A free keyword research routine needs four tools and one sheet. Search Console tells you what you already appear for, Google Trends shows which way interest is moving, Keyword Planner offers ideas and monthly search estimates, and autocomplete exposes phrasings you had not considered. Each answers a different question, and a sheet that records which tool said what keeps them from contradicting each other. A solo site owner can run the routine in an hour a week. Examples come from one niche, indoor plants.
One question per tool
Asking all four tools the same question produces conflicting answers. Give each one a separate job and merge the results.
Table: Tool, the question it answers and its limit
| Tool | Question | Limit |
|---|---|---|
| Search Console Performance report | Which queries showed my site? | Only queries where your site appeared |
| Google Trends | Is interest rising, falling or seasonal? | Relative scale, low-volume terms show as zero |
| Keyword Planner | What new keywords exist and what are monthly search estimates? | An ads tool with account setup |
| Autocomplete | How do people finish this phrase? | No volume, varies by person |
Using a tool outside its job
Reading volume from Trends, or demand size from autocomplete, is the easiest mistake. Trends values are relative, and autocomplete is a suggestion list. Use those two for direction and ideas, not for counting.
Why the order matters
Practical suggestion: Search Console first, then Trends, then Keyword Planner, with autocomplete last. Starting with Search Console puts real data first. Trends sets direction, Planner adds new keywords and estimates, and autocomplete widens the list. Reverse the order and your own data drowns among outside guesses.
What you already earn, from Search Console
The Performance report (Opens in a new tab) comes with tabs for queries, pages, countries, devices, search appearance and dates. Grouping by queries groups data by what users typed, and each row carries clicks, impressions, click-through rate and average position. The default view covers the past three months.
What to look for first
Sort by impressions and look for two groups. One is queries with high impressions and low clicks, where your page shows up but is not chosen. The other is topics that appear in the table yet have no page of their own. Both come from real search behavior, which makes them the soundest leads in a free stack.
The boundary of the report
The report only contains queries where your site appeared in results. A topic where you do not show is absent. So Search Console measures the depth of what you cover and does not find new topics; for those, turn to Trends, Planner and autocomplete.
Separating search types
The search type filter splits web, image, video and news. For keyword research keep the web filter on. Mixing image or video data into the same table inflates impressions in a misleading way.
Reading average position correctly
The report defines average position as the average position of the topmost result from your site, and in the table it is the average for the URL or grouping shown in that row. Even if two of your pages show for one query, the row represents the topmost result. Practical suggestion: do not read position as a ranking promise; read it together with impressions and clicks.
Reading query and page together
Select a query and switch to the Pages tab to see which URL showed. Write the page address beside the query in the sheet. The decision "I already have a page for this" or "this keyword shows on the wrong page" is then ready in the table.
Exporting
Reports offer an export button for chart and table data, and values shown as "~" or "-" become zeros in the download. Sort the download by impressions and carry the first hundred queries into your research sheet.
Reading demand with Trends and Planner together
Trends gives direction over time and Planner gives monthly search estimates. Read together, they say how large a topic is and where it is heading.
The Trends scale
According to the Trends data FAQ (Opens in a new tab), each data point is divided by the total searches of the geography and time range it represents, and the results are scaled from 0 to 100 based on a topic's proportion to all searches on all topics. Two regions with the same relative interest can have different total volumes. A reading of 100 is therefore not a search count. Check on the screen what 100 refers to in the range you selected. Practical suggestion: read the chart's own explanation before treating a 100 from one period or term as equal to a 100 from another.
Sampling, zeros and time zones
The page says Trends uses a sample of searches, which it considers sufficient given the billions of searches processed daily. Trends only shows data for popular terms, so low-volume terms appear as "0", and repeated searches from the same person in a short period are removed. Ranges of 30 days or longer use UTC, while ranges of 7 days or shorter use your local time zone. A zero on a narrow term, then, does not mean nobody searches for it.
