{"id":3828,"date":"2026-06-19T08:29:10","date_gmt":"2026-06-19T07:29:10","guid":{"rendered":"https:\/\/www.gironi.it\/blog\/principal-component-analysis-pca\/"},"modified":"2026-07-17T09:06:00","modified_gmt":"2026-07-17T08:06:00","slug":"principal-component-analysis-pca","status":"publish","type":"post","link":"https:\/\/www.gironi.it\/blog\/en\/principal-component-analysis-pca\/","title":{"rendered":"PCA (Principal Component Analysis): reduce SEO data complexity without losing insights"},"content":{"rendered":"\n<div style=\"background-color: #f8f9fa;padding: 20px;border-radius: 8px;margin-bottom: 30px;border-left: 4px solid #4a90e2\">\n<h3 style=\"margin-top: 0\">In this article:<\/h3>\n<ul style=\"margin-bottom: 0\">\n<li><a href=\"#what-is\">What PCA is<\/a><\/li>\n<li><a href=\"#foundations\">The mathematical foundations<\/a><\/li>\n<li><a href=\"#applications\">Practical applications<\/a><\/li>\n<li><a href=\"#concrete-use\">PCA in web marketing<\/a><\/li>\n<li><a href=\"#implementation\">Implementing PCA in R<\/a><\/li>\n<li><a href=\"#verification\">Verification and interpretation<\/a><\/li>\n<li><a href=\"#faq\">FAQ<\/a><\/li>\n<\/ul>\n<\/div>\n\n\n\n<p class=\"wp-block-paragraph\">SEO and web marketing analysis almost always presents us with the same problem: we have too many metrics and we don&#8217;t know which ones really matter. Search volume, CPC, competition, CTR, bounce rate, time on page, conversions by source, average position \u2014 the list grows with every new tool we add to our stack.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The problem is not a lack of data: it&#8217;s that these metrics are often correlated with each other (more impressions lead to more clicks, which lead to more conversions, which increase costs). When variables move together, most of the information they contain is redundant. <strong>Principal Component Analysis (PCA)<\/strong> is the tool we have to cut through this redundancy: it reduces a set of many correlated variables to a few uncorrelated <strong>components<\/strong> that capture most of the original information.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In clearer terms: PCA finds the &#8220;directions&#8221; in which our data vary the most, and lets us project everything else onto those few dimensions. It is like looking at a three-dimensional object from its most informative angle instead of from a random viewpoint.<\/p>\n\n\n\n<!--more-->\n\n\n\n<h2 class=\"wp-block-heading\">What PCA Is<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">To grasp the idea of dimensionality reduction, picture a city with many interconnected streets. PCA works like an urban-planning system that identifies the main traffic arteries. By focusing on these &#8220;main roads&#8221;, we get a clear view of the city&#8217;s structure without having to analyse every single side street.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In the context of web marketing and data analysis, PCA is a powerful tool for several reasons. It is <strong>effective for visualising and exploring high-dimensional datasets<\/strong>, making it easy to <strong>spot trends, patterns or outliers<\/strong>. It is also commonly used in the data pre-processing stage for machine learning, extracting the most informative features from large datasets while preserving relevant information. Another advantage is its ability to *<em>minimise or eliminate <\/em>multicollinearity<em> and <\/em>overfitting***, frequent problems in web marketing datasets with many potentially correlated variables.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"foundations\">The Mathematical Foundations<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">To understand how PCA works we need to familiarise ourselves with a few concepts. Nothing scary \u2014 we take them one at a time.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Variance and covariance.<\/strong> Variance measures how spread out a single variable is around its mean. Covariance measures how two variables move together: positive means they tend to increase or decrease together, negative means they move in opposite directions. PCA looks for directions (components) that maximise variance \u2014 because more variance means more information.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The <strong>covariance matrix<\/strong> summarises the covariances between all pairs of variables. It has the variances on the diagonal and the covariances off the diagonal. It is the starting point of PCA.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Eigenvalues and eigenvectors.<\/strong> This is the mathematical core. The principal components are simply the eigenvectors of the covariance matrix. An eigenvector identifies a direction in the data space; the associated <strong>eigenvalue<\/strong> tells us how much variance that direction captures.