A website carbon check answers a narrow question: roughly how much CO₂e is linked to one visit to one page. That's the right unit for comparing pages and tracking changes, but it isn't what people usually ask for. A director writing an annual sustainability update, or a client filling in a supplier questionnaire, wants one number for the whole website over a year. You can get a fair estimate in under an hour with the analytics you already have, as long as you are clear about how it was worked out.
The method is simple: grams per page view, multiplied by how many times each page was viewed, added up across the site. The care goes into choosing which pages to check and into saying plainly what the total does not cover.
Step 1: get a year of page views
Use a full twelve months rather than one month multiplied by twelve. Most small business sites are seasonal: a garden centre's spring, a tax adviser's January. A whole year smooths that out.
In Google Analytics 4 the figure is in the Pages and screens report: choose Reports, then Engagement > Pages and screens, and set the date range to the last twelve months. The column you want is Views, which Google defines as the number of web pages your users saw, with repeated views of the same page counted. That matches the unit of a carbon check: one load of one page. Export the table (page path and views) to a spreadsheet. Other analytics tools have an equivalent “pageviews by page” report; the method is the same.
Two cautions. If you only collect analytics from visitors who accept cookies, the views are an undercount, so say so. And if your analytics were added part-way through the year, use the period you actually have and state it.
Step 2: check one page per template
You don't need to check every URL. Most sites are built from a handful of templates: the home page, service or product pages, blog posts, a contact page. Pages built from the same template usually weigh about the same, with the photos and embeds making the difference. Sort your exported views from highest to lowest and check:
- every page in the top ten by views, because they carry most of the total;
- one typical page from each template that isn't already covered;
- any page you know is unusually heavy, such as a gallery or a page with a video.
Run a GreenTracer check on each and note the grams of CO₂e per page view from the report. Each check measures a first, uncached load and uses the heavier of the mobile and desktop runs, so the figures are like for like. Check them on the same day if you can; results can shift a little between runs for the reasons in why two checks of the same page can differ.
Step 3: multiply, then add the rest
For each checked page, multiply its grams per view by its yearly views. That gives the checked pages' share of the total. For everything else (the long tail of pages you didn't check), work out the average grams per view across the pages you did check, weighted by their views, and multiply it by the remaining views. Then add the two parts and divide by 1,000 to get kilograms.
If your checked pages are lighter or heavier than the long tail is likely to be (for example, your blog posts all carry large photos but you only checked one), use the figure for the matching template instead of the average. Either way, write down which you used.
A worked example
An illustrative small business site, with twelve months of views from analytics. The grams per view use GreenTracer's Sustainable Web Design Model v4 figure of about 0.148 g CO₂e per MB transferred on a standard host:
- Home page: 2.0 MB, about 0.296 g per view, 36,000 views: about 10.7 kg.
- Services page: 1.2 MB, about 0.178 g per view, 14,400 views: about 2.6 kg.
- A typical blog post: 3.5 MB, about 0.518 g per view, 9,600 views: about 5.0 kg.
- Everything else: 24,000 views at the weighted average of the three checked pages (about 0.303 g): about 7.3 kg.
That's about 25.5 kg CO₂e a year for the site. On a host listed as green by the Green Web Foundation, where the model removes the data-centre operational share (about 0.121 g per MB), the same traffic comes to about 20.8 kg. Notice where the weight is: the home page alone is over 40% of the total, so it is the page to improve first. A 1 MB saving there is worth more over a year than the same saving on a page few people visit, which is the point made in does a page nobody visits have a carbon footprint?
What the yearly figure does and doesn't mean
Be clear about three limits when you share the number.
- It treats every view as a first visit. The Sustainable Web Design Model includes a return visitor ratio and a data cache ratio for visitors whose browsers already hold some files. Repeat visitors often download much less, because of the HTTP cache. A check measures an uncached load, so multiplying it by every view leans high for sites with many returning visitors.
- It uses data transfer as a stand-in. The model, as the Green Web Foundation's CO2.js documentation explains, estimates energy from the data sent. It doesn't meter your server, and the model's authors say that its broad system boundaries make it likely to overestimate actual emissions.
- It covers page views only. Downloads such as PDFs, emails, background tasks and your own staff's use of the admin area aren't in it. See why website carbon is an estimate for the wider picture.
None of this makes the figure useless. It is consistent, repeatable and based on published figures, which is what you need to see whether next year's number goes down. Just don't present it as a measurement or compare it with a figure worked out a different way.
Writing the figure down so it stays useful
Keep a short note with the spreadsheet: the date range of the views, which pages you checked and when, the figure used for the long tail, whether the host was listed as green, and the model (SWDM v4). Download the PDF of each page's report so you can show where the grams came from. Next year, repeat the same steps with the same choices. If the total changes, you will be able to tell whether it was page weight, traffic or hosting that moved it; tracking improvements honestly covers how to keep those comparisons fair.
When you describe it, say something like “estimated at about 25 kg CO₂e for the year, based on page weight and page views using the Sustainable Web Design Model v4”. Avoid words such as “neutral” or “offset” that the figure can't support. The quickest way to lower the total is still the heaviest, most visited page: start with the one-hour checklist.