Comparing terms
The FAQ suggests relative comparison to judge popularity. Practical suggestion: plot your own terms against each other, such as "low light houseplants" and "pet friendly houseplants", and read which one holds more interest and how each moves across the year. For topic choice with Trends, see Google Trends for content topic selection.
Keyword Planner is an ads tool
Google's Keyword Planner help page (Opens in a new tab) describes it as a Google Ads tool for researching keywords for Search campaigns. It can suggest keywords, estimate monthly searches and average cost, group keywords by theme and forecast performance, and it notes that actual results depend on factors such as your bid, budget and ad quality. Treat its figures as ads-context estimates and not as an organic traffic forecast.
Account setup and billing
The page says you must complete account setup by entering billing information before using basic features such as keyword ideas. Access to the tool depends on that setup. If your interface is not in English, label names may look different. In English they read "Discover new keywords" and "Get search volume and forecasts".
Discovering keywords
"Discover new keywords" generates ideas from seed words or a site, and entering a keyword and a URL together returns a larger set. For a plant shop, you could enter your category page address together with the word "monstera".
Uploading a list for volume and forecasts
"Get search volume and forecasts" returns volume and forecast data for a list you enter or upload. For a simple list the file needs one column with the header "Keyword". Words gathered from Search Console and autocomplete can be merged into one list and queried together.
Filters and the refine panel
"Add filter" narrows results by competition, impression share and keyword text, and the "Refine keywords" panel groups ideas by theme, brand or category so you can untick irrelevant groups.
Limits worth writing down
The page states that keywords with very low search volume, or considered sensitive, are not discoverable or forecastable. The documentation has no statement on how search volume ranges are displayed, so read the range notes in the interface itself. Forecasts refresh daily, are based on the last 7 to 10 days and are adjusted for seasonality. Google's pages can change, so check the current help page before building a process on any figure.
Choosing region and range
Trends data can be viewed worldwide or down to city-level geography. A plant shop that ships nationwide reads country level; one that delivers to a single city reads city level. Choose the time range by purpose: a longer range for seasonal shape, a shorter one for the interest in a new product.
Carrying Planner output into the sheet
Write the monthly estimate into the sheet as a rough class, not a number: low, medium, high. Practical suggestion: set the class boundaries yourself and note them in a corner of the sheet, so someone reading it two months later knows what the classes mean. Putting estimates beside real Search Console impressions also shows how well an estimate agrees with your site's situation.
Measuring one keyword in two places
If a keyword has impressions in Search Console and an estimate in Planner, compare the two. Impressions far under the estimate suggest your page is weak for that keyword. Impressions close to the estimate suggest it is well placed. The two figures do not measure the same thing, yet the ratio still points to priorities.
Widening the list with autocomplete
The completions that appear while you type show how people continue a phrase. They are an idea source and not a volume source.
Expansion method
Add each letter of the alphabet after a seed phrase and note the suggestions. Then try question stems such as "how to repot", "why does my" and "when to water". Record each suggestion in the sheet with "autocomplete" in the source column.
Suggestions differ by person
Suggestions can vary with location, language and history. Practical suggestion: try the phrase in a private window and in a signed-out browser, and trust phrases that show up in both lists more. A fuller method needs its own treatment.
One research sheet for everything
Merging the tools is where the value appears. Each keyword is a row, and each tool fills its own columns.
Table: Research sheet columns
| Column | Source | Purpose |
|---|---|---|
| Keyword | All tools | Single list |
| Origin | Typed by hand | Which tool surfaced it |
| Impressions and clicks | Search Console | Real data for your site |
| Direction | Trends | Rising, falling or seasonal |
| Monthly search estimate | Keyword Planner | Size idea |
| Intent label | Typed by hand | Question, comparison, price, local |
| Page decision | Typed by hand | Improve, create or skip |
A decision rule
Practical suggestion: a keyword with Search Console impressions goes to an existing-page improvement first. A keyword absent from Search Console but rising in Trends and present in Planner is a candidate for a new page. A keyword seen only in autocomplete waits as an idea until a second tool backs it.
