2-variable framework for content refresh
The more pages you have on your site, the more demanding you need to prioritize which ones should be improved first.
You rarely publish all of the website pages in 1 day. Also, rarely are all of your pages a high quality.
Most websites with written content (without programmatic SEO) develop evolutionarily — they improve the content gradually with each new unit.
Usually after 1-2 years of publishing content:
The website already has enough data on what type of content and which specific pages bring conversions.
The website already has enough pages that do not meet the newly established quality standards, and as a rule, are poorly ranked in organic search.
Updating the old content became a better ROI activity than creating a new one.
So, you have to choose which pages to update and in which order.
I use 2 variables to prioritize landing pages: conversion potential and content gap.
1/ Conversion potential
I use my own Looker Studio SEO forecasting template to merge GSC and GA4 data to calculate how many additional conversions the specific page will get if it were to rank #1 by all keywords it ranks now.
If the website is big and has many types of different pages, I create custom dimensions with a list of page segments and a separate table to get the same answer on the segment level.
Here are my formulas I use to calculate the clicks, conversions, and revenue potential.
Lost Сlicks = (Impressions * 0.2) — URL Clicks
Lost Key Events = ((Impressions * 0.2) — URL Clicks) * Session key event rate
Lost Revenue = ((Impressions * 0.2 — URL Clicks) * Session key event rate * (Event value/Key events))
The value 0.2 is a pessimistic assumption for the CTR of the #1 position.
These are very valuable formulas because they focus attention on 2 things at once:
How well a particular page converts overall
Is there still organic traffic potential on this page?
You should remember that this SEO forecasting approach is not 100% accurate. There are too many things we don’t count:
impact of brand keywords;
broad keywords, by which landing pages rank, but never will get the #1 position;
lower CTR for many search queries where AI overviews are added.
However, it’s still the best approach because real conversion data is involved.
2/ Content gap
It’s not enough to KNOW which page can bring you more conversions.
It’s important to have an option for content improvement.
So, on this stage, I visit the page and review manually how much new value I can add to this page.
Sometimes a page has a great quality and you can’t improve it anymore. In this case, you may need more internal and external backlinks to this page or SEO experiments on the product level.
The most popular ways to add more value to the page on the content level are:
1/ Give the answers as fast as possible (whether by changing the order of paragraphs or creating a summary of the article).
2/ Add internal and external outbound links to relevant sources.
3/ Add more structure (h2-h6, tables, bullet points).
Note that you don’t have to create new content. You just need to organize better the existing one.
4/ Add screenshots and videos (to help a reader find the solution faster).
5/ Add more real-life experience (quotes, screenshots).
6/ Add data studies if they exist on this topic.
7/ Delete outdated information.
8/ Delete information that doesn’t have additional value.
For example, long paragraphs about “What is X” when the user looks for “How to use X”.
I don't try to add words to make the article longer. I believe that the best articles are the ones that say more with fewer words.
So, finally, you always have to start content refresh with a page that:
has a good conversion rate;
has a lot of impressions, but few clicks;
has a lot of opportunities for improvement.
How do you prioritize pages for content refresh?
P.S. Don’t miss my free Looker Studio templates:
GA4 Backlink Analysis (with monitoring traffic from AI chats).