How to Refresh and Prune Underperforming Keywords: A Step-by-Step Guide

TL;DR: Refreshing and pruning underperforming keywords involves auditing your current keyword list, identifying low performers based on clear metrics, deciding whether to optimize or remove them, and implementing changes systematically. This guide walks you through the exact process to clean up your Google Ads campaigns and stop wasting budget on keywords that aren't pulling their weight. Whether you're managing your own campaigns or handling multiple client accounts, this process should become a regular part of your PPC maintenance routine—ideally every 30-60 days depending on spend volume.

Most Google Ads accounts I audit have the same problem: 20-30% of keywords are basically dead weight. They're consuming budget, triggering impressions, maybe even getting clicks—but they're not moving the needle on actual business goals. The frustrating part? These underperformers often hide in plain sight, buried in campaigns that look "okay" at the aggregate level.

Here's what usually happens: you launch campaigns with promising keyword lists, they run for a few months, and you focus on the obvious winners and losers. Meanwhile, a whole middle tier of keywords quietly drains your budget without contributing meaningful conversions. Sound familiar?

The solution isn't complicated, but it does require a systematic approach. You need to regularly refresh keywords that have potential and ruthlessly prune the ones that don't. This isn't a one-time spring cleaning—it's ongoing maintenance that separates profitable campaigns from money pits.

Let's walk through exactly how to do this, step by step.

Step 1: Pull Your Keyword Performance Data

Before you can make smart decisions about which keywords to keep, refresh, or kill, you need the right data in front of you. This means pulling a comprehensive performance report that covers enough time to be statistically meaningful.

Start by exporting your keyword data from Google Ads. For most accounts, 30-60 days is the sweet spot. If you're running low-volume campaigns (under 1,000 clicks per month), extend that window to 90 days or even longer. The mistake most advertisers make is pruning keywords based on a week or two of data—that's how you accidentally kill keywords that might have converted if you'd given them more time.

Here are the metrics you absolutely need to include in your export:

Impressions and Clicks: Basic engagement metrics that show whether your keywords are even getting seen and clicked.

CTR (Click-Through Rate): Tells you if your ad copy is resonating with the search intent behind each keyword.

Conversions and Conversion Rate: The ultimate performance indicators—are these keywords actually driving business results?

Cost and CPA (Cost Per Acquisition): What you're spending versus what you're getting back.

Quality Score: Google's assessment of your keyword relevance, which directly impacts your ad costs and positions. Understanding what causes low quality score and how to fix it is essential for this analysis.

Why does the timeframe matter so much? Because you're looking for patterns, not anomalies. A keyword that got 50 clicks and zero conversions in one week might convert beautifully over a month. Conversely, a keyword that converted once out of 10 clicks might just be lucky—you need more data to know if that's repeatable.

Here's a quick tip that saves hours: pull your Search Terms Report at the same time. This shows you the actual search queries triggering your keywords. In most accounts I audit, the problem isn't the keyword itself—it's that the keyword is matching to irrelevant search terms. You can't make smart pruning decisions without understanding the difference between search terms and keywords.

Once you have your data exported, organize it in a spreadsheet where you can sort and filter easily. This becomes your working document for the entire pruning process.

Step 2: Define Your Underperformance Criteria

Here's where most people get stuck: what actually counts as "underperforming"? The answer depends entirely on your campaign goals and your specific account performance.

Set clear thresholds before you start making decisions. For example, you might flag keywords that have received 100+ clicks with zero conversions, or keywords with a CPA that's 3x higher than your target. The specific numbers matter less than having consistent criteria you can apply across your account.

Different campaign types need different benchmarks. If you're running lead generation campaigns, you might be comfortable with a 2-3% conversion rate. For ecommerce, that number could be 5-10% depending on your product and price point. Brand awareness campaigns might prioritize CTR and impression share over direct conversions.

The "spend without return" rule is my go-to starting point: any keyword that's consumed more than your target CPA in spend without generating a single conversion is an immediate prune candidate. If your target CPA is $50 and a keyword has spent $150 with zero conversions, it's not "getting warmed up"—it's wasting money. Learning to prioritize keywords by ROI potential helps you make these decisions faster.

But here's the nuance most guides skip: you need to account for external factors before labeling something underperforming. Did you just raise bids significantly? That keyword might need time to stabilize. Is this a seasonal product that naturally performs better in Q4? Don't prune it in January based on December data.

In practice, I use a tiered system:

Red Flag Keywords: High spend, zero conversions, poor quality score (below 5/10). These get pruned immediately.

Yellow Flag Keywords: Moderate spend, below-target conversion rate, but showing some signs of life. These are refresh candidates.

Green Flag Keywords: Meeting or exceeding performance targets. Leave these alone or scale them up.

Gray Flag Keywords: Not enough data yet. These go into a "monitor" bucket for the next audit cycle.

