透過集合功能整理內容
你可以依據偏好儲存及分類內容。
使用 Search Console 泡泡圖改善 SEO
分析搜尋成效資料一直是個挑戰,尤其當您有大量長尾查詢時,會更難以理解以及視覺化。泡泡圖可協助您瞭解,對您的網站來說有哪些查詢成效較佳,哪些有待改善。
如果您想測試本文討論的技巧,可以將資料連結至 Looker Studio,然後嘗試調整圖表設定。
如果您尚未讀過將 Search Console 連結至 Looker Studio,以及使用 Looker Studio 監控搜尋流量,建議參閱這兩篇文章,進一步瞭解在 Looker Studio 中可以利用 Search Console 執行哪些作業。
瞭解圖表內容
如果您擁有多項指標和維度,泡泡圖就是絕佳的視覺化圖表,可讓您更有效率地看出資料中的關係和模式。在這裡顯示的範例中,您可以在同一個檢視畫面中查看查詢和裝置維度的點閱率 (CTR)、平均排序和點擊次數。
本節將詳細說明部分圖表元素,釐清圖表含有哪些內容,不含哪些內容。
資料來源
本頁顯示的泡泡圖使用 Search Console 資料來源提供的「網站曝光」表格,其中包含依網站和查詢匯總的搜尋成效資料。
篩選條件和資料控制項
圖表提供五個自訂選項,協助您有效控管資料:
- 資料控制:選擇想要分析的 Search Console 資源。
- 日期範圍:選擇想要在報表中查看的日期範圍,系統預設顯示過去 28 天。
- 查詢:納入或排除要聚焦的查詢。您可以按照在 Search Console 中篩選資料的方式,輕鬆篩選資料。
- 國家/地區:納入或排除國家/地區。
- 裝置:納入或排除裝置類別。
軸
圖表中的軸代表平均排序 (Y 軸) 和網站點閱率 (X 軸)。軸有三個重大轉換:
- 倒轉 Y 軸方向:Y 軸顯示的是平均排序,倒轉代表 1 位於頂端。就大部分圖表來說,最佳位置在右上角,因此用圖表顯示平均排序時,倒轉 Y 軸較符合直覺。
- 對數刻度:兩個軸都採用對數刻度,可讓您深入瞭解圖表中極端值的查詢情形 (極低的點閱率、平均排序或兩者兼具)。
- 參考資料線:參考資料線的作用是醒目顯示高於或低於特定門檻的值。查看平均值、中位數或特定百分位數,可能會讓人注意到偏離模式的情況。
泡泡
圖表中每個泡泡都代表一項查詢,樣式屬性說明如下:
- 大小:以泡泡大小表示點擊次數,可協助您快速瞭解哪些查詢帶來大量流量 — 泡泡越大,表示該查詢產生的流量越多。
- 顏色:以泡泡顏色表示裝置類別,可協助您瞭解行動裝置和電腦兩者間的搜尋成效差異。您可以採用任何維度當做顏色,不過,值的數量越多,會越難識別出模式。
分析資料
這種視覺化呈現方式的目標是要突顯查詢最佳化商機。圖表會顯示查詢成效,其中 Y 軸代表平均排序,X 軸代表點閱率,泡泡大小代表總點擊次數,而泡泡顏色代表裝置類別。
紅色平均參考虛線代表各個軸的平均值,這兩條線將圖表分割成不同象限,代表四種類型的查詢成效。您的象限可能會與本文中分享的圖表看起來不太一樣,因為畫面是由您的網站查詢分布情形決定的。
圖表顯示四個查詢群組,您可以分析這些內容,決定要在改善 Google 搜尋成效時投注心力在哪些方面。
- 高排名,高點閱率:這部分沒什麼需要進行處理,效果已經很好。
- 低排名,高點閱率:這些查詢似乎很貼近使用者需求,雖然排名低於網站上的平均查詢排名,點閱率卻很高。如果查詢的平均排序升高,就有可能大幅影響您的成效。針對這類查詢,建議您著重於改善 SEO。舉例來說,園藝網站第 2 象限的熱門查詢可能是「如何製作木屋」。請檢查您是否已經有關於這個主題的網頁,然後透過以下兩種方式繼續操作:
- 如果沒有相關網頁,建議建立一個網頁,集中收錄您網站上有關該主題的所有資訊。
- 如果已有相關網頁,建議增加內容,進一步滿足使用者的需求。
- 低排名,低點閱率:查看低點閱率 (包含低排名與高排名) 的查詢時,最有趣的部分是查看泡泡大小,瞭解哪些查詢雖然點閱率低,但仍帶來大量流量。雖然在此象限的查詢似乎不值得您投資改善,但這些查詢主要可分為兩組:
- 「相關查詢」:如果此處的查詢對您來說很重要,這是個好的開始,因為這表示這些查詢已經顯示在 Google 搜尋中了。將這些查詢排在完全不會顯示在搜尋結果中的查詢之前,因為這類查詢比較容易最佳化。
- 「不相關查詢」:如果您的網站內容與這些查詢無關,也許這正是個好機會,可以微調您的內容,鎖定能帶來相關流量的查詢。
- 高排名,低點閱率:這些查詢的點閱率偏低可能有各種不同理由。查看最大的泡泡,找出下列跡象:
- 您的競爭對手可能使用了結構化資料標記,而且搭配複合式搜尋結果顯示,這有可能吸引使用者點擊他們的結果,而非點擊您的結果。建議您針對 Google 搜尋中最常見的視覺元素進行最佳化。
- 您已最佳化的查詢,或「意外」獲得排名的查詢,可能是與您的網站相關但使用者不感興趣的查詢。這對您來說可能不是問題,在這種情況下,您可以忽略這些查詢。如果不想讓使用者透過這些查詢找到您 (例如含有令人反感的字詞),可以試著調整內容,針對這些流量的相關字詞,移除某些可能被視為同義詞或相關查詢的說法。
- 使用者可能已經找到所需資訊,例如公司的營業時間、地址或電話號碼。請檢查所使用的查詢和包含相關資訊的網址。如果您網站的目標之一是吸引使用者前往商店,這種情況就沒什麼問題;如果您認為使用者有必要造訪網站瞭解更多資訊,可以試試改善標題和說明,清楚傳達這項訊息。詳情請參閱下一節。
我們之所以未提及裝置類別,是因為裝置類別可當成查詢成效的額外線索。舉例來說,假設使用者在街上使用導航尋找某個地點,這時有某些查詢會比較貼近使用者需求;在此情況下,該查詢可能在行動裝置上展現高成效,而在電腦上展現低成效。
改善特定查詢的 SEO
找到值得投入時間和心力的查詢後,請務必建立相關網頁,或對相關網頁進行最佳化。
透過本頁面顯示的視覺化圖表找到查詢後,您可以透過 Search Console 使用者介面建立特定查詢的查詢篩選器,或是使用 Looker Studio 建立資料透視表。透過這兩種方法,都可以查看由特定查詢帶來流量的所有網頁。確認要最佳化的查詢及其相關網址後,請參考搜尋引擎最佳化 (SEO) 入門指南,瞭解如何最佳化內容。以下提供幾個訣竅:
- 確認標題元素、說明中繼標記和 alt 屬性皆具體、準確且具有描述性。
- 使用標題元素強調重要文字,並為內容建立階層式結構,方便使用者和搜尋引擎瀏覽您的文件。
- 思考使用者為了尋找您的某些內容而可能搜尋的其他字詞,例如同義詞和相關查詢。您可以運用 Google Ads 提供的關鍵字規劃工具,藉此找出新的關鍵字變化版本,並查看每個關鍵字的大致搜尋量。您也可以利用 Google 搜尋趨勢,從與您網站相關的新熱門話題與查詢中尋找靈感。
除非另有註明,否則本頁面中的內容是採用創用 CC 姓名標示 4.0 授權,程式碼範例則為阿帕契 2.0 授權。詳情請參閱《Google Developers 網站政策》。Java 是 Oracle 和/或其關聯企業的註冊商標。
上次更新時間:2025-08-04 (世界標準時間)。
[null,null,["上次更新時間:2025-08-04 (世界標準時間)。"],[[["\u003cp\u003eVisualize Search Console data with a bubble chart to reveal query optimization opportunities by analyzing click-through rate, average position, and clicks.\u003c/p\u003e\n"],["\u003cp\u003eIdentify high-potential queries by focusing on those with low positions but high click-through rates, indicating relevance and potential for increased traffic with improved ranking.\u003c/p\u003e\n"],["\u003cp\u003eInvestigate queries with low click-through rates and high positions to determine if they are unrelated to your site or require optimization for better user engagement.