> For the complete documentation index, see [llms.txt](https://docs.smaq.io/documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.smaq.io/documentation/features/search-term-analysis.md).

# Search Term Analysis

Search Term Analysis is SMAQ's deepest paid-search feature. It reads every search term that triggered your Google Ads keywords over a window, classifies them, and surfaces:

* **Wasted spend** — search terms with significant spend and no conversions, especially off-brand.
* **Underutilized winners** — search terms with strong conversion rates but capped by low impression share.
* **Drift** — search terms broadening your keyword's intent over time.
* **Brand vs off-brand split** — based on your on-brand / off-brand keyword list.

Available as:

* The [Search Term Analysis automation](/documentation/features/automations/reference.md#search-term-analysis) — runs on a schedule.
* The `/search-terms/` page in any project — interactive view.

> **\[Screenshot needed]** Search Term Analysis page showing a table of search terms with spend, conversions, conversion rate, and a "Wasted" / "Underutilized" / "Healthy" classification column.

### Setup

#### Configure on-brand / off-brand keywords

SMAQ needs to know which terms are "brand" (people searching for you by name) and which are "off-brand" (generic intent).

1. In project settings, find the **Search Term Analysis** section (or trigger the configuration prompt on first run).
2. **On-brand keywords**: list your brand variations — company name, common misspellings, signature products.
3. **Off-brand keywords**: leave blank for "everything not on-brand," or specify particular off-brand themes you want monitored.

#### Run the analysis

* **On the Search Terms page**: click **Analyze** to run on-demand. SMAQ pulls the last 30 days by default — adjust if needed.
* **As an automation**: see [Search Term Analysis automation](/documentation/features/automations/reference.md#search-term-analysis) — runs on a schedule and emails findings.

### Reading the output

Output groups search terms into buckets:

* **Wasted (off-brand, no conversions)** — kill candidates. Add as negative keywords.
* **Wasted (on-brand, no conversions)** — weird. Often signals tracking issues or duplicate ad showing.
* **Strong but underutilized** — search terms with conversion rate well above account average but limited by low impression share. Push more aggressively.
* **Healthy** — performing in line with account averages. No action.
* **Drift** — search terms now matching your keyword that didn't 30 days ago. Worth reviewing for intent shift.

> **\[Screenshot needed]** Search term table with the four classifications color-coded and one "Wasted off-brand" row expanded showing AI rationale.

### AI rationale

Click any search term to see the AI's classification rationale: spend, conversion rate, impression share, brand match score, and the reasoning chain.

### Actions

For each search term, you can:

* **Add as negative keyword** — generates the negative keyword text formatted for Google Ads. (SMAQ doesn't push to Google Ads on your behalf — copy and paste into your Google Ads account.)
* **Mark as reviewed** — clears it from future runs even if it would otherwise be flagged again.
* **Add to project knowledge base** — useful for terms you want to remember context on.

### Configuration fields (automation form)

When setting up the automation specifically:

* **Additional instructions** — free text to bias the AI (e.g. "Focus on terms below $100 spend").
* **On-brand keywords** — same as project setup but per-automation if you want to scope differently.
* **Off-brand keywords** — same.
* **Lookback days** — default 30.
* **Min spend** — minimum spend per term to consider (default $10 — filters out long-tail noise).
* **Min clicks** — minimum clicks (default 5).
* **Max search terms** — cap the output (default 100; raise for deep audits).

### Tips

* **Run weekly, not daily.** Search term data is noisy at the daily level.
* **Update on-brand keywords periodically.** New product launches, new sub-brands.
* **Trust the off-brand wasted bucket the most.** It's the highest-confidence cleanup.

### Tier availability

* **Starter** — Search Term Lite (capped weekly, narrow scope).
* **Growth / Scale / Agency** — full Search Term Analysis.

### Related

* [Google Ads connection](/documentation/connections/google-ads.md).
* [Keyword Exploration](/documentation/features/automations/reference.md#keyword-exploration) — finds new opportunities; companion to Search Term Analysis which cleans up existing.
* [Quality Score Monitor](/documentation/features/automations/reference.md#quality-score-monitor).