How to Forecast Conversions Per Keyword Using Google Keyword Planner (Step-by-Step)

TL;DR: Google Keyword Planner's forecasting feature lets you estimate how many conversions a keyword might drive before you spend a cent. This guide walks you through exactly how to use it—from pulling up the right report to interpreting the numbers and turning them into a real bidding strategy. Whether you're a freelancer pitching a new client, an agency building out a campaign, or a marketer trying to justify budget, this workflow will save you from guessing. We'll cover the actual steps inside Planner, what the forecast data means in practice, where it tends to mislead you, and how to sanity-check the numbers before committing spend. No fluff, no vague advice—just a clear, repeatable process you can run in under 30 minutes.

In most accounts I audit, keyword lists are built on gut feel and search volume alone. Advertisers pick terms that look right, set a bid, and hope the conversions follow. The problem is that hope isn't a bidding strategy. Planner's forecasting tool gives you something better: a directional model of what your keywords might actually deliver before you touch your budget.

The catch? Most people either don't know the conversion forecast feature exists, or they run it wrong and end up with numbers that are more misleading than helpful. This guide fixes that.

Step 1: Access the Forecasting Tool Inside Google Keyword Planner

Head to your Google Ads account and navigate to Tools → Planning → Keyword Planner. You'll see two options on the landing screen: "Discover new keywords" and "Get search volume and forecasts." These are not the same thing, and picking the wrong one is the most common starting mistake.

"Discover new keywords" is for research and ideation. It's useful for expanding a keyword list, but it won't give you the forecast columns you need. "Get search volume and forecasts" is the one you want. This is the tool that models impressions, clicks, cost, and conversions based on your bid inputs. If you're new to the platform, our guide on how to use Google Keyword Planner covers the full interface in detail.

Once you're in, paste in your keyword list directly or upload a CSV. You can run forecasts on a single keyword or hundreds at once. If you're working with a large list, the CSV import saves time and keeps things clean.

Before you hit the forecast button, configure three settings that dramatically affect your output:

Location: Set this to match where your actual campaign runs. A forecast for "emergency plumber" in London will look very different from the same keyword in a rural market.

Language: Match your campaign's language targeting. Leaving this on the default when you're targeting a specific language skews volume estimates.

Network: This one trips people up constantly. The default is often "Google Search Network" plus search partners. If your campaigns only run on Google Search, switch it to Google Search only. Running a forecast on a broader network than you actually use inflates your projected impressions and clicks.

Once you've set these correctly and run the forecast, you should land on the Forecasts tab showing columns for impressions, clicks, cost, CTR, and average CPC. If you're seeing this table, you're in the right place. Conversion data won't appear yet—that comes in the next step.

Step 2: Configure Bids and Budget to Unlock Conversion Estimates

Here's why the Conversions column is blank by default: Planner needs a bid input to model traffic volume, and it needs your account's historical conversion rate to calculate estimated conversions. Without a bid, it can't estimate how much traffic you'd buy. Without conversion history, it has no CVR to apply.

To set your bid, look for the bid slider or the manual entry field at the top of the Forecasts view. You can set a single blanket CPC across your entire keyword list, or adjust bids per keyword if you want more granular control. Start with a blanket bid that reflects your typical Max CPC for the campaign type you're planning.

The relationship here is straightforward: a higher bid wins more auctions, which means more impressions, more clicks, and therefore more projected conversions. But the returns diminish as you push the bid up. You'll see this in the forecast curve—clicks increase steeply at first, then flatten out as you reach the ceiling of available traffic at that bid level. Understanding how to forecast keyword CPC accurately will help you set more realistic bid inputs from the start.

Next, set a daily budget. This allows Planner to calculate total projected conversions over a date range, which defaults to 30 days. If you're building a monthly forecast for a client proposal, keep the default. If you're modeling a shorter test period, adjust the date range accordingly.

Now, the important caveat: the Conversions column only populates if your Google Ads account has conversion tracking history. Planner uses your account's historical CVR to model conversion estimates. If your account is brand new or has no conversion data, that column will stay empty.

What do you do if your account is new? Don't guess randomly. Instead, use Planner's click estimates and apply a manual CVR based on industry benchmarks from credible sources like Google's own Think with Google research, WordStream's industry reports, or your own experience managing similar accounts. The formula is simple: Projected Conversions = Projected Clicks × Your Estimated CVR. We'll build this out properly in Step 5.

When the setup is correct, the Conversions column populates alongside Cost per Conversion. That's your signal to move to the next step.

Step 3: Read the Forecast Data Without Getting Fooled by It

This is where most advertisers go wrong. They see the Conversions column light up and take the numbers at face value. Before you do that, you need to understand what Planner is actually telling you—and what it isn't.