\u003c/p\u003e\n"],["\u003cp\u003eOptimize content for identified queries by refining titles, meta descriptions, headings, and incorporating relevant keywords to enhance search visibility and user experience.\u003c/p\u003e\n"]]],["A bubble chart visualizes query performance using Search Console data. It displays click-through rate (CTR) on the x-axis, average position on the y-axis, click volume as bubble size, and device category as bubble color. Customization options include data control, date range, query, country, and device filters. Analyzing the quadrants formed by average CTR and position helps identify optimization opportunities. Focus on queries with low position but high CTR, and address top-position, low-CTR queries by improving content or structured data.\n"],null,["# How to Create a Search Console Bubble Chart | Google Search Central\n\nImproving SEO with a Search Console bubble chart\n================================================\n\n\nAnalyzing Search performance data is always a challenge, but even more so when you have plenty of\nlong-tail queries, which are harder to visualize and understand. A\n[bubble chart](https://support.google.com/datastudio/answer/7207785)\ncan help you understand which queries are performing well for your site, and which could be improved.\n\n\nIf you'd like to test the techniques discussed here, you can\n[connect your data to Looker Studio](https://datastudio.google.com/reporting/1e5b5f6a-38d7-4547-a54b-69594681a09b/page/xFbeC/preview) and play with the chart settings. \n\n\nIf you haven't read [connecting Search Console to Looker Studio](/search/blog/2022/03/connecting-data-studio) and\n[monitoring Search traffic with Looker Studio](/search/blog/2022/03/monitoring-dashboard),\nconsider checking them out to understand more about what you can do with Search Console in Looker\nStudio.\n\nUnderstanding the chart\n-----------------------\n\n\nA bubble chart is a great visualization when you have multiple metrics and dimensions because it\nlets you to see relationships and patterns in your\ndata more effectively. In the example shown here, you can see\n[click-through rate](https://support.google.com/webmasters/answer/7576553#choosingmetrics)\n(CTR), average position, and clicks for the query and device dimensions in one view.\n\n\nThis section goes into detail on some of the chart elements to clarify what the chart shows, and what it doesn't.\n\n### Data source\n\n\nThe bubble chart shown in this page uses the Site Impression table available through the\n[Search Console data source](https://support.google.com/datastudio/answer/7314895),\nwhich includes [Search performance data](https://support.google.com/webmasters/answer/7576553)\naggregated by site and queries.\n\n### Filters and data controls\n\n\nThere are five customization options in the chart to help you control your data effectively:\n\n1. **[Data control](https://support.google.com/datastudio/answer/7415591)**: Choose the Search Console property you'd like to analyze.\n2. **Date range**: Choose the date range you'd like to see in the report; by default you'll see the last 28 days.\n3. **Query** : Include or exclude queries to focus on. You can [filter your data](/search/blog/2021/06/regex-negative-match) similar to how you do it in Search Console.\n4. **Country**: Include or exclude countries.\n5. **Device**: Include or exclude device categories.\n\n### Axes\n\n\nThe axes in the chart are average position (y-axis) and site CTR (x-axis). There are three\nsignificant transformations in the axes:\n\n6. **Reverse y-axis direction**: Since the y-axis shows average position, inverting it means that 1 is at the top. For most charts, the best position is in the top right corner, so it is more intuitive to invert the y-axis when using it to display average position.