: Self-competition, a mismatched search intent or an AI Overview sitting above the result can all push the click-through rate down.

What actually belongs in a snippet?

The answer is surprisingly simple:

The user's question. Not the company's self-portrait.

A typical company title reads something like this:

Running Shoes Overpronation | Nordlicht Laufsport – Your Specialist Store Since 1998

The problem: The user wanted to know which running shoes suit overpronation. The title tells them about the company instead.

Better:

Running Shoes for Overpronation: 6 Models Compared

And the description should say concretely what awaits the visitor:

Six models compared by weight, drop and price.
Plus a simple self-test you can do at home.

This is not an SEO trick. It is simply a better answer to the search query.

But doesn't Google rewrite the snippet anyway?

Yes.

Studies by Ahrefs, Portent and others arrive at high rewrite rates for meta descriptions and title tags, depending on the dataset. Google itself also points out that snippets are frequently generated straight from the page content.

That does not mean your own description is pointless.

Google decides per search query which excerpt fits best. One page can therefore end up with different snippets for different queries.

And part of the rewriting is genuinely avoidable: A description that answers the query precisely and actually summarises the content of the page gives Google less reason to use a different text.

In short:

Your meta description is not an instruction to Google. But it is an offer.

And that offer should be a good one.

What about AI search?

This is where it gets speculative.

AI search systems work with sources during research and receive, among other things, titles, URLs and text excerpts. So it is plausible that this information plays a part in the pre-selection.

What is not established is that a better meta description leads to more AI citations. As long as no solid study supports it, nobody should claim otherwise.

For SEO the benefit is clear regardless: A precise title and a good description help the user place the result.

How to measure whether the change worked

This is where most mistakes happen.

Change the meta description, wait three days, celebrate a higher CTR?

That is not enough.

Because the position may have moved at the same time. On top of that, Google needs time to pick up the change, and Search Console delivers its data with a delay.

So the measurement should be kept as clean as possible:

1. Before: Record impressions, clicks, CTR and position.
2. Change: Adjust the title and/or description.
3. Wait: At least 14 days, ideally longer — and wait for enough impressions.
4. After: Compare the same metrics again.

What matters is not only whether the CTR went up, but whether the difference is large enough not to be chance.

A rise from 3.3 to 5.9 per cent across several thousand impressions is a different thing from a rise from 3.3 to 4.1 per cent across 200.

And sometimes the result reads:

No demonstrable difference.

That is not a bad result. It simply means this change apparently achieved nothing.

This is precisely the answer that is accepted far too rarely in SEO testing.

What Rankmio makes of this

Rankmio therefore does not start from "low CTR equals change the snippet".

First it checks whether the snippet is even a plausible cause. Among other things, it looks at the query, the page, the position and the Search Console data.

Only when a sensible angle emerges does a proposal get written.

Afterwards the change is measured again.

The result can be:

Better · Worse · No demonstrable difference

The third answer comes up most often, and it is a real answer.

A good SEO analysis tells you not only what you can change. It also tells you what you are better off leaving alone. That is the difference between a task list and a busywork list.

What it comes down to

1. Check the cause first. Not every low CTR is a snippet problem.
2. Look at query and page together. Averages across a whole website only help so much with a concrete decision.
3. Answer the user's question. Don't put the company at the centre.
4. Measure your changes. Don't cheer after three days — check whether the effect actually holds up.

Frequently asked questions

Quick quiz: is it the snippet?

Sources

• Google Search Central: How to Write Meta Descriptions
• Ahrefs: How Often Does Google Rewrite Meta Descriptions?
• Zyppy: Google Rewrites 61% of Page Title Tags
• Portent: How Often Google Rewrites Our Title Tags
• SparkToro: In 2026, Less than One Third of Google Searches Still Send a Click