Ranking third does not mean what it did a decade ago. Above the third organic result there may now be a generated answer, four ads, a question block, and a video carousel — each one an opportunity for the searcher to finish their task before your listing is ever seen. This SERP feature opportunity calculator models that pressure: it starts from a normal click-through rate for your position and reduces it feature by feature, so you can see the click-through rate you should realistically expect rather than the one a generic curve promises.
Arb Digital's SEO team uses this before setting any CTR expectation, because the most common cause of a "failing" campaign is a forecast built on a results page that no longer exists. The rankings improved exactly as planned. The page above them changed shape.
What This SERP Feature Opportunity Calculator Does
Enter search volume, your organic position, and a baseline position-1 click-through rate. Tick the features that appear on the live results page for the query. The calculator applies each feature as a proportional reduction, returns the realistic click-through rate that remains, and converts it into expected monthly clicks.
Add your actual clicks and impressions from Search Console and it does something more useful still: it compares your real click-through rate against the modelled expectation. If you are beating the model, your title and description are doing genuine work and the opportunity is to replicate that elsewhere. If you are well below it, the listing itself is the problem, not the results page.
The weights shown next to each feature are modelled defaults, not measurements from any published study. They reflect the relative prominence and answer-completeness of each feature type, and the sensitivity control scales all of them together so you can tune the model to your own audience rather than accept someone else's assumptions.
How to Use It
- Search the query yourself, in an incognito window, on the device type most of your audience uses. Mobile results carry more features above the fold than desktop, which changes the answer.
- Tick every feature you can see above or beside the organic list. Only tick the featured snippet box if a competitor owns it — if you own it, it works in your favour, not against you.
- Set your baseline CTR. Use your own measured position-1 click-through rate from Search Console rather than the default, which is only a placeholder.
- Adjust sensitivity if needed. Specialist professional audiences tend to scroll past features that stop casual searchers, so a lower sensitivity often fits business-to-business queries.
- Add your real clicks and impressions to see whether your listing is outperforming or underperforming the model.
The Formula: How Feature Loss Is Modelled
Baseline click-through rate follows a decay curve by position: CTR = Position-1 CTR ÷ Position. Features then apply multiplicatively rather than additively:
Realistic CTR = Baseline CTR × (1 − w₁s) × (1 − w₂s) × … where w is each feature's weight and s is your sensitivity setting.
Take position 3 with a 28% position-1 baseline: the starting point is 28 ÷ 3 = 9.33%. With an AI overview (0.32), an ads block (0.18), and a people-also-ask block (0.08) at sensitivity 1, the remaining share is 0.68 × 0.82 × 0.92 = 0.513. Realistic CTR is 9.33% × 0.513 = 4.79%. Against 12,000 monthly searches that is 575 expected clicks instead of 1,120 — the features cost roughly 545 clicks a month at an unchanged ranking.
Why Features Multiply Instead of Adding Up
Adding percentages is the intuitive approach and it breaks immediately. Four features at 30% each would remove 120% of your clicks, which is nonsense. Multiplying models what actually happens: each feature takes a share of the audience that is still available after the features above it have taken theirs.
This has a practical consequence worth internalising. The first feature on a results page does the most damage; the fifth does very little, because it is competing for the attention of a much smaller remaining group. A page that already carries an AI overview and an ads block is not made dramatically worse by an image pack. Conversely, the arrival of a single prominent feature on a previously clean results page can halve your clicks overnight with no ranking change at all.
The Zero-Click Question, Answered Honestly
Plenty of searches end without a click to any website, and that share differs enormously by query type. Anyone quoting a single universal percentage is generalising across queries where it cannot hold: a search for a definition, a conversion, or a phone number is often answered on the results page, while a search for a comparison, a purchase, or a detailed how-to usually is not.
The important thing is what you do with it. Queries whose answer fits in two lines were never going to send much traffic, whatever your position, and building content strategy around them wastes effort. Queries requiring depth, judgement, or a transaction still send clicks, and those are where ranking work pays. Sorting your keyword list by how completely the results page can answer the query is a more useful filter than sorting by volume, and running the survivors through the SEO traffic forecast calculator gives a far more honest projection.
