How to read a core update without panicking

Google began rolling out a core update on 2 June and said something it has never said before: this one comes in two parts, and the second part arrives in July. Some of the improvements Google wanted to ship were not ready in time, so the June release went out without them.

That matters more than it sounds. Google has been explicit that a site which drops in June may recover in July when the second half lands, and that a site which gains in June may give some of it back. Which means the single worst thing you can do right now is look at ten days of data and start rewriting your website.

Meanwhile the page experience update starts rolling out within days and continues until the end of August. Two overlapping changes, neither of them finished, both of them landing on the same charts. Here is how we are reading it.

What Google actually said

Three things, and they are worth keeping straight because the rest of the internet has already blurred them together.

• The core update started on 2 June and takes a couple of weeks to finish.
• A second core update follows in July, containing the parts that did not make this one. Movement from June may reverse then.
• The page experience update is separate, starts mid-June, and rolls out gradually until the end of August. It is a tiebreaker, not a ranking overhaul.

Core updates are not penalties. Nothing in your site was flagged. Google reassessed what it considers the best answer for a set of queries and your pages moved relative to everyone else’s. That distinction matters because penalties have fixes and core updates mostly do not, at least not on the timescale people want.

What we are seeing in the accounts we run

Cautiously, because the rollout is not done. Organic positions on informational pages are moving around more than service pages. Local pack results have been steadier than organic listings through the first week, which is usually how core updates go, since proximity and the listing itself carry so much of the local result. A couple of sites are up on queries they had given up on. A couple are down on queries they never really deserved.

Calls and booked appointments, which is what we actually report on, have not moved much anywhere. That gap between “rankings look different” and “nothing changed at the front desk” is the normal state of affairs during a core update, and it is the reason we ask owners not to make decisions during one.

How to read the movement without guessing

If you want to know whether something real happened to you, this is the order we work in.

1. Wait for the rollout to finish. Two weeks minimum. Ranking data inside a rollout is noise wearing a suit.
2. Compare 28 days to 28 days, not this week to last week. Then compare the same window to last year, because last June was not a normal month for any local business and half the “drops” people panic about are just a reopening comparison.
3. Split local from organic. Search Console tells you about organic results. Your listing insights tell you about the map. If organic is down and discovery views on the listing are flat, your problem is a page, not your visibility in the area.
4. Go page by page, not sitewide. In Search Console, filter to the pages that lost clicks and look at which queries they lost. Losing a blog post about a general question is not the same as losing the page you want people to book from.
5. Check the front desk. Call volume, form submissions, booked appointments. If those are flat and a rank tracker is red, the rank tracker is a hobby, not a business problem.

Page experience is not this

When the page experience rollout starts and someone sees a dip in the same week, it will get blamed for everything. Two reasons to resist.

The rollout is gradual over roughly ten weeks, so there is no day when it lands on you. And Google has repeatedly described the effect as small, applied between pages that are otherwise comparable. If you lost a position to a competitor whose page is slower than yours, speed is not the story.

The work on Core Web Vitals is still worth doing, but do it because a page that loads fast and does not jump around books more appointments, not because you expect a ranking jump in July. That is the same argument we make about all site speed work.

What not to do this month

• Do not mass rewrite content. Especially not because a tool told you the “content score” is low.
• Do not delete or noindex pages because they lost traffic during an unfinished rollout.
• Do not buy links. Every core update produces a wave of people offering to fix your rankings for a monthly fee.
• Do not change your business name on the listing to include keywords. It is against the guidelines, competitors report it, and losing the listing costs more than any core update will.
• Do not fire anyone, including your agency, in week two of a two part update.

What to do this month

• Export Search Console data for the last 28 days now, so you have a clean before while the rollout is still running.
• Wait until the rollout is confirmed complete, then compare 28 days to 28 days and to the same period last year.
• Separate organic movement from listing performance before you diagnose anything.
• Pick the two or three pages that actually lost booking related queries and look at those pages honestly against the results now ranking above them.
• Keep the Core Web Vitals work moving on its own schedule, aiming at the templates rather than individual pages.
• Do nothing structural until July, when the second half of the update has landed.

Summer is a good season for most of the businesses we work with, and the schedule is a better signal than a ranking chart is. Keep publishing the pages your customers actually need, keep the listing accurate as hours change over the summer, and treat the next six weeks as a period for observing rather than reacting. That patience is most of what local search work is.