The key is documenting your criteria upfront. Write down exactly what qualifies a keyword for each category in your account. This prevents you from making emotional decisions or changing the rules mid-audit because you're attached to certain keywords.

Step 3: Categorize Keywords into Refresh, Prune, or Monitor

Now that you have your data and your criteria, it's time to sort your keywords into three buckets: refresh, prune, or monitor. This is where the real decision-making happens.

Refresh candidates are keywords with potential that aren't currently delivering. These typically have decent traffic volume and engagement (good impressions, reasonable CTR) but are falling short on conversions. The problem isn't necessarily the keyword itself—it's how you're executing on it.

What usually happens here is you've got a solid keyword matched to mediocre ad copy or a landing page that doesn't quite align with search intent. Maybe you're using broad match and catching too many irrelevant variations. These keywords deserve a second chance with better execution.

Prune candidates are the budget drainers. High spend, no conversions, poor quality scores—these keywords are actively hurting your account performance. They're either fundamentally misaligned with your offering, too competitive for your budget, or triggering irrelevant traffic you can't control.

In most accounts I audit, 15-25% of keywords fall into this category. That's a significant chunk of wasted spend you can reallocate to your winners.

Monitor candidates are keywords that simply don't have enough data yet to make a confident decision. Maybe they've only received 20 clicks, or they launched two weeks ago. Don't make the mistake of pruning these prematurely—flag them for review in your next audit cycle.

Here's a simple spreadsheet system I use: add a column called "Action" and label each keyword as "Refresh," "Prune," or "Monitor." Then add a "Reason" column where you note why you made that decision. This documentation becomes invaluable when you're reviewing your work 60 days later or explaining decisions to a client. You can also import keywords via CSV to streamline this workflow.

The categorization process should be methodical, not rushed. Sort your keyword list by spend (highest to lowest) and work through your biggest budget consumers first. This is where you'll find the most impactful opportunities—a single high-spend underperformer can free up more budget than pruning 20 low-spend keywords.

As you categorize, pay attention to patterns. Are all your broad match keywords underperforming? That's a match type strategy problem, not a keyword problem. Are certain ad groups consistently producing prune candidates? That might indicate a targeting or messaging issue at the ad group level.

Step 4: Refresh Keywords Worth Saving

Refreshing underperforming keywords is where the real optimization skill comes in. These are keywords that have shown some promise but aren't quite delivering—and with the right adjustments, they can become profitable contributors to your campaigns.

Start by testing different match types. If you're running a keyword on broad match and it's triggering lots of clicks but few conversions, tighten it up to phrase match or exact match. This immediately improves relevance by reducing the range of search queries that can trigger your ad. Understanding how phrase match and exact match differ is crucial for making these decisions effectively.

The mistake most agencies make is leaving keywords in the same match type forever because "that's how we set it up." Match types aren't set-it-and-forget-it—they're levers you should actively adjust based on performance data.

Next, look at ad group structure. If your refresh candidate is buried in an ad group with 15 other keywords, create a dedicated single-keyword ad group (SKAG) or at least a tighter themed ad group. Learning how to cluster keywords by theme for ad groups lets you write ad copy that directly addresses the search intent behind that specific keyword, which typically boosts both CTR and quality score.

Landing page optimization is the third lever. Pull the search terms report for your refresh candidates and look at what people are actually searching for. Then ask yourself: does your landing page immediately address that search intent? If someone searches "affordable CRM for small business" and lands on your generic CRM homepage, that's a disconnect. Send them to a page specifically about pricing or small business solutions.

Finally, consider bid adjustments. Sometimes the issue isn't the keyword itself—it's that you're not bidding competitively enough to get quality traffic. If your keyword has a decent quality score but low impression share due to rank, try increasing your bid by 20-30% and monitor the results. Conversely, if you're in position 1-2 but getting lots of clicks with no conversions, try lowering your bid to position 3-4 where traffic is often more qualified.

Give your refreshed keywords at least 30 days to show results before making another decision. Track them separately so you can measure whether your optimization efforts actually moved the needle. If they still underperform after a refresh attempt, it's time to move them to the prune list.

Step 5: Prune Keywords That Are Draining Budget

Pruning is the part everyone hesitates on, but it's often the most impactful action you can take. Every dollar you stop wasting on underperformers is a dollar you can reinvest in your proven winners.

First decision: pause or remove? I typically pause keywords rather than delete them outright. Pausing preserves your historical data and lets you reactivate the keyword later if circumstances change (new product launch, seasonal shift, etc.). The only time I fully remove keywords is when they're completely irrelevant to the business and will never make sense to run. If you need guidance on removal, check out how to delete keywords in Google Search.

Here's the critical step most people skip: add your pruned keywords to negative keyword lists. If you just pause a keyword that was running on broad match, it can still trigger through other broad match keywords in your account. You haven't actually stopped the bleeding—you've just made it harder to track.

