The PPC Bestiary

Professionals pair broad match with strong conversion data, an aggressive negative keyword list, and a Smart Bidding strategy that has enough conversion volume to actually learn from, then watch the search terms report weekly to keep it fed on the right things.

Used to protect known, high-converting queries and to isolate bid control for terms with proven intent, usually built from search terms report data rather than guesswork.

Built as the middle layer of an account structure, catching variations the exact match tier misses while still staying anchored to a specific concept, then trimmed with negatives as its search terms report reveals drift.

Maintained as a living list, reviewed on a set cadence against the search terms report, with shared negative lists applied across campaigns so the same junk query never has to be blocked twice.

Improved indirectly by tightening ad group themes, writing ad copy that mirrors search intent, and making sure the landing page actually delivers what the ad promised, then read as a diagnostic rather than chased as a score.

Only switched on once conversion tracking is verified accurate and the account has enough conversion volume to support the chosen strategy, then given room to exit its learning phase before judging results.

Set using a target derived from actual historical account performance rather than an arbitrary goal, and given a wide enough range to keep bidding active instead of stalling out.

Reserved for lead generation or single-price offers where conversions really are comparable in worth, and swapped for Target ROAS once the business can track conversion value.

Given clear audience signals, tightly curated asset groups, account-level negative keyword lists and brand exclusions where relevant, and checked against search term insights and new customer data rather than judged on ROAS alone.

Reviewed on a regular schedule to harvest new keyword ideas, spot mismatches between intent and match type, and build the negative keyword list from real evidence instead of assumption.

Monitored alongside the lost impression share breakdown to diagnose whether the fix is more budget or better Ad Rank, rather than treated as a vanity number on its own.

Improved by working the inputs that are actually controllable, ad relevance, expected click-through rate, landing page experience, and enabled extensions, rather than trying to win purely by raising the bid.

Verified before anything else in an account audit, checked against actual business records for accuracy and duplication, because Smart Bidding built on broken tracking optimizes toward a fiction.

Implemented through the Google tag, Tag Manager, or the Conversions API alongside a consent management setup, treated as a data quality upgrade rather than an optional extra.

Left alone for the period it needs to gather sufficient conversion data, with major structural or budget changes batched and spaced out rather than made reactively day to day.

Isolated into its own campaign or measured with brand exclusion segments and incrementality checks, so blended ROAS numbers do not get mistaken for the real efficiency of non-branded prospecting.

Run with placement exclusion lists, managed placements or curated audiences, and separate reporting from Search so its very different click quality never gets averaged into the same conclusions.

Segmented by intent and funnel stage with frequency capping applied, so the shade reminds rather than haunts, and paired with exclusion lists for people who already converted.

Kept synced automatically from the site's actual inventory and pricing system, audited for disapprovals and missing attributes on a regular schedule rather than set up once and forgotten.

Used as a checklist for asset variety and completeness, then set aside in favor of actual conversion and click-through data when deciding which creative is truly working.

Monitored by watching for clicks with unusually low conversion rates or suspicious traffic patterns by location and device, with IP exclusions and Google's invalid click reporting tool used when a genuine pattern is confirmed.

Applied for genuinely short, unusual events with an estimated conversion rate change based on real past performance, then removed once the event ends so it does not distort ordinary bidding afterward.

Checked first whenever performance shifts unexpectedly, and reviewed after handoffs between agencies or team members to understand exactly what was changed and when.

Set using monthly and campaign-level budget planning rather than reacting to single-day spend swings, with shared budgets used deliberately across campaigns that should draw from the same pool.