MDX failed to compile
The article body contains characters MDX can't parse. Common culprits: unescaped { / }, or < followed by a letter (looks like a JSX tag). Edit the body and re-save to fix.
[next-mdx-remote] error compiling MDX: Unexpected character `3` (U+0033) before name, expected a character that can start a name, such as a letter, `$`, or `_` More information: https://mdxjs.com/docs/troubleshooting-mdx
Most beginners pick Broad Match because it sounds comprehensive. More coverage, more reach, more chances to sell. Then the budget's gone by noon and there's nothing to show for it. Keyword Match Types aren't complicated. But picking the wrong one early is one of the fastest ways to drain a campaign before it has a chance to prove itself. ## What match types actually do There are three options, and each one controls how loosely Google interprets your keyword before showing your ad: **Broad Match** — Google shows your ad for searches related to your keyword, including synonyms, similar topics, and variations. "Related" is doing a lot of heavy lifting there. Google's definition of related is almost always wider than yours. **Phrase Match** — shows for searches that include the meaning of your keyword, with some flexibility on surrounding words. More control than Broad, more reach than Exact. The middle ground most campaigns should eventually live in. **Exact Match** — shows for searches that match your keyword's intent closely. Close variants like typos and plurals are included, but Google doesn't wander. You know exactly what someone typed before your ad appeared. <Callout variant="tip" title="Start with Exact — always"> Launch with Exact Match. If impressions are too low after two weeks, add Phrase Match. Only test Broad once you have 30–50 conversions per month and Smart Bidding is active. Not before. </Callout> ## Why Broad Match burns budgets Say you're selling project management software and you add `team productivity` as a Broad Match keyword. Google might show your ad for: - "productivity podcast recommendations" - "home gym workout for productivity" - "productivity tips for students" None of those people will ever buy your software. But you'll pay for every click anyway. Broad Match earns its place only when Smart Bidding has real signal to work with — at least 30–50 conversions per month. Without that data, you're funding Google's learning, not your sales pipeline. ## Phrase Match: the underused middle ground Phrase Match uses quotes around your keyword: `"project management software"`. It shows ads for searches that carry the same intent, even if extra words appear before or after. With `"project management software"` as Phrase Match, you'd show for: - "best project management software for small teams" ✓ - "project management software free trial" ✓ - "project management software comparison 2026" ✓ But not "how to manage a project without software" — different intent entirely. This is where most campaigns should move once they have two to three weeks of clean Exact Match data. You know what's converting. Phrase lets you capture more of it without guessing. ## Start with Exact Match — here's why Exact Match uses brackets: `[project management software]`. It's the tightest option and the one that gives you the cleanest data. Every dollar is traceable. Every impression tells you something real. When a campaign starts with Exact Match, you build a picture of what actually converts before you ever spend money on broader traffic. The downside: volume can be lower if you pick niche terms. That's fine. Low volume with high relevance beats high volume with a 0.2% Conversion Rate any day. The workflow that works: run Exact Match for two weeks → open the Search Terms Report → find relevant queries you hadn't thought of → add them as new Exact or Phrase keywords → repeat. ## The trap nobody mentions <Callout variant="warn" title="Broad Match + Smart Bidding isn't a shortcut"> Google will pitch Broad Match paired with Smart Bidding as the smart move because the algorithm "learns better." True — but only when you have historical conversion data. A new account with 8 conversions per month running Broad + Target CPA is handing Google a blank check to experiment. Smart Bidding needs data to be smart. It isn't smart by default. </Callout> The second trap: running any Broad Match without a Negative Keywords list. Your Search Terms Report will always surface irrelevant queries. Without negatives, you're paying for traffic you'd never want in the first place. Make this a weekly habit: open the Search Terms Report → identify queries that spent money but would never convert → add them as Negative Keywords. This one habit alone can cut wasted spend by 20–30% without touching a single bid or budget. ## Quick reference | Situation | Use | |---|---| | New account, <30 conversions/month | Exact Match only | | Have data, want more reach | Add Phrase Match | | 30+ conversions/month + Smart Bidding active | Test Broad Match (separate campaign) | | Testing a new keyword | Exact first, scale with Phrase | ## What to do next Open your Search Terms Report right now. Look at what queries have actually spent your budget in the last 14 days. If more than 40% of the spend went to searches that wouldn't convert, your match type is too wide. Tighten from the top — find the keywords spending the most on irrelevant queries and either switch them to Exact, add negatives, or pause them. Fix the leak before you scale the spend. <!-- INTERNAL LINKS TO ADD --> <!-- [insert link to "Negative Keywords: how to stop paying for bad traffic in Google Ads"] --> <!-- [insert link to "Smart Bidding on Google Ads: Target CPA vs Target ROAS — which to use"] -->