Google News and Discover for volatile editorial traffic

Discover traffic swings 40 to 70 percent month over month at every publisher we have worked with, and almost none of that movement is explained by anything on the page. So we do not sell Discover optimisation as though it were a dial you turn. We build the eligibility conditions properly once, then spend the engagement on the decision that actually varies week to week: which stories are worth the attempt. Eligibility, done once and done correctly: The max-image-preview:large robots directive, without which Discover will not show a large image and the story will not surface in any volume worth counting. Lead images at least 1,200 pixels wide, and not the same crop as the in-article header, because the card uses a different aspect ratio and a cropped face or a clipped headline destroys the click. NewsArticle markup with datePublished and dateModified that tell the truth, since backdating updates to farm freshness is detected and it costs more than it earns. Publisher Center configured. News sitemaps built to the 48-hour window and the 1,000 URL limit, and a check that your CMS is not generating them from the wrong date field, which is the single most common defect we find and it is usually silent. Authors as entities: In games and sports coverage this is where most publishers are weakest. An author page with a headshot and two sentences is not an entity. We build profiles with a defined beat, outbound links to the same person on platforms already recognised, and consistent bylines across the archive, then verify whether the knowledge graph has actually picked them up rather than assuming that publishing the page was sufficient. The editorial call: Top Stories and Discover are different systems and they reward different things. A patch note has a six-hour window in Top Stories and almost no life in Discover. A well-photographed piece on a roster change, a doping ruling or an equipment recall has the reverse profile. We sit in your editorial meeting weekly and say which is which, and more often we say a story is not worth covering at all because three larger outlets already have it and a rewrite will not outrank them. For live coverage — matches, tournaments, launch events — LiveBlogPosting markup with a genuine update cadence, or do not run a live blog. Not included: We do not write or edit your stories and we do not supply headlines. We will reject a headline for being unclear or for promising more than the piece delivers, and we will tell you what a truthful headline costs in clicks, but the desk decides. We do not run social distribution or newsletters. We do not arrange paid syndication. Who this is not for: Sites publishing fewer than five original pieces a week. Discover rewards consistent supply and there is nothing here for a monthly title. Aggregators and rewrite operations, because we will not help a site get better at republishing other people's reporting. Publishers who want a guarantee: nobody can promise Discover inclusion, the Search Console report does not even appear below a traffic threshold, and an agency offering that guarantee is selling a coin flip with an invoice attached. What you get: An eligibility audit with every defect ranked by the traffic it is costing, a CMS specification your developers implement, an author entity plan, a weekly editorial session with a written call on each candidate story, and monthly reporting that keeps Discover, Top Stories and standard search apart instead of merging them into one line that conceals which one actually moved.