How long does it take to recover from a Google update?
Problem
- A Google core update can take a third of your traffic overnight, with no penalty notice and nobody to appeal to.
- Google itself says recovery can take months, may wait for the next core update, and is not guaranteed.
Solution
- Read your visits before and after the update off your WebCounter counter, page by page.
- Put them into the free calculator below and it shows the visits and rands lost while you wait, and where you settle.
Key points
- A 30% drop on 10,000 visits a month costs 35,250 visits over two years - about R5,640 - in the default example.
- Traffic settles at 9,100 a month in month 10, 91% of before.
- 900 visits a month never come back if you win back 70% of the drop.
- Every month sooner the recovery starts saves about R336.
- Winning nothing back costs R11,520 - twice the default case.
In does posting more often grow your traffic we saw that a site grows by stacking posts that rank. A Google update can knock that stack down in a week. This post asks what a drop after a Google update costs you in visits and rands - and how long it takes to get them back.
Why a core update drop is slow to fix
A core update is not a penalty. It is Google re-judging pages across the web.
- Google says it makes broad core updates "several times a year", and that a drop usually means other pages now look more helpful than yours, not that you broke a rule.
- Improvements can take months to show, and Google says some sites may only see them after the next core update.
- Google also says there is "no guarantee" that changes you make will bring the visits back.
- So most of the cost is the waiting - every month at the low point is a month of lost visits.
Work out what the drop costs you
Type your visits a month before the drop, the share you lost, the months before recovery starts, the share of the drop you win back and how many months it takes, what 1,000 visits earn you and how long to count. The calculator shows the visits and earnings lost, the month you settle and what is still missing, and charts your visits against before.
Prefer a full page? Open the Google update recovery calculator on CalcSA. For a drop that comes back every year, the cost of a seasonal traffic dip calculator goes deeper.
Worked example: the calculator's defaults
A site at 10,000 visits a month loses 30% in a core update. It waits an estimated 4 months before recovery starts, then wins back an estimated 70% of the drop evenly over 6 months. 1,000 visits earn an estimated R160 (about 1.8 pages a visit at R90 per 1,000 page views). Counted over 24 months:
- The low point is 7,000 visits a month, for months 1 to 4.
- Traffic settles at 9,100 visits a month in month 10 - 91% of before.
- 35,250 visits are lost over the two years - about R5,640.
- 900 visits a month are still missing after the recovery, for as long as nothing changes.
- Each month sooner the recovery starts saves 2,100 visits - about R336.
Three more cases, same defaults except the one shown:
- Recovery starts after 1 month instead of 4: 28,950 visits lost, about R4,632, and settled by month 7.
- You win the whole drop back: 19,500 visits lost, about R3,120, and back to 10,000 a month in month 10.
- You win nothing back: 72,000 visits lost, about R11,520 - 3,000 visits a month gone for good.
Estimates. The 4 months before recovery starts, the 70% won back, the 6 months it takes and the R160 per 1,000 visits are assumptions, as in the calculator. Google publishes no recovery times. Your WebCounter counter shows your real visits before and after.
Two years after a 30% drop
Same 10,000 visits a month and 30% drop. Visits a month if you win back all of it, 70% of it, or none, starting after 4 months:
- All of it back
- 70% back
- None back
Show the numbers as a table
| Month | All of it back | 70% back | None back |
|---|---|---|---|
| 0 (before) | 10,000 | 10,000 | 10,000 |
| 1 | 7,000 | 7,000 | 7,000 |
| 4 | 7,000 | 7,000 | 7,000 |
| 5 | 7,500 | 7,350 | 7,000 |
| 7 | 8,500 | 8,050 | 7,000 |
| 10 | 10,000 | 9,100 | 7,000 |
| 24 | 10,000 | 9,100 | 7,000 |
- The flat stretch at 7,000 is the same in every case - you cannot shorten the wait for the next update.
- What you can change is the height you climb back to, and that gap lasts for years.
- Fix the pages that lost the most first, then check again a week after the next core update has finished rolling out.
Reading your counter for this
Your WebCounter counter shows visits by page and by source, so you can see exactly what an update took. No counter yet? Get your free WebCounter counter - it takes a minute and costs nothing.
- Size of the drop: compare search visits for the weeks after the update finished with the same weeks before.
- Which pages lost: sort pages by the change - a drop is usually a few pages, not the whole site.
- Rule out other causes - a seasonal dip, a broken page or a tracking change can look like an update.
- Count bots out - a crawler spike before the update makes the drop look bigger than it is.
Sources
- Google Search Central, Google Search's core updates - "several times a year", recovery that can take months or wait for the next core update, and "no guarantee" - checked October 2026 via a summary of the page, not quoted in full.
- Google, Search Status Dashboard - start and end dates of each core update's rollout.
- The 4 months before recovery, the 70% won back over 6 months and the R160 per 1,000 visits are our estimates - no published benchmark for core update recovery times was found.
- Calculator: CalcSA, Google update recovery.
Estimates only. Whether and when a site recovers depends on the update, your niche and what you change; your own counter is the test.
Count your page views - free
WebCounter gives your site a free counter and a 30-day report of page views and real visitors, with bots filtered out. No cookies on your visitors.
Get your free counter