Writing your own research questions
Before starting the weekly routine, write three or four questions: Which product category am I writing for? What does the reader ask about it? Which of my pages is weak? With what intent would a new page open? Placed at the top of the sheet, the questions stop the keyword list from growing without purpose and show that each row serves a question.
A one-hour weekly routine
The hour is not split evenly. Search Console gets the biggest share because its data is the soundest basis for decisions.
Table: The weekly hour
| Minutes | Task | Output |
|---|---|---|
| 15 | First fifty queries by impressions in Search Console | New rows in the sheet |
| 10 | Compare two or three terms in Trends | Direction column |
| 15 | Upload the list to Keyword Planner | Estimate column |
| 10 | Autocomplete expansion | New idea rows |
| 10 | Intent label and page decision | This week's single task |
One task per week
End the hour with one job: improve a page, open a draft or park an idea. A routine that produces four tasks gets dropped within weeks, while a small steady one lasts.
When the routine slips
A busy week can derail the routine. Practical suggestion: do not try to make up the missed week. Next week, run only the Search Console step and leave the other tools for the week after. Cramming three weeks of work into one day lowers the quality of the sheet and leads to abandoning the routine. Marking the skipped week with a date in the sheet helps you remember the gap a month later.
Example scenario: an indoor plant shop
Example scenario: a shop sells monstera, snake plants and pothos with pots. In the Performance report the owner sees impressions for a snake plant care page. In Trends two terms move differently across the year. Keyword Planner returns new variants for a pot category from an uploaded list. Autocomplete suggests several question stems for a care guide.
Table: Example scenario, tool output and decision
| Tool | Finding | Decision |
|---|---|---|
| Search Console | Snake plant care page earns impressions, few clicks | Revisit title and opening paragraph |
| Trends | The two terms follow different paths | Move the rising term's page forward |
| Keyword Planner | New pot variants appear | Consider a pot category page |
| Autocomplete | Question stems hint at a guide | Park the idea and confirm with a second tool |
The owner picks the first row as the week's task. The other three findings wait in the sheet and spread across later weeks.
Checklist
Checklist: free tool workflow
- Is Search Console set to web search with the date range noted?
- Were Trends terms compared in the same range and region?
- Is the Planner account set up with billing information?
- Are zeros in Trends read as low volume and not as no searches?
- Does every keyword record its origin?
- Does every keyword have an intent label and a page decision?
- Is there a single task for the week?
Missteps
Reading Trends as volume
Trends gives a relative scale from 0 to 100. A reading of 80 does not mean a number of searches.
Reading Planner as an organic forecast
Planner is built for Search campaigns, and its estimates arise in an ads context. Use them as a direction signal in SEO decisions.
Treating zero as absence
Trends shows low-volume terms as "0", which does not prove nobody searches them.
Leaving the origin blank
Without an origin column you cannot interpret the sheet three months later.
Producing an excess of tasks
One task keeps the routine alive. For further reading, see how to do keyword research step by step and the note on keyword research for Turkish suffixes and characters.
Data hygiene: matching one keyword across tools
Words from the four tools do not arrive in one shape. Search Console returns queries in lowercase and inflected form, Planner suggests base forms, and autocomplete gives long phrases. If these differences go unmanaged, one topic gets counted as three rows.
Cleaning steps
- Lowercase every keyword and trim leading and trailing spaces.
- Merge spelling variants into one row, for instance "houseplant" and "house plant".
- Remove duplicates, but record both tools in the origin column.
- Fold plural and singular forms under one head keyword.
- Sort by impressions and review the first hundred rows.
After cleaning, the sheet is shorter and easier to read. The aim is not linguistic perfection; it is to stop the sheet from counting one topic three times.
The illusion of duplicate rows
When one topic sits on three rows, the sheet suggests three separate opportunities. Practical suggestion: add a group column beside each head keyword and write one decision per group. "Low light plants", "plants for low light" and "best low light houseplants" then share one decision.
Where the free stack runs out
Free tools are a sound start for a small or mid-sized site, but they do not cover every need. Writing the gaps down shows when another source is worth having.