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The formula is straightforward:<\/p>\n\n\n\n\\( \\Sigma \\mathbf{v} = \\lambda \\mathbf{v} \\\\ \\)\n\n\n\n<p class=\"wp-block-paragraph\">where \\( \\Sigma \\) is the covariance matrix, \\( \\mathbf{v} \\) is the eigenvector (the direction of the component), and \\( \\lambda \\) is the eigenvalue (the variance along that direction). Finding the principal components means solving this equation for \\( \\Sigma \\).<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"900\" height=\"825\" src=\"https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-rotazione-en-1.png\" class=\"wp-image-4265\" alt=\"PCA: the principal axes capture the direction of maximum variance\" srcset=\"https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-rotazione-en-1.png 900w, https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-rotazione-en-1-300x275.png 300w\" sizes=\"auto, (max-width: 709px) 85vw, (max-width: 909px) 67vw, (max-width: 1362px) 62vw, 840px\" \/><figcaption class=\"wp-element-caption\">PCA: the principal axes capture the direction of maximum variance<\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">The figure makes the idea immediate. The grey dots are two correlated variables: as X increases, Y tends to increase too. The red arrow (PC1) is the direction in which the data vary the most \u2014 it is the principal axis. The green arrow (PC2) is orthogonal and captures the remaining variance. This is the geometric idea of PCA: finding the axes that &#8220;explain&#8221; most of the variability in the data.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Explained variance.<\/strong> This is the metric that tells us how valuable each component is. The proportion of variance explained by a component is:<\/p>\n\n\n\n\\( \\text{variance explained}_k = \\frac{\\lambda_k}{\\sum_{i=1}^{p} \\lambda_i} \\\\ \\)\n\n\n\n<p class=\"wp-block-paragraph\">where \\( \\lambda_k \\) is the eigenvalue of the \\( k \\)-th component and \\( p \\) is the total number of components. The <strong>cumulative explained variance<\/strong> is the sum of the first \\( k \\) proportions: it tells us how much original information we keep by retaining only \\( k \\) components.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">As a side note: criteria like the <strong>Kaiser rule<\/strong> (keep only components with eigenvalue &gt; 1) and the <strong>scree plot<\/strong> (the ordered eigenvalues graph, with the &#8220;elbow&#8221; as the cut-off point) help choose the number of components to retain.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"applications\">Practical Applications<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">PCA is a versatile technique with a wide range of applications. In <strong>image processing<\/strong>, it is used for compression. In <strong>genomics<\/strong>, it helps identify the most critical genes. In <strong>finance<\/strong>, for risk analysis and portfolio optimisation. In <strong>healthcare<\/strong>, for medical image analysis. In <strong>security<\/strong>, for biometric systems. In <strong>climatology<\/strong>, for analysing large environmental datasets.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For <strong>data analysis and marketing<\/strong> specifically, PCA makes it possible to simplify complex datasets, reduce noise, extract the most significant features, and improve the performance of predictive models. Its ability to visualise high-dimensional data in two or three dimensions makes it easier to spot patterns, trends and outliers.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">An important clarification: <strong>PCA does not eliminate some variables while keeping others<\/strong>. Instead, it builds new variables \u2014 the principal components \u2014 obtained as <strong>linear combinations<\/strong> of the original ones. It is not feature selection: it is feature extraction. The difference is subtle but crucial.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"concrete-use\">PCA in Web Marketing<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Let&#8217;s see how PCA applies to concrete problems in our daily work.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Keyword analysis.<\/strong> A keyword dataset has numerous metrics: search volume, competition, CPC, average position on Google and Bing. By applying PCA, we can condense these dimensions into a few components that capture the underlying themes. For example, we might find that one component captures the &#8220;potential value&#8221; of the keyword (high volume + high CPC) and another captures &#8220;competitiveness&#8221; (high competition + low ranking).<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Traffic metric analysis.<\/strong> Sessions, bounce rate, time on page, conversions by source \u2014 PCA can reveal latent variables driving performance. One component might capture user engagement and another the effectiveness of different traffic sources.