Where to start: the customer’s language
Most businesses pick keywords from their own product names while customers type something else. A clinic owner says “hair removal laser” and the customer writes “laser price near me”. Before any tool, review five real sources: WhatsApp and DM messages, phone questions, customer reviews, phrases your salespeople hear most, and competitor pages.
The output is a raw list of 50 to 100 phrases. Do not worry about duplicates or typos; the typos often show how people really write. In Persian, watch for variants such as with or without a half-space and the Arabic versus Persian forms of “ی” and “ک”.
Expanding the list with free tools
- Google autocomplete: type your phrase and note the suggestions; try appending letters of the alphabet.
- “People also ask”: a good source for headings and FAQ sections.
- Search Console: if you have a site, the Performance report shows phrases you already appear for; positions 8 to 20 are the best quick wins.
- Google Keyword Planner: gives approximate volume ranges. Do not read the numbers as exact; use them to compare phrases relatively.
Search intent: four buckets
Every phrase carries an intent, and your page must satisfy it. Look at the first-page results: if they are all articles, your sales page probably will not rank.
| Intent | Example | Page type |
|---|---|---|
| Informational | “what is local SEO” | Blog article |
| Commercial | “best site builder for a shop” | Comparison article |
| Transactional | “order website design Tabriz” | Service or product page |
| Navigational | “TETA Team” | Brand page |

Choosing and prioritising
Your list is now large and your resources small. Build a simple score that combines commercial value, approximate volume and difficulty. No fancy tool is needed; a CSV file and a few lines of Python are enough (for heavier work see data analysis with Python and pandas).
import pandas as pd
# keywords.csv columns: keyword, volume, difficulty (0-100), intent_fit (1-3)
df = pd.read_csv("keywords.csv")
df["score"] = df["volume"] * df["intent_fit"] / (df["difficulty"] + 10)
print(df.sort_values("score", ascending=False).head(20))Set intent_fit yourself: 3 means the phrase is close to a sale, 1 means purely informational. The 10 in the denominator avoids division by zero and an unfair bonus for near-zero difficulty. The formula is not sacred; it gives a first ordering and the final judgement is yours.
Do not underrate long-tail phrases
A phrase such as “e-commerce website with online payment in Tabriz” has little volume but also less competition, and its visitor is usually closer to buying. For a new site, ten precise long-tail phrases often beat one broad phrase you cannot compete for.
Mapping keywords to pages
Build a simple table: page URL, primary keyword, and two to four related phrases. If two pages share a primary keyword they compete with each other (cannibalisation); merge them or refocus one on a different phrase.
Use the primary keyword naturally in the title, H1, first paragraph and URL, and spread related phrases across subheadings. Writing the page itself is the topic of content SEO writing, and the links between pages are covered in internal linking.
Review and measurement
Six to eight weeks after publishing, check in Search Console which phrases the page appears for. You will often see phrases you did not predict; they feed the next round of research. Rewrite pages with many impressions and few clicks using a better title and description. If you would rather hand research and content to a team, our SEO service covers exactly this cycle.
Frequently asked questions
Which free tools can I use for keyword research?
Google autocomplete, “People also ask”, Google Search Console and Keyword Planner are enough to start. They let you find phrases, intent and quick wins at no cost.
How many keywords should one page target?
Each page has one primary keyword plus a few related phrases. The number matters less than covering one clear search intent fully without competing with another page.
Does low search volume mean a keyword is worthless?
No. Precise low-volume phrases usually have less competition and a higher chance of converting. A phrase’s value depends on how close it is to the customer need, not only on the volume number.
How often should keyword research be repeated?
For a small business, a quarterly review in Search Console and a fuller pass every six months is usually enough. Repeat it sooner if your product or market changes.

Comments
No comments yet. Be the first to share your thoughts.