Here's a quick breakdown of each column and what it means in practice:

Impressions: How often your ad is estimated to show at your set bid and budget. This is a ceiling, not a guarantee.

Clicks: Estimated clicks based on projected impressions and an estimated CTR. Planner models CTR using historical data for similar keywords and ad positions. For a deeper look at how these numbers are calculated, see our guide on forecasting clicks and impressions from keywords.

Cost: Projected spend over the date range at your set bid and budget. Useful for budget planning but treat it as an estimate, not a bill.

CTR: The click-through rate Planner is modeling. Worth checking—if it looks wildly high or low compared to your actual account CTR, that's a signal to investigate.

Avg. CPC: The average cost per click Planner expects you to pay at your set Max CPC. This will typically be lower than your Max CPC due to the auction dynamic.

Conversions: This is the critical one. And here's the thing most people don't realize: Planner applies your account's blended historical CVR to every single keyword's projected clicks. It does not model per-keyword conversion rates. If your account converts at 3% overall, Planner applies that 3% to a branded keyword, a generic keyword, and an informational keyword alike. That's a significant limitation.

Cost/Conv: Projected cost per conversion at your set bid. This is your most actionable column for identifying which keywords are worth pursuing.

To get the most out of this data, sort the table by Cost/Conv ascending. This surfaces the keywords that are projected to deliver conversions most efficiently at your current bid. Then filter out keywords with very low projected clicks—say, fewer than 10 clicks over 30 days. Forecasts on low-volume keywords are statistically unreliable and can skew your decision-making.

The biggest mistake you can make here is treating this forecast as a guarantee. It's a directional model. Use it to prioritize, not to promise results to a client.

Step 4: Segment Keywords by Intent to Refine Your Conversion Estimates

Because Planner applies a blended CVR to everything, treating your entire keyword list as one homogenous group leads to bad decisions. A branded keyword will not convert at the same rate as a generic category keyword. An informational query will not convert like a high-intent transactional term. If you lump them together, your forecast will be wrong in both directions—optimistic on some terms, pessimistic on others.

The fix is to segment your keywords by intent tier and run separate forecasts with different CVR assumptions for each group. Here's a practical framework:

Branded keywords: Your brand name, branded product terms. These typically convert at a much higher rate than non-branded terms because the user already knows who you are. Model these separately with a higher CVR assumption.

High-intent transactional: Terms like "buy X now," "X free trial," "X pricing," or "hire X agency." The user is close to a decision. These warrant a higher CVR assumption than generic terms.

Mid-intent research: Terms like "best X for Y," "X vs Z," "X reviews." The user is evaluating options. These convert, but at a lower rate than transactional terms. Apply a moderate CVR assumption.

Low-intent informational: Terms like "what is X," "how does X work." These rarely convert directly. Model them with a low CVR or consider whether they belong in a conversion-focused campaign at all.

To put this in concrete terms: a keyword like "PPC management software free trial" is a high-intent transactional term. You'd model it with a higher CVR assumption. A keyword like "what is PPC management" is informational. If you apply the same CVR to both, you'll either overbid on the informational term or underbid on the transactional one.

Match type assumptions matter here too. Exact match on a high-intent term gives you tighter control over what search queries trigger your ad, which typically produces better conversion rates than broad match on the same term. Understanding how keyword match type affects Google Ads performance is essential when you interpret Planner's click estimates for broad match keywords—apply a more conservative CVR because the actual queries that trigger your ad will be more varied in intent.

Run separate Planner forecasts for each intent group with the corresponding bid assumptions. By the end of this step, you should have two to four segmented forecast groups with different projected conversion rates, giving you a much more realistic total picture of what your keyword list might deliver.

Step 5: Export and Build a Keyword-Level Conversion Forecast Model

Now it's time to pull everything out of Planner and build a working model you can actually use for decisions. In the top-right corner of the Forecasts view, you'll find the export button. Download the CSV.

When you open the file, you'll see more columns than you need. Here's what to keep:

Keyword: Obviously.

Match Type: Critical for understanding how Planner modeled the traffic.

Impressions, Clicks, Cost: Your core traffic and spend estimates.

Conversions and Cost/Conv: Planner's baseline estimates, which you'll be overriding with intent-specific CVRs.

Delete the rest. A cleaner spreadsheet means fewer mistakes.

Now add a manual CVR override column. Based on your intent segmentation from Step 4, assign a CVR estimate to each keyword. Label the column clearly—something like "Adjusted CVR (Intent-Based)." Then add a formula column for adjusted projected conversions:

Projected Conversions = Projected Clicks × Adjusted CVR

Add another column for adjusted Cost per Conversion: Projected Cost / Adjusted Projected Conversions. This gives you a keyword-level CPA estimate that accounts for intent, not just Planner's blended account average.