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>User segmentation.<\/strong> By analysing behavioural data with many variables, PCA identifies natural groupings of users, enabling more defined segments.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Campaign performance analysis.<\/strong> Impressions, clicks, conversions, cost, CTR, CPA \u2014 PCA reveals the key factors that determine campaign success.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"implementation\">Implementing PCA in R<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Let&#8217;s now run PCA on real data, with two examples that mirror the scenarios we just described.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">First, let&#8217;s set up the keyword positioning data:<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>set.seed(123)\nn_keywords &lt;- 100\nkeywords &lt;- paste0(\"keyword_\", 1:n_keywords)\nsearch_volume &lt;- round(runif(n_keywords, min = 100, max = 10000))\ncompetition &lt;- runif(n_keywords, min = 0.1, max = 0.9)\ncpc &lt;- round(rnorm(n_keywords, mean = 2.5, sd = 1), 2)\nranking_google &lt;- round(rnorm(n_keywords, mean = 15, sd = 10), 0)\nranking_bing &lt;- round(rnorm(n_keywords, mean = 12, sd = 8), 0)\n\nkeyword_data &lt;- data.frame(\n  Keyword = keywords,\n  Search_Volume = search_volume,\n  Competition = competition,\n  CPC = cpc,\n  Ranking_Google = ranking_google,\n  Ranking_Bing = ranking_bing\n)\n\nhead(keyword_data)<\/code><\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Result:<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>    Keyword Search_Volume Competition  CPC Ranking_Google Ranking_Bing\n1 keyword_1          2947   0.5799912 1.79             37            6\n2 keyword_2          7904   0.3662588 2.76             28            6\n3 keyword_3          4149   0.4908904 2.25             12            4\n4 keyword_4          8842   0.8635791 2.15             20            4\n5 keyword_5          9411   0.4863219 1.55             11            9\n6 keyword_6           551   0.8122802 2.45             10           15<\/code><\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Now the campaign performance data:<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>set.seed(456)\nn_campaigns &lt;- 50\ncampaign_ids &lt;- paste0(\"campaign_\", 1:n_campaigns)\nimpressions &lt;- round(runif(n_campaigns, min = 1000, max = 100000))\nclicks &lt;- round(impressions * runif(n_campaigns, min = 0.01, max = 0.1))\nconversions &lt;- round(clicks * runif(n_campaigns, min = 0.005, max = 0.05))\ncost &lt;- round(clicks * runif(n_campaigns, min = 0.1, max = 2), 2)\nctr &lt;- round((clicks \/ impressions) * 100, 2)\ncpa &lt;- round(cost \/ conversions, 2)\ncpa[is.nan(cpa)] &lt;- 0\n\ncampaign_data &lt;- data.frame(\n  Campaign_ID = campaign_ids,\n  Impressions = impressions,\n  Clicks = clicks,\n  Conversions = conversions,\n  Cost = cost,\n  CTR = ctr,\n  CPA = cpa\n)\n\nhead(campaign_data)<\/code><\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Result:<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>  Campaign_ID Impressions Clicks Conversions    Cost  CTR    CPA\n1  campaign_1        9866    873          14 1093.32 8.85  78.09\n2  campaign_2       21841   1788          20 3360.17 8.19 168.01\n3  campaign_3       73563   2866          66 2764.48 3.90  41.89\n4  campaign_4       85361   4121          73 1422.12 4.83  19.48\n5  campaign_5       79051   3432         133 1623.28 4.34  12.21\n6  campaign_6       33864   3064         126 6047.70 9.05  48.00<\/code><\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Now run PCA with <code>prcomp()<\/code>. It is essential to <strong>scale the data<\/strong> (<code>scale. = TRUE<\/code>) before applying PCA: otherwise variables with larger scales (thousands of impressions vs fractions of CPC) would dominate the analysis.<\/p>\n\n\n\n<div class=\"wp-block-group has-background\" style=\"background-color:#f5f7f9;margin-top:2.5rem;margin-bottom:2.5rem;padding-top:1.5rem;padding-right:1.5rem;padding-bottom:1.5rem;padding-left:1.5rem\"><div class=\"wp-block-group__inner-container is-layout-flow wp-block-group-is-layout-flow\">\n<p>If one variable is expressed in euros and another in percentages, PCA will be dominated almost entirely by the variable with the larger numerical scale. For this reason, in most cases it is essential to standardise the data before the analysis. It is the most common mistake made by those approaching PCA.