Free Keyword Research Template: Every Column Explained

A keyword research template is a spreadsheet where each row is one search query and each column records a fact you need before deciding to target it: where the keyword came from, how much demand it has, what already ranks, and what it’s worth to your business. My free keyword research template below has 14 columns, five filled example rows and a CSV you can paste into Google Sheets or Excel. Read the two tables side by side and you’ll see why each column exists.

This sheet covers the research and scoring stage only. Deciding which page each keyword belongs to is a separate job, keyword mapping, and it starts once this sheet is done.

The Keyword Research Template at a Glance

I split the template into two views so it fits on screen. The left side (columns A to H) holds research facts. The right side (columns I to N) holds your decisions. The example rows use a made-up online coffee roaster, and every volume and score in them is illustrative. They show the format, not real search data.

Left side: the research columns

Right side: the decision columns

Here’s the full template as CSV. Paste it into cell A1 of a tab named Keywords, then split it into columns if it lands in one (Data > Split text to columns in Sheets, Data > Text to Columns in Excel).

Keyword,Source,Date Pulled,Volume,Trend,Intent,SERP Check,KD,Business Value,Ranking Chance,Priority,Cluster,Target Page,Notes
how to make cold brew coffee,Autocomplete,2026-09-12,10K-100K,Peaks in summer,Informational,"Recipe posts, 2 videos",38,1,2,2,Brewing guides,New guide,Low buying intent. Link to cold brew beans
best coffee beans for espresso,Keyword Planner,2026-09-12,1K-10K,Flat,Commercial,"Blog roundups, 1 retailer list",29,3,2,6,Espresso,New roundup,Include our own beans honestly
ethiopian yirgacheffe coffee beans,Competitor pages,2026-09-14,1K-10K,Flat,Transactional,Product pages from roasters,18,3,3,9,Single origin,Existing product page,Improve product copy first
coffee subscription,Keyword Planner,2026-09-12,10K-100K,Rises in December,Commercial,"Big brand pages, review roundups",61,3,1,3,Subscriptions,Subscription page,Revisit in 6 months
why does my espresso taste sour,Search Console,2026-09-15,100-1K,Flat,Informational,"Forum threads, 3 short guides",12,2,3,6,Espresso,New guide,We already get impressions

Why Do the Source and Date Columns Come First?

Look at the Source column in the left table. Row 5 came from Search Console, which means real people already found this site through that query. Rows 2 and 4 came from Keyword Planner, which means they’re estimates. Those two kinds of evidence deserve different levels of trust, and without a Source column you can’t tell them apart a month later.

Date Pulled matters for the same reason. Numbers drift. Tools refresh their databases on their own schedules, seasonal queries swing month to month, and a volume figure you copied in January can be badly out of date by the time you actually brief the article in June. When I open an old sheet and see no dates, I treat every number in it as a guess.

If you have an existing site, Search Console is the best source you’ll get for free. I walk through that process in my guide to Google Search Console keyword research.

The Demand Columns: Volume and Trend

The Volume column in the example shows ranges like 1K–10K instead of single numbers. That’s deliberate. Google’s Keyword Planner help says search counts are averaged over a 12-month period by default, and Keyword Planner can display a range such as 1K–10K instead of one number. Paid tools each model volume their own way, so the same keyword can show very different figures depending on where you look. Write down what the tool gave you, name the tool in Source, and stop pretending the number is precise.

Trend is the column most templates skip. Google’s Trends FAQ explains that its figures are scaled on a range of 0 to 100 relative to all searches, so Trends shows shape, not size. Use it to answer one question: when does interest peak? In row 1, a summer peak tells you to publish the cold brew guide by spring, not in July.

Is the SERP Check Column Worth the Extra Time?

Yes. It’s the column I’d keep if I could only keep one. For every keyword, search it in a private window and write down what the top five results actually are: product pages, roundups, videos, forums, big brands.

Compare rows 2 and 3 in the left table. Both are about coffee beans, but row 2’s results are blog roundups while row 3’s are product pages. That single observation tells you row 2 needs an article and row 3 needs a product page. No volume or difficulty score gives you that. Not even close.

The KD column sits at the end on purpose. Difficulty scores come from third-party tools, each tool calculates them its own way, and many lean heavily on the backlinks of pages that already rank. I keep KD as a tiebreaker only. The SERP Check column tells me far more about my real chances, especially the “forum threads” note in row 5, which usually signals that nobody has written a good answer yet.

Correct vs Incorrect: The Same Row Filled Two Ways

Here’s row 4 filled in badly, then filled in properly. Read the two rows column by column.

The incorrect row looks precise and scores a perfect 9. But the volume has no source, nobody looked at the results page, and the Ranking Chance of 3 is wishful thinking. The correct row shows that the page one results are dominated by large subscription brands, so a small roaster’s chance drops to 1 and the Priority falls from 9 to 3.

Same keyword. Opposite decisions. The only difference is two columns that took about 2 minutes to fill, and that is the whole argument for using a keyword research template instead of a loose list pasted from whichever tool happened to be open that afternoon.

The Column Most People Get Wrong: Priority

Priority is the product of two scores, each on a 1–3 scale. Business Value asks how close the keyword is to a sale or lead: 3 for buying intent on something you sell, 1 for loosely related traffic. Ranking Chance asks whether a site like yours can realistically reach page one, based on the SERP Check.