From here, you can calculate projected ROAS if you know your average order value or lead value: (Projected Conversions × Average Value) / Projected Cost. For each keyword, you now have enough data to make a go/no-go decision. Keywords with a projected CPA well below your target CPA are strong candidates to bid on. Keywords where the projected CPA exceeds your target need either a lower bid or a negative keyword treatment. Our guide on prioritizing keywords by ROI potential walks through a similar decision framework you can apply here.

Once you've identified your high-value keywords from this model, the next step is getting them into your campaigns quickly. This is where tools like Keywordme come in—it lets you add keywords directly into Google Ads and apply match types without leaving the interface, which means you go from forecast to live campaign without the usual export-import-reformat cycle.

Your success indicator here is a clean spreadsheet with each keyword's projected conversions, adjusted CPA estimate, and a clear go/no-go decision column.

Step 6: Validate the Forecast Against Real Performance Data

A forecast model is only as good as your willingness to test it against reality. This step is where you close the loop.

If you have existing campaign history, start here: pull the Search Terms Report in Google Ads and compare which terms are actually driving conversions against what Planner projected. Look for patterns. Are the high-intent terms you modeled at a higher CVR actually converting at a higher rate? Are the informational terms you flagged as low-intent showing up in your search terms with zero conversions and high spend? Learning how to optimize match types using the Search Terms Report can help you tighten up which queries actually trigger your ads during this validation phase.

If you're starting fresh with no campaign history, set a short test budget and run your top-forecast keywords for two to three weeks. Keep the keyword list tight—focus on your highest-priority terms from the forecast model. After the test period, pull actual CPA data and compare it to your Planner estimates.

There are two red flags to watch for during validation:

Actual CPA is significantly higher than projected: This often means Planner used a blended CVR that doesn't reflect the actual intent of those specific keywords. Your intent segmentation from Step 4 should help catch this, but if you see a consistent gap, revisit your CVR assumptions.

Actual click volume is much lower than projected: This usually signals a bid competitiveness issue. Your Max CPC may not be high enough to win the auctions Planner was modeling at. Adjust your bids in Planner, re-run the forecast at the higher bid level, and use the revised model to update your campaign structure.

The iteration loop is: run forecast → launch campaign → compare actuals → adjust assumptions → re-run forecast. Each cycle makes your forecast model more accurate because you're replacing Planner's blended CVR estimates with real data from your own account.

For agencies managing multiple accounts, this validation step is especially valuable. Spotting patterns across accounts—which keyword intent tiers consistently over-deliver or under-deliver against Planner forecasts—helps you build better baseline assumptions that make future forecasts more reliable from the start. Pairing this with a process to refresh and prune underperforming keywords ensures your keyword list stays aligned with what the data actually shows.

Putting It All Together: Your Conversion Forecasting Checklist

Here's the full workflow condensed into a quick-reference checklist:

1. Access the right tool: Use "Get search volume and forecasts" in Keyword Planner, not "Discover new keywords." Set location, language, and network to match your actual campaign settings.

2. Configure bids and budget: Enter a Max CPC bid and daily budget to unlock the Conversions column. If your account lacks conversion history, use manual CVR estimates from credible industry sources.

3. Read the data critically: Understand that Planner applies your account's blended CVR to every keyword. Sort by Cost/Conv, filter out low-volume terms, and treat the output as directional, not definitive.

4. Segment by intent: Group keywords into branded, high-intent transactional, mid-intent research, and low-intent informational tiers. Run separate forecasts per group with different CVR assumptions.

5. Build your model: Export the CSV, add an intent-based CVR override column, calculate adjusted projected conversions and CPA per keyword, and assign go/no-go decisions.

6. Validate against real data: Cross-reference your forecast with actual search term performance. Iterate on bid and CVR assumptions until your model reflects reality.

A few notes on when to re-run this process: before any new campaign launch, when entering a new market or location, and when seasonal trends shift significantly. A forecast built in January for a summer campaign needs to be refreshed closer to the launch date.

One final thing worth saying plainly: Planner forecasts are a starting point, not a finish line. Their real value is in prioritizing which keywords to test and setting realistic expectations with clients before any spend is committed.

Once your forecast model is built and you're ready to move from planning to execution, adding keywords, applying match types, and building negative lists is where the real time gets lost. Start your free 7-day trial of Keywordme and do all of that directly inside Google Ads, without spreadsheets, without switching tabs, and without the usual friction that slows down campaign launches. After that it's just $12/month—a straightforward trade for the time it saves.