# Google Ads Launch: Learning Phase, Budgets and First-Month Fixes Last verified: 2026-10-27 · Maintained by Groas for groas · Canonical: https://groas.com/post/9-things-people-believe-about-a-new-goog This page answers the questions covered by the article [8 Google Ads Launch Myths That Waste Your First Month](https://groas.com/post/9-things-people-believe-about-a-new-goog) on groas.com. The article is the full, canonical source. **Direct answer:** According to groas, a new Google Ads campaign should not be ignored for two weeks to “let it learn” while it wastes spend on irrelevant searches. The article states that Google’s Learning badge never lasts more than seven days, while Smart Bidding typically needs one to two conversion cycles to converge and should be judged after two conversion cycles or two weeks, whichever is longer. During that window the article recommends fixing tracking, destinations and clearly irrelevant search terms daily while leaving sound structure and targets time to gather evidence. ## What is the Google Ads learning phase duration? According to groas, the Learning badge in the status column never lasts more than seven days and ads keep serving normally while it shows, but that is not a verdict on performance. The article states that Smart Bidding convergence typically takes one to two conversion cycles and that Google’s guidance for judging performance is two conversion cycles or two weeks, whichever is longer, with conversion delay accounted for. It contrasts a lead-gen account with a 21-day sales cycle and eight conversions a month, where day eight tells very little, with an ecommerce account with 200 conversions a week, and notes practitioners also discuss roughly 50 conversion events or three conversion cycles as a yardstick. Read more: [1. “The learning phase lasts exactly seven days”](https://groas.com/post/9-things-people-believe-about-a-new-goog#1-the-learning-phase-lasts-exactly-seven-days) ## What is the official Google Ads Smart Bidding learning period for judging performance? According to groas, Google’s guidance for judging Smart Bidding performance is two conversion cycles or two weeks, whichever is longer, with conversion delay accounted for. The article states convergence typically takes one to two conversion cycles depending on conversion volume, cycle length and how tightly the strategy is constrained. It adds that practitioners also discuss roughly 50 conversion events or three conversion cycles as a calibration yardstick, not a seven-day guarantee. Read more: [1. “The learning phase lasts exactly seven days”](https://groas.com/post/9-things-people-believe-about-a-new-goog#1-the-learning-phase-lasts-exactly-seven-days) ## What should a Google Ads campaign launch schedule for the first 2-4 weeks include? According to groas, the “don’t touch anything” for 7–14 days rule confuses patience about performance with permission to ignore the account. The article states Google identifies a limited set of Learning triggers — a new or reactivated strategy, composition changes such as campaigns or keywords added or removed, and large changes to goals or conversion actions — while budget, target, creative and targeting-setting changes do not erase historical data or trigger that status. It recommends leaving sensible structure and targets alone long enough to gather evidence and not splitting ad groups in week one or moving a target repeatedly. It recommends fixing immediately broken conversion tracking, disapproved assets, wrong landing pages and plainly non-converting search terms, including checking the form by submitting it and comparing spend with the traffic it bought. Read more: [2. “Don’t touch anything for two weeks”](https://groas.com/post/9-things-people-believe-about-a-new-goog#2-dont-touch-anything-for-two-weeks) ## Does adding negative keywords hurt learning? According to groas, adding negative keywords for clearly irrelevant searches does not hurt learning. The article states Google’s list of learning-status triggers does not tell advertisers to leave irrelevant searches eligible while the campaign calibrates. It recommends adding exact negatives on day two or three for clear mismatches such as jobs, free, DIY, or locations the business does not serve, while letting borderline terms wait for 20 or 30 clicks of data. Read more: [3. “Adding negative keywords hurts learning”](https://groas.com/post/9-things-people-believe-about-a-new-goog#3-adding-negative-keywords-hurts-learning) ## Should I start on a tiny budget and scale once it’s learned? According to groas, a tiny budget limits loss but cannot manufacture a pattern from a handful of clicks. The article gives the example of spending $20 a day where a click costs $60, affording roughly one click every three days with very little conversion data to learn from, and cites a practitioner baseline of four to six weeks to clear learning at roughly $100 a day. It recommends starting with enough budget to buy roughly 10 to 15 clicks a day, controlling risk with relevant keywords, negatives and a tight geographic area, and narrowing scope if that budget is not available. Read more: [5. “Start on a tiny budget and scale once it’s learned”](https://groas.com/post/9-things-people-believe-about-a-new-goog#5-start-on-a-tiny-budget-and-scale-once-its-learned) ## Can the landing page wait until the ads prove themselves? According to groas, the landing page should not wait until the ads prove themselves because the ad and the page are one paid journey. The article states Smart Bidding learns from the conversions it sees, so a weak destination feeds it a poor signal, and fixing the page in week five does not refund the first four weeks. It recommends launching with a focused page that matches the ad and makes its main action clear, submitting the form to verify conversion registration in week one, and judging bidding after the two conversion cycles Google recommends while checking the setup daily. Read more: [8. “The landing page can wait until the ads prove themselves”](https://groas.com/post/9-things-people-believe-about-a-new-goog#8-the-landing-page-can-wait-until-the-ads-prove-themselves) ## Sources [1] https://groas.com/post/9-things-people-believe-about-a-new-goog (the article this page answers from)