# Recovering a publisher's lost Google Discover traffic

> A digital publisher lost almost all of its Google Discover traffic after a Google core update, and Discover was where most of its readers came from. The impressions went to zero and stayed there for months. I rebuilt the quality signals Discover needs, held the site steady through the wait, and when Google next re-evaluated, Discover switched back on across their titles and the main site reached its best month on record.

**Digital publisher · Google Discover · SEO recovery · Solo engagement**

## Before → After

Before: Dropped from Google Discover after a core update, with its main traffic channel sitting at zero.

- **0**: Discover impressions, for months
- **97%**: of articles given Discover-ready images
- **Same day**: every title, the day Google re-evaluated
- **Record**: the main site's best Discover month

## The stakes

Most of this publisher's readers didn't arrive from a search box. They came from Google Discover, the feed of articles Google shows people on their phones. Then a Google core update rolled through and their titles fell out of Discover completely. Not down a little, gone. The impressions went to zero and stayed there. For a publisher that runs on Discover, that is the main source of readers vanishing overnight, with no query to rank for and no button to press to bring it back.

## The insight

**Discover doesn't come back because you fixed it. It comes back when Google re-evaluates you, and only if you pass.**

Discover isn't search. Search answers a query, so if you rank, you show up. Discover is a recommendation feed that decides on its own whether to put you in front of someone. Once a site falls below the quality bar it stops getting picked, and normal recrawling doesn't undo that. The thing that re-opens the door is one of Google's broad core updates, when it re-scores sites from scratch. So the job wasn't to fix Discover directly. It was to make the site clearly good enough to be recommended again, and to have all of it in place before the next core update looked.

## What I did

I rebuilt the signals Discover actually cares about. The articles had no proper images, so I gave them images at the size Discover needs, covering almost the whole catalogue. I cleaned up the things that quietly tell Google a site is low quality: broken images, broken structured data, a schema bug that was telling Google the articles belonged to a different site, missing author information, missing descriptions. I trimmed the thinnest pages that were dragging the whole site's quality down, with a safety check so nothing that was still earning traffic got touched. Every change was reversible and tested on a small batch before it went out. Then came the harder part: I stopped changing things. Every big change restarts Google's re-evaluation clock, so once the fixes were in I held the site steady and waited, watching the numbers with a tracker I built to pull the data every day.

## The result

When Google's next core update rolled out, Discover switched back on across their titles on the same day. The main site held, and went on to its best Discover month on record, above its previous seasonal peak. Some of the smaller titles came back and then faded. A channel that had been at zero for months was live again.

**What this proves:** Getting back into Google Discover isn't a growth trick. It's making the site good enough to be recommended again, then being patient enough to let Google notice. You can't force the timing, but you can be ready when it comes.

Lost Discover or search traffic after an update? Let's talk it through. Book a call: https://calendly.com/nerijus-masikonis/30min