1, step 1: collect keywords;
(1, collecting keywords that are constantly derived;
(2) Most people don't know the existing keywords (I don't know that these words have conversion rate);
(3) At present, a better collection tool: Baidu bidding background; (Methods: Collect related words through the bidding background of core keywords)
2. Step 2: screen keywords;
(1, cut off some words that obviously don't meet the needs of users.
Case: Among the keywords related to medical plastic surgery, acne products (such as acne masks) obviously do not meet the needs of users in the medical industry. They are only targeted search products, which will not transform medical plastic surgery and have a weak effect on traffic conversion, and should be removed from the shelves.
3. Step 3: keyword test;
4. Step 4: Keyword classification and reasoning;
1, because keywords are constantly derived. (New words are constantly derived and old words are constantly disappearing; )