Table: Needs the free stack leaves open
| Need | What the free stack offers | What to do |
|---|---|---|
| Query data for a new site | Search Console is empty until data accrues | Start with Trends and autocomplete |
| Which keywords a rival appears for | Rival data is outside the job of these four tools | Read rival pages by hand |
| Exact volume | Planner gives estimates, read the interface for ranges | Treat volume as direction |
| Competitive difficulty | Difficulty is judged by reading the results page | Read the results page by hand |
When to consider a paid tool
Practical suggestion: if the weekly routine has run steadily for three months and the same missing data blocks your decisions each time, weigh a paid tool. Buying one before the routine exists ties the tool to an empty sheet. Build the method first and choose the tool against a real need.
From the sheet to a content plan
The sheet is not only a word list; it should become next month's publishing plan. The conversion takes three moves: grouping, prioritizing and tying each group to a page.
Grouping and prioritizing
Split groups by intent label: question, comparison, price, problem solving. Priority comes from Search Console impressions, the Trends direction and whether a page already exists. A group with a weak page and rising impressions moves to the top. A group seen only in autocomplete joins the waiting list.
One page per group
Keywords sharing an intent gather on one page. Squeezing two different intents onto one page blurs which question the page answers. When deciding, look at the keyword cluster; the method of building clusters deserves its own guide.
Writing the working title
The working title for each group should mirror the question the reader asks. A title such as "How to choose a pot with drainage" carries the intent of the matching group. Derive the title from the group's intent, not from the word list.
A four-week setup calendar
The first four weeks go to establishing the routine, and each week has a single focus.
Table: Four-week setup
| Week | Focus | Output |
|---|---|---|
| 1 | First hundred Search Console queries and sheet setup | Blank template and first rows |
| 2 | Trends comparison and direction column | Ten groups with a direction |
| 3 | Planner account and list upload | Monthly estimate column |
| 4 | Autocomplete and intent labels | First content plan |
At the end you hold a filled sheet and a first plan. Later weeks refresh the plan and add new keywords.
Reporting the findings
If someone else will read the sheet, a short summary helps. It states what each tool found, what will be done and what is waiting, in three lines.
Report skeleton
Line one gives this week's finding. Line two gives the task. Line three names the parked ideas and the second tool that could back them. Share the raw sheet as a link and keep the summary readable.
The monthly review
Once a month, reread the sheet. Check which parked ideas gained support from a second tool. Move ideas that no tool backed for three months into an archive tab. That keeps the list short and actionable.
Frequently Asked Questions
Are free tools enough for keyword research?
Practical suggestion: they are a sound start for a small or mid-sized site. Search Console gives real data, Trends gives direction, Keyword Planner gives monthly estimates and autocomplete widens the list.
Is Keyword Planner really free?
The help page says you must complete account setup by entering billing information before using basic features.
What does 100 mean in Trends?
Values are divided by total searches for the geography and time range and scaled from 0 to 100. A 100 is not a search count; check on the screen what 100 refers to in the range you selected. Practical suggestion: read the chart's own explanation before comparing 100s across periods or terms.
Why does a term show zero in Trends?
The page says Trends only shows data for popular terms, so low-volume terms appear as "0".
Can Search Console find new topics?
The Performance report covers only queries where your site appeared. New topics come from Trends, Planner and autocomplete.
How long should a weekly session last?
Practical suggestion: one hour and one resulting task. A short routine is easier to sustain than a rare long one.
Conclusion
A free keyword research routine gives each tool a single job. Search Console supplies real queries, Trends supplies relative direction, Keyword Planner supplies monthly search estimates and autocomplete supplies ideas. The results meet in one sheet where every keyword records its origin, an intent label and a page decision, and the week ends with one task. Limits belong in the sheet too: Trends is relative, Planner is an ads tool and Search Console covers only queries where you appeared.
Turning the sheet into pages takes time. For well planned, original articles on the topics you pick, see our content writing service.
Post a comment
Please log in to leave a comment. Login