\n7. **Log scale** : Using a [logarithmic scale](https://en.wikipedia.org/wiki/Logarithmic_scale) for both axes lets you to have a better understanding of queries that are in the extremities of the chart (very low CTR, average position, or both).\n8. **[Reference lines](https://support.google.com/datastudio/answer/9921462)**: The reference line is very helpful to highlight values that are above or below a certain threshold. Looking at the average, median, or a certain percentile can call attention to deviations from the pattern.\n\n### Bubbles\n\n\nEach bubble in the chart represents a single query, with the following\n[style properties](https://support.google.com/datastudio/answer/7207785#style-properties):\n\n- **Size**: Using the number of clicks as the bubble size helps you see in a glance which queries are driving the bulk of the traffic---the larger the bubble the more traffic the query generates.\n- **Color**: Using the device category as the bubble color helps you understand the differences between mobile and desktop Search performance. You can use any dimension as the color, but as the number of values increases, the harder it is to recognize patterns.\n\nAnalyzing the data\n------------------\n\n\nThe goal of this visualization is to help surface query optimization opportunities. The chart\nshows query performance, where the y-axis represents average position, the x-axis represents CTR\nthe bubble size represents total number of clicks, and the bubble color represents\ndevice category.\n\n\nThe red Average dashed reference lines show the average for each of the axes, which split the chart\ninto quadrants, showing four types of query performance. Your quadrants are likely to look\ndifferent than the one shared in this post; they'll depend on how your site queries\nare distributed.\n\n\nThe chart shows four groups of queries, which you can analyze to help you decide where to invest\nyour time when optimizing your Google Search performance.\n\n1. **Top position, high CTR**: There's not much you need to do for those; you're doing a great job already.\n2. **Low position, high CTR** : Those queries seem relevant to users; the queries get a high CTR even when ranking lower than the average query on your website. If the query average position moves up, it could have a significant impact on your performance---focus on improving SEO for these queries. For example, a top query in quadrant 2 for a gardening website could be \"how to build a wooden shed.\" Check if you have a page about this already, and proceed in two ways:\n - If you don't have a page, consider creating one to centralize all the info you have in the website about the subject.\n - If you do have a page, consider adding content to better address that user need.\n3. **Low position, low CTR** : When looking at queries with low CTR (both with low and top position), it's especially interesting to look at the bubble sizes to understand which queries have a low CTR but are still driving significant traffic. While the queries in this quadrant might seem unworthy of your effort, they can be divided into two main groups:\n - **Related queries**: If the query in question is important to you, it's a good start to have it appearing in Search already. Prioritize these queries over queries that are not appearing in Search results at all, as they'll be easier to optimize.