<\/p>\n<\/div><\/div>\n\n\n\n<pre class=\"wp-block-code\"><code>pca_keywords &lt;- prcomp(keyword_data[, 2:6], scale. = TRUE)\nsummary(pca_keywords)\n\npca_campaigns &lt;- prcomp(campaign_data[, 2:7], scale. = TRUE)\nsummary(pca_campaigns)<\/code><\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Results:<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># summary(pca_keywords) \u2014 keywords (5 variables, 5 components)\nImportance of components:\n                          PC1    PC2    PC3    PC4    PC5\nStandard deviation     1.1381 1.0298 0.9894 0.9305 0.8941\nProportion of Variance 0.2591 0.2121 0.1958 0.1732 0.1599\nCumulative Proportion  0.2591 0.4712 0.6670 0.8401 1.0000\n\n# summary(pca_campaigns) \u2014 campaigns (6 variables, 6 components)\nImportance of components:\n                          PC1    PC2    PC3     PC4    PC5     PC6\nStandard deviation     1.7837 1.2229 0.9303 0.49392 0.4250 0.18138\nProportion of Variance 0.5303 0.2492 0.1442 0.04066 0.0301 0.00548\nCumulative Proportion  0.5303 0.7795 0.9238 0.96442 0.9945 1.00000<\/code><\/pre>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1425\" height=\"675\" src=\"https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-scree-en.png\" class=\"wp-image-4255\" alt=\"Variance explained by PCA: keywords and campaigns\" srcset=\"https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-scree-en.png 1425w, https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-scree-en-300x142.png 300w, https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-scree-en-1024x485.png 1024w, https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-scree-en-1200x568.png 1200w\" sizes=\"auto, (max-width: 709px) 85vw, (max-width: 909px) 67vw, (max-width: 1362px) 62vw, 840px\" \/><figcaption class=\"wp-element-caption\">Variance explained by PCA: keywords and campaigns<\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">The figure tells the story better than the numbers. For keywords (left) the variance is fairly evenly distributed: the metrics are largely uncorrelated and PCA cannot compress them much without losing information. For campaigns (right), the first two components explain almost 80% of the variance \u2014 the metrics are strongly correlated and two dimensions suffice to describe nearly everything.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"975\" height=\"750\" src=\"https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-scree-standalone-en.png\" class=\"wp-image-4261\" alt=\"Scree plot: variance explained by campaign data\" srcset=\"https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-scree-standalone-en.png 975w, https:\/\/www.gironi.it\/blog\/wp-content\/uploads\/2026\/07\/pca-scree-standalone-en-300x231.png 300w\" sizes=\"auto, (max-width: 709px) 85vw, (max-width: 909px) 67vw, (max-width: 1362px) 62vw, 840px\" \/><figcaption class=\"wp-element-caption\">Scree plot: variance explained by campaign data<\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">The enlarged plot on the campaign data confirms: PC1+PC2 exceed 80% cumulative variance. The dashed line at 80% is a common threshold \u2014 below it lie components that contribute little.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The <strong>loadings<\/strong> (<code>pca_keywords\\( rotation<\/code>) show the correlation between original variables and components, helping interpret the meaning of each component. The <strong>scores<\/strong> (<code>pca_keywords \\)x<\/code>) represent the projection of the original data onto the new space.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For further visualisation, you can use the scree plot (<code>plot(pca_keywords)<\/code>) and the biplot (<code>biplot(pca_keywords)<\/code>), which displays both scores and loadings in the plane of the first two components.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"verification\">Verification and Interpretation<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Interpreting principal components requires domain knowledge. If in the keyword data PCA the first component has high positive loadings for search volume and CPC, it might represent &#8220;high-potential keywords&#8221;. If the second component is dominated by ranking, it might represent &#8220;actual visibility&#8221;. Interpretation is always context-dependent.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Keep in mind that <strong>principal components do not have a &#8220;natural&#8221; meaning<\/strong>. They are mathematical constructs that we must interpret by looking at the loadings of the original variables. There is no predefined label for PC1: we have to build it, based on the data and the context.<\/p>\n\n\n\n<div class=\"wp-block-group has-background\" style=\"background-color:#f5f7f9;margin-top:2.5rem;margin-bottom:2.5rem;padding-top:1.5rem;padding-right:1.5rem;padding-bottom:1.5rem;padding-left:1.5rem\"><div class=\"wp-block-group__inner-container is-layout-flow wp-block-group-is-layout-flow\">\n<p>It is important to keep PCA&#8217;s limitations in mind. It assumes <strong>linear relationships<\/strong> between variables, and it is <strong>sensitive to data scale<\/strong> (which is why we always scale before applying it). For non-linear relationships, techniques like t-SNE and UMAP may be more appropriate.<\/p>\n<\/div><\/div>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"faq\">FAQ<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>When does it make sense to use PCA?