Owning the Feature Instead of Fighting It
Every feature on the page is also a placement you could occupy. Featured snippets are drawn from ordinary indexed pages, and Google's documentation on featured snippets explains that they are selected automatically rather than requested — the practical lever is a page that answers the specific question cleanly and early, in a format that matches the snippet type shown.
The same logic applies to other placements. Video carousels reward having a video that answers the query. Image packs reward properly described, appropriately sized images. Rich results for reviews, recipes, products, and FAQs depend on valid structured data, and the eligibility requirements for each are set out in Google's search gallery documentation. Marking up a page correctly with the schema markup generator or the FAQ schema generator costs almost nothing and occasionally converts a feature from a competitor into an asset.
When a Feature Is Good News
Two cases invert the usual reading. If you own the featured snippet, the block that suppresses everyone else's click-through is amplifying yours — leave that box unticked and treat the position as stronger than the number suggests. And when a local map pack appears, a business with a well-maintained profile may collect calls and direction requests that never register as website clicks at all, meaning the true return is invisible to a CTR model.
People-also-ask blocks can also work in your favour when your own pages populate several of the expanded answers. The block still costs some direct clicks, but the visibility it provides across multiple related questions can be worth more than the clicks it absorbs. As always, the results page itself is the evidence — look at it before deciding whether a feature is a threat.
Comparing the Model Against Your Real Numbers
The gap between modelled and actual click-through rate is where the opportunity sits. Running well below the model with normal impressions usually means the listing is failing to earn attention it is already receiving: a truncated title, a description that does not match the query, or a URL that looks irrelevant. Those are cheap fixes with fast feedback — draft them against the SERP snippet preview and check length limits with the meta description length checker before publishing.
Running above the model consistently is worth investigating just as carefully. Whatever those listings are doing — a distinctive title pattern, a date, a number, a specific promise — is a repeatable advantage, and copying it across your weaker pages is one of the cheapest gains available in search.
Arb Digital's SEO team targets snippet, video, and rich-result placements deliberately, and rewrites the listings that are already earning impressions but not clicks.
SEO Services Content MarketingCommon Mistakes to Avoid
- Checking the results page from your own logged-in browser — personalisation and location change what you see.
- Ignoring the mobile results page, which usually pushes organic results much further down than desktop does.
- Treating a CTR drop as a ranking problem when impressions held steady and only the page around you changed.
- Adding feature penalties together instead of compounding them, which produces impossible results.
- Abandoning a keyword because features exist without checking whether you could own one of them.
Related Free Tools From Arb Digital
Project the traffic that remains with the SEO traffic forecast calculator, write listings that beat the model using the SERP snippet preview and the meta tag generator, and chase rich results with the schema markup generator. If a page lost clicks without losing rankings, the content decay calculator will tell you whether the trend is worth acting on. More in the free online tools hub.
Frequently Asked Questions
It is any element on a search results page that is not a standard organic listing — generated answers, ads, featured snippets, question blocks, video carousels, image packs, map packs, and product carousels all qualify. Each occupies space and attention above or beside the organic results.
They are modelled defaults chosen to reflect how prominent each feature is and how completely it answers a query. They are not measurements from a published study, which is why the sensitivity control exists — adjust them to match what you observe in your own data.
Because each feature only competes for the attention that remains after the ones above it. Adding percentages would eventually remove more than one hundred per cent of clicks, which is impossible, while compounding reflects diminishing effects.
No. If you own the featured snippet, hold several answers in a question block, or appear in a map pack with a strong profile, the same features can increase your visibility. Only count features held by competitors as a reduction.
Compare clicks against impressions in Search Console. Steady impressions with falling clicks points to a change on the results page. Falling impressions points to a change in your ranking or in search demand.
You cannot request one, but you can improve the odds by answering the specific question concisely and early on the page, using a format that matches the snippet type shown, and making sure the page already ranks on the first page for that query.
Not automatically. Queries needing comparison, depth, or a transaction still generate clicks. Queries with a short factual answer were always low-yield. Judge each query by how completely the results page can satisfy it rather than by the presence of a feature alone.
Outputs here are modelled planning estimates, not measured click-through rates — always validate them against your own Search Console data before making budget decisions.