When you prune a keyword, add the exact match version to your negative keyword list at the campaign or account level. This ensures it can't sneak back in through another keyword's matching behavior. Understanding how match types work for negative keywords is essential for this step.

Document everything. Create a simple note in your spreadsheet explaining why you pruned each keyword: "Spent $340 over 90 days, zero conversions, QS 3/10, triggering mostly irrelevant search terms." This documentation is invaluable when you're reviewing your account six months later or when a stakeholder questions why a keyword was removed.

Now here's the fun part: reallocate that saved budget. Calculate how much monthly spend you're cutting by pruning these keywords, then systematically increase bids on your top performers. This is where pruning becomes actively profitable, not just defensive—you're not just stopping losses, you're amplifying wins.

In most accounts I manage, a single pruning session frees up 15-25% of the keyword budget. That's significant money you can push toward keywords with proven ROI, often doubling down on your best performers and scaling them more aggressively than you could before.

Step 6: Set Up a Recurring Audit Schedule

Keyword pruning isn't a one-time project—it's ongoing maintenance that should happen regularly. The accounts that perform best over time are the ones with systematic, recurring optimization processes built in.

How often should you audit? It depends on your account size and spend velocity. For accounts spending $5,000+ per month, I recommend monthly keyword audits. For smaller accounts ($1,000-$5,000/month), every 60 days works well. If you're spending under $1,000/month, quarterly audits are probably sufficient—you need enough data volume between audits to make meaningful decisions.

Create a simple checklist you follow each audit cycle:

1. Export keyword performance data for the audit period

2. Pull Search Terms Report for the same timeframe

3. Apply your underperformance criteria and categorize keywords

4. Implement refresh actions (match type changes, ad copy updates, bid adjustments)

5. Pause prune candidates and add them to negative lists

6. Document all changes and reallocate freed budget

7. Set calendar reminder for next audit

Track improvements over time to prove the ROI of your optimization work. Create a simple dashboard that shows total keyword count, average CPA, and wasted spend percentage before and after each audit. This data becomes incredibly valuable when you're justifying your management fees or explaining to stakeholders why ongoing optimization matters.

There are also warning signs that indicate you need an immediate audit outside your regular schedule. If your CPA suddenly spikes 30%+ without a clear external reason, that's a red flag. If you launch a major campaign or product line, audit keywords 2-3 weeks after launch to catch early underperformers. If you make significant changes to your landing pages or offer, review keyword performance to see how it impacts conversion rates. For new launches, having a solid strategy for picking keywords for a new product launch can prevent many pruning headaches down the road.

The beauty of a recurring schedule is that each audit gets faster and easier. You're making incremental improvements rather than trying to fix months of accumulated waste. Your account stays lean, your budget stays focused, and your performance trends upward consistently rather than in sporadic jumps.

Putting It All Together

Let's recap the complete keyword refresh and pruning process in a quick, scannable checklist you can reference during your next audit:

Step 1: Export keyword data covering 30-90 days (include impressions, clicks, CTR, conversions, cost, CPA, quality score)

Step 2: Define clear underperformance criteria based on your campaign goals (e.g., 100+ clicks with zero conversions, CPA 3x above target)

Step 3: Categorize each keyword as Refresh (has potential, needs optimization), Prune (draining budget, no potential), or Monitor (insufficient data)

Step 4: Refresh promising keywords by adjusting match types, improving ad group structure, optimizing landing pages, or adjusting bids

Step 5: Pause underperformers, add them to negative keyword lists, document your reasoning, and reallocate saved budget to top performers

Step 6: Schedule recurring audits (monthly for high-spend accounts, 60-90 days for smaller accounts) and track improvements over time

Remember: keyword pruning isn't about being ruthless for the sake of it. It's about focusing your budget where it actually drives results. Every account has winners and losers—your job is to identify them quickly, double down on what works, and cut loose what doesn't.

The most important thing is to start. Don't wait for the perfect criteria or the perfect timing. Pull your data, sort by highest spend, and start making decisions on your top 20 budget consumers. That alone will often uncover 10-15% wasted spend you can immediately redirect.

For agencies managing multiple client accounts, this process can feel overwhelming when you're doing it manually across dozens of campaigns. The good news? Tools exist to speed this up dramatically. Start your free 7-day trial with Keywordme and optimize Google Ads campaigns 10X faster—without leaving your account. Remove junk search terms, build high-intent keyword lists, and apply match types instantly, right inside Google Ads. No spreadsheets, no switching tabs, just quick, seamless optimization for just $12/month.

Whether you're optimizing manually or using tools to accelerate the process, the fundamental principle remains the same: regular keyword maintenance separates profitable campaigns from money pits. Make it part of your routine, track your improvements, and watch your account performance climb steadily over time.