PPC Campaign Checklist

Pick one primary objective: prospecting (top-of-funnel awareness/traffic), conversion (purchase), or retargeting. Set a target ROAS or TACOS based on the SKU's contribution margin — a 30% margin product can't sustain a 2.0 ROAS goal. Document MER expectations if blended performance matters more than channel-level ROAS.

Match channel to objective: Google Search/Shopping for high-intent demand capture, Meta/TikTok for prospecting and creative-driven discovery, Amazon Sponsored Products for marketplace conversion, Pinterest for category discovery. Brand-defense Sponsored Brands campaigns belong on Amazon if you have Brand Registry.

For Google: pull search-term volume and competition from Keyword Planner; segment by exact, phrase, and broad match intent. For Amazon: use Helium 10 Cerebro or Jungle Scout to harvest competitor ASINs and search-volume data. For Meta/TikTok: define interest stacks, lookalike sources (1%, 2-5%), and exclusion audiences (recent purchasers, current subscribers).

Pre-empt wasted spend on irrelevant queries — "free", "diy", "cheap", competitor brand names you don't want to bid on, and out-of-category modifiers. Apply at the campaign or shared library level. On Amazon, add as negative exact / negative phrase at the ad group level.

Google Responsive Search Ads need 15 headlines and 4 descriptions. Lead with the offer or differentiator, not the brand name. Mirror the landing page H1 to protect Quality Score. Avoid superlative claims ("best", "#1") that trigger FTC substantiation issues or platform policy review.

Pull approved imagery from the PIM (Akeneo, Plytix) — never grab from the website without checking license terms. Meta and TikTok favor UGC-style vertical 9:16 video; Google Shopping pulls images from the merchant feed. Verify Amazon main image complies with white-background and 85%-fill requirements or the listing gets suppressed.

Vary one element per variation — hook, image, CTA — so you can isolate what's driving performance. Avoid kitchen-sink testing where every variation differs on every dimension; you'll have no read on what worked.

Check every substantive claim against the substantiation file — "clinically proven", "FDA approved", "#1 selling" all need backup. Influencer / UGC creative needs #ad or "paid partnership" disclosure per FTC Endorsement Guides. Health, finance, and supplement categories trigger extra platform review on Meta and Google.

Use Google Tag Assistant, Meta Pixel Helper, or TikTok Pixel Helper to confirm Purchase, AddToCart, and InitiateCheckout fire correctly. Confirm GA4 enhanced ecommerce events match Shopify analytics within 5%. For Amazon, verify Attribution tags on off-Amazon traffic. Check server-side / Conversions API setup for iOS 14+ signal recovery.

Default to "people in or regularly in your targeted locations" — not "or interested in", which leaks budget to international searchers. Exclude states where you don't ship (alcohol, CBD, supplements with state restrictions). For multi-country campaigns, separate by country to control bid and currency.

Pull 30-90 days of conversion-by-hour data before applying schedule modifiers — don't guess. B2B-skewing brands often see weekday-business-hours peaks; gift / impulse categories peak evenings and weekends. Avoid dayparting brand campaigns; they should always be on.

Start manual CPC for new campaigns to gather conversion data, then transition to Target ROAS or Maximize Conversions once you have 30+ conversions in 30 days. Smart Bidding starves on low conversion volume. On Amazon, dynamic-bids-down-only is safer for new keywords than fixed bids.

Within the first 2 hours, confirm impressions are accruing and the pixel is firing on the live ad. Check that landing pages return 200 and load on mobile under 3s — campaigns can deliver against broken pages and burn budget. Watch first-day spend pacing for runaway CPCs.

Day 1-3 is too early to judge ROAS — focus on CTR (above 1% on Search, above 0.8% on Display/Social), CPC vs. forecast, and impression share lost to budget vs. rank. Triple Whale or Northbeam dashboards roll this up across channels for blended MER.

Wait until 100 clicks per ad group before judging conversion rate; smaller samples are noise. Compare to baseline pre-campaign conversion rate; if site CR is dropping with new traffic, the issue is targeting or message-match, not the landing page.

Pull the search-term report weekly (Google) or customer search-term report (Amazon). Add irrelevant queries as negatives; promote high-converting discovered terms to exact-match in their own ad group. This is where most of the optimization lift hides in the first month.

Walk the funnel: impression share lost (auction issue), CTR (creative issue), CR (landing page or audience issue), AOV (offer issue). Lower target ROAS by 20% or switch to Maximize Conversions for two weeks to gather data, then re-evaluate. Don't churn the bid strategy weekly — Smart Bidding needs ~14 days to relearn.

Use platform-native split tests (Meta A/B Test, Google Drafts and Experiments) — manually splitting budget between ad sets contaminates with audience overlap. Test for 7 days minimum at the same daily budget. Document winning variations for the creative library.

Promote high-converting search terms into single-keyword ad groups (SKAGs) for Google Search, or harvest into exact-match on Amazon. On Meta, scale lookalikes from purchaser data once you have 500+ conversions in the source audience. Trim audiences with frequency above 4 to fight creative fatigue.

Pull Microsoft Clarity or Hotjar session recordings for paid-traffic visitors. Check Core Web Vitals (LCP under 2.5s, INP under 200ms, CLS under 0.1) — slow mobile pages tank Quality Score and conversion rate together. Test offer placement, social proof above the fold, and one-click checkout (Shop Pay, Apple Pay).