Why multiply? Because a zero-chance keyword shouldn’t rank high just because it’s valuable. Look at row 4 again: value 3, chance 1, priority 3. Adding would give it 4 out of 6 and push it up the list. Multiplying keeps it near the bottom, where it belongs for now.

My honest take: don’t build a clever weighted formula with ten inputs. I’ve seen those sheets, and nobody trusts the output because nobody can explain it. Two scores you filled in yourself, multiplied, beat a formula you can’t defend.

What Formulas Keep the Sheet Clean?

Most keyword lists are merged from several tools, so they’re full of duplicates that differ only by capital letters or a trailing space. Google’s help page for the UNIQUE function warns you to “ensure that cells including text do not have differing hidden text” such as trailing spaces. That’s why the formulas below clean the text before removing duplicates.

These assume the CSV layout above, on a tab named Keywords, with Business Value in column I, Ranking Chance in column J and Priority in column K.

Priority (put in K2 and fill down, Sheets and Excel):
=IF(COUNT(I2:J2)<2,"",I2*J2)

Duplicate flag (put in O2 and fill down, Sheets and Excel):
=IF(COUNTIF($A$2:A2,A2)>1,"Duplicate","")

Clean, deduplicated keyword list on a new tab (Google Sheets):
=ARRAYFORMULA(UNIQUE(FILTER(LOWER(TRIM(Keywords!A2:A1000)),Keywords!A2:A1000<>"")))

Same list in Excel 2021, Excel 2024 or Microsoft 365:
=UNIQUE(FILTER(LOWER(TRIM(Keywords!A2:A1000)),Keywords!A2:A1000<>""))

All scored rows sorted by Priority, highest first (Google Sheets):
=SORT(FILTER(Keywords!A2:N1000,Keywords!K2:K1000<>""),11,FALSE)

Same sort in Excel 2021, Excel 2024 or Microsoft 365:
=SORT(FILTER(Keywords!A2:N1000,Keywords!K2:K1000<>""),11,-1)

The duplicate flag marks the second and later copies of a keyword, so you can filter for “Duplicate” and delete those rows. COUNTIF ignores letter case, which is exactly what you want here. UNIQUE, FILTER and SORT need Excel 2021, Excel 2024 or Microsoft 365. Older desktop versions can use Remove Duplicates on the Data tab instead.

How Do You Adapt the Keyword Research Template for an Existing Site?

A site that already ranks has data a new site doesn’t. Use it. Add three columns after Keyword: Impressions, Clicks and Average Position, all pulled from the Search Console Performance report. Then the sheet shows which keywords you’re already close on, which is usually where the fastest wins are.

One limit to know. When you export from the Search Console interface, Google’s export documentation says the data is truncated to 1,000 rows. For a small site that’s plenty. For a bigger one, filter by page or folder and export in batches, or use the Search Console API.

What Should You Do With the Rows You Reject?

Don’t delete them. Move them to a second tab called Parking Lot and add a column called Reason. I use four fixed values so the tab stays filterable:

• Wrong intent: the results page wants something you don’t offer, like a recipe when you sell beans.
• Out of scope: a real query, but outside what the business does.
• No chance yet: page one is all big brands or sites with far stronger link profiles.
• Seasonal: worth doing, just not now. Add the month to revisit in Notes.

In my experience, a good share of those rows come back. A keyword with Ranking Chance 1 today can move to 2 once your site earns a few links, and a “no chance yet” reason tells you exactly why you parked it. When I review a sheet every 3 months, the Parking Lot is the first tab I open.

What the Two Tables Tell You Together

Read left to right, each row tells a short story: where the keyword came from, whether demand is real and seasonal, what Google already rewards, and what the keyword is worth to you. If any column in that chain is empty, the Priority number at the end is a guess. A filled keyword research template won’t write your content, but it will stop you from spending three weeks on a page that was never going to rank.

If you’d rather not spend a week filling one of these in, our keyword research service delivers the finished sheet, and you can try it with 5 keywords free. And once your top rows are clear, the next step is grouping them into pages; my topical map template picks up exactly where this sheet ends.

Frequently Asked Questions

How Many Keywords Should Go in a Keyword Research Template?

Enough to cover the topics you can realistically publish on in the next 6 months. For most small businesses that’s 50 to 200 rows after deduplication. A list of 5,000 keywords you’ll never score is worse than 80 you actually understand.

Can I Fill It In Without Paid SEO Tools?

Yes. Search Console, Google autocomplete, Google Trends and competitor pages are free, and my free keyword gap walkthrough shows how I pair a Search Console export with a competitor’s pages to spot what they cover and you don’t. Keyword Planner is free too, but Google’s help page says you need to finish Google Ads account setup, including billing information, to use it. Put “free source” data in the Source column and your Priority scores still work.

Does This Sheet Replace a Keyword Map?

No. This sheet decides which keywords are worth pursuing. A keyword map then assigns each chosen keyword to one URL, and that’s a different process with its own rules. The Target Page column here is only a first guess to carry forward.

How Often Should I Update the Sheet?

Every 3 months for most sites. Refresh the Search Console columns, re-check the SERP for your top 20 rows, and review the Parking Lot. Update the Date Pulled cell whenever you replace a number.

Last updated: September 2026 by Mizanur Rahman