\n - **Unrelated queries**: If your site doesn't cover content related to this query, maybe it's a good opportunity to fine tune your content or focus on queries that will bring relevant traffic.\n4. **Top position, low CTR** : Those queries might have a low CTR for various reasons. Check the largest bubbles to find signs of the following:\n - Your competitors may have [structured data markup](/search/docs/appearance/search-result-features) and are showing up with rich results, which might attract users to click their results instead of yours. Consider optimizing for the most common [visual elements in Google Search](/search/docs/appearance/visual-elements-gallery).\n - You may have optimized, or be \"accidentally\" ranking, for a query that users are not interested in relation to your site. This might not be an issue for you, in which case you can ignore those queries. If you prefer people not to find you through those queries (for example, they contain offensive words), try to fine-tune your content to remove mentions that could be seen as synonyms or related queries to the one bringing traffic.\n - People may have already found the information they needed, for example your company's opening hours, address, or phone number. Check the queries that were used and the URLs that contained the information. If one of your website goals is to drive people to your stores, this is working as intended; if you believe that people should visit your website for extra information, you could try to optimize your titles and description to make that clear. See next section for more details.\n\n\nWe haven't mentioned the device categories because they can be used as additional signs of query\nperformance. For example, suppose some queries are more relevant when people are navigating in the\nstreet, trying to find a location; in that case, the query might have a high performance\non mobile devices, but a low performance on desktop.\n\nImproving SEO for specific queries\n----------------------------------\n\n\nOnce you find queries that are worth the time and effort, make sure to optimize or create pages\nrelated to those queries.\n\n\nAfter you find the queries using the visualization shown in this page, you can\n[create a query filter](https://support.google.com/webmasters/answer/7576553#filteringdata)\nfor specific queries using the Search Console user interface, or create a\n[pivot table](https://support.google.com/looker-studio/answer/7516660)\nusing Looker Studio; in both ways, you can check all the pages that are receiving\ntraffic for a specific query. After you know the queries you want to optimize and their related\nURLs, use the [SEO starter guide](/search/docs/fundamentals/seo-starter-guide)\nto optimize your content. Here are some tips:\n\n- Ensure that your [title](/search/docs/appearance/title-link#page-titles) elements, [description meta tags](/search/docs/appearance/snippet#meta-descriptions), and alt attributes are descriptive, specific, and accurate.\n- Use heading elements to emphasize important text and help create a hierarchical structure for your content, making it easier for users and search engines to navigate through your document.\n- Think about other words that a user might search for to find a piece of your content, for example, synonyms and related queries. You can use the [Keyword Planner](https://ads.google.com/home/tools/keyword-planner/) provided by Google Ads to help you discover new keyword variations and see the approximate search volume for each keyword. You can also [use Google Trends](/search/docs/monitor-debug/google-trends) to find ideas from rising topics and queries related to your website."]]