<\/strong><br> When we have many correlated variables and want to reduce them to a few interpretable dimensions. It is ideal for exploration, visualisation, and pre-processing for machine learning. It does not make sense when variables are already few and independent.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>How many components should I keep?<\/strong><br> It depends on the cumulative explained variance. A rule of thumb: stop when the scree plot curve flattens (the &#8220;elbow&#8221;), or when cumulative variance reaches 70-80%. The Kaiser rule (eigenvalue &gt; 1) is another criterion, but should be used flexibly.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Why do I need to scale the data before PCA?<\/strong><br> Because PCA maximises variance. If one variable is measured in thousands (impressions) and another in hundredths (CTR), the first would artificially dominate the analysis. Scaling (mean 0, standard deviation 1) puts all variables on an equal footing.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Does PCA work with non-linear data?<\/strong><br> No, standard PCA assumes linear relationships. For non-linear structures there are variants like Kernel PCA, or non-linear techniques such as t-SNE and UMAP.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"conclusion\">Conclusion<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Principal Component Analysis gives us an elegant way to untangle the complexity of our data. It reduces dimensionality, reveals hidden patterns, improves predictive models, and makes visualisable what would otherwise be a jumble of numbers.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Its power, however, also lies in its limitations: it works well when relationships are linear, requires scaling, and the interpretation of the components is our responsibility, not the algorithm&#8217;s. Used with awareness, it is one of the most versatile tools in the web marketing data analysis toolkit.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h3 class=\"wp-block-heading\">Further Reading<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Principal Component Analysis is covered with exemplary clarity in <a href=\"https:\/\/www.amazon.it\/dp\/1461471370?tag=consulenzeinf-21&amp;ascsubtag=una-introduzione-allanalisi-delle-componenti-principali-pca\" rel=\"nofollow sponsored noopener\" target=\"_blank\"><em>An Introduction to Statistical Learning<\/em><\/a> by James, Witten, Hastie and Tibshirani, alongside other unsupervised learning techniques.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>In this article: What PCA is The mathematical foundations Practical applications PCA in web marketing Implementing PCA in R Verification and interpretation FAQ SEO and web marketing analysis almost always presents us with the same problem: we have too many metrics and we don&#8217;t know which ones really matter. Search volume, CPC, competition, CTR, bounce &hellip; <a href=\"https:\/\/www.gironi.it\/blog\/en\/principal-component-analysis-pca\/\" class=\"more-link\">Leggi tutto<span class=\"screen-reader-text\"> &#8220;PCA (Principal Component Analysis): reduce SEO data complexity without losing insights&#8221;<\/span><\/a><\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_uag_custom_page_level_css":"","footnotes":""},"categories":[161],"tags":[],"class_list":["post-3828","post","type-post","status-publish","format-standard","hentry","category-statistics"],"lang":"en","translations":{"en":3828,"it":3349},"uagb_featured_image_src":{"full":false,"thumbnail":false,"medium":false,"medium_large":false,"large":false,"1536x1536":false,"2048x2048":false,"post-thumbnail":false},"uagb_author_info":{"display_name":"Paolo Gironi","author_link":"https:\/\/www.gironi.it\/blog\/author\/autore-articoli\/"},"uagb_comment_info":0,"uagb_excerpt":"In this article: What PCA is The mathematical foundations Practical applications PCA in web marketing Implementing PCA in R Verification and interpretation FAQ SEO and web marketing analysis almost always presents us with the same problem: we have too many metrics and we don&#8217;t know which ones really matter. Search volume, CPC, competition, CTR, bounce&hellip;","_links":{"self":[{"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/posts\/3828","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/comments?post=3828"}],"version-history":[{"count":5,"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/posts\/3828\/revisions"}],"predecessor-version":[{"id":4388,"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/posts\/3828\/revisions\/4388"}],"wp:attachment":[{"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/media?parent=3828"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/categories?post=3828"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.gironi.it\/blog\/wp-json\/wp\/v2\/tags?post=3828"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}