In paid search, site speed rarely looks like a technical problem at first. It usually shows up as rising CPL, weaker landing-page performance, and growing friction after the click.
That is what makes speed expensive: the click is already paid for by the time the landing page fails to load, reassure, or convert fast enough.
This article shows how speed affects CPL through post-click reach, landing-page conversion, and landing-page experience - and how to fix those losses in the right order.
Key takeaways
Key takeaways in 30 seconds
- The cost of a lead increases not because the site is âslow in general,â but because some of the paid clicks do not reach a full view of the page and form.[3]
- In Google Ads, landing page quality is included in the Quality Score diagnostic, and ad and landing page quality can affect CPC, position, and overall ad performance.[1]
- For landing pages, it is not the abstract scores that are critical, but the custom thresholds: LCP up to 2.5 seconds, INP up to 200 ms, and CLS up to 0.1 at the 75th percentile.[5]
- Even with a constant cost per click, speeding up the landing page can reduce the CPL through an increase in conversion: in a Portent study, a site with a loading time of 1 second gave 3 times higher lead generation conversion than a site with a loading time of 5 seconds.[8]
- A slow page is especially dangerous for automated strategies: if the algorithm receives few reliable conversions, it has a harder time optimizing bids based on the real likelihood of the application.[10]
- A practical next step is to test ad landings against real user data, link speed to CPL across campaigns, and fix those pages first where latency is multiplied by the most expensive traffic.[6]
- Where is speed included in the cost of a lead?
- Why a paid click does not mean a visit?
- How delay turns into BYN losses
- How speed relates to Quality Score and auction
- What speed metrics are important for advertising landing
- Why mobile traffic suffers first
- Where to look for a problem in Google Ads, Yandex Direct and analytics reports
- What to fix first
- How to check that the speed has actually reduced the CPL
- Frequently asked questions
- Conclusion
Where is speed included in the cost of a lead?
The cost of a lead in contextual advertising is usually calculated simply: costs are divided by the number of applications. But inside this formula there is a layer that is often not visible in the advertising account: how many paid clicks actually saw the page, understood the offer, waited for the form and were able to submit the application. Google explicitly states that effective landing pages are critical to conversions from Google Ads, and improving the speed of mobile landing pages is one easy way to get better results from mobile advertising.[3]
In practice, speed affects CPL not as a separate technical parameter, but as a funnel multiplier. If a click costs 3 BYN, landing conversion is 4%, and the site consistently opens quickly, one lead costs about 75 BYN. If due to delays, heavy scripts and unstable form the conversion drops to 2%, the same click turns into 150 BYN per lead. The advertising rate may not have changed, but the economy has worsened by half.
It is important to separate the two situations. First: the user didnât even wait for useful content. Second: the user saw the page, but due to slow response, interface shifts or late loading of the form, did not take the target action. Core Web Vitals shares exactly these levels of experience: loading, interactivity and visual stability.[5]
Why a paid click does not mean a visit?
Contextual advertising sells a click, but businesses need contact. There is a short but expensive section between a click and a lead: the browser opens the page, the main content is loaded, styles, scripts, widgets, analytics, forms and event handlers are connected. If the user leaves in this area, the click may remain in the advertising statistics, but not even a normal visit appears in the business funnel.
Google Ads in landing page reports allows you to look at mobile optimization and analyze the pages to which advertising traffic goes.[3]PageSpeed Insights adds two types of data to this: laboratory data, useful for diagnostics, and field data, which reflects real-world user experiences across devices and networks.[6]This is fundamental for advertising: a test on an office laptop does not show how the page opens for a user on the mobile Internet at the moment of an actual click.
The more expensive the click, the more painful this gap is. For SEO, a slow page is also harmful, but in context, every lost download is immediately paid for. Therefore, an advertising landing page should be assessed not only by design and text, but by its ability to quickly bring the user to the first meaningful screen and action.
| Area after click | What does the advertising office see? | What's happening in business |
|---|---|---|
| Click on ad | Consumption fixed | The user has just started the transition |
| The page takes a long time to respond | The click has already been paid | The user can leave before viewing the offer |
| The content appears, but the form loads later | The visit may be included in analytics | The intention cools down, some leads are lost |
| The page opened quickly and immediately gave action | Click and visit agreed | The chance of an application increases at the same click price |
How delay turns into BYN losses
The basic formula for management analysis looks like this: CPL = CPC / landing conversion. But for a slow site, itâs more useful to break down conversion into two factors: the reach to meaningful views and the conversion of those who actually saw the page. Then the formula becomes more accurate: CPL = CPC / (share of reaching users ÃÂ conversion after viewing).
Let's say a campaign receives 1000 clicks for 3 BYN. Consumption - 3000 BYN. If 92% of users waited for the page, and 5% of them left a request, the business receives 46 leads, CPL about 65 BYN. If, due to speed, only 75% reach the page, and the conversion of those who see it drops to 3.5%, you get 26 leads and a CPL of about 115 BYN. The loss of 50 BYN on each lead occurred not in the bid or in the ad, but in the technical area after the click.
This math explains why "lower your cost per click" isn't always the right first response. If a campaign buys relevant demand, but the landing page is slow to display the offer and form, lowering the bid may reduce the volume of traffic, but will not correct the cost per lead. In such a scenario, speed optimization returns economics without reducing coverage.
With the same CPC, the difference between 46 and 26 leads changes the CPL from approximately 65 BYN to 115 BYN. This is a management model that needs to be recalculated using actual campaign, page, and analytics data.[8]
How speed relates to Quality Score and auction
Speed can't honestly be boiled down to "improve PageSpeed and Google will lower CPC." The mechanics are subtler. Google describes the Quality Score as a diagnostic score from 1 to 10 that compares the quality of your ad and landing page to competitors for keywords. Components include expected CTR, ad relevance, and landing page quality.[1]
However, Google specifically states that ad quality predicts how much users will like a search ad and its landing page. High quality ads may receive higher rankings, lower costs, and better performance; The higher the quality, the lower the actual cost per click usually is.[2]That is, speed is important not as a âpoint for the sake of a point,â but as part of the advertising experience: the user receives a relevant answer faster, performs the action more easily, and returns to the search less often.
In Yandex Direct, logic also does not come down to the rate. The impression documentation states that for similar ads, the ad with the best combination of CTR, quality score, and bid per click may appear in the SERP.[9]If the landing page is technically unstable or the site is unavailable, the advertising system can limit impressions by monitoring site availability.[9]
OK
M.I.D.
RISK
What speed metrics are important for advertising landing
An advertising landing page does not require the cult of one PageSpeed score. It is more important for businesses to understand at what stage the user loses the feeling of control: he waits a long time for the first screen, tries to click on the form without reacting, or misses due to the shifting of elements. Core Web Vitals provides a working map: LCP measures loading of the largest visible element, INP measures interactivity, CLS measures visual stability.[5]
PageSpeed Insights categorizes user experience into Good, Needs Improvement, and Poor thresholds, and shows the 75th percentile. This is important: you need to look not only at the average, but at the experience of most users, including more complex devices and networks.[6]For an ad landing, the 75th percentile is closer to the reality of the campaign than a single test after clearing the cache.
A good LCP should be up to 2.5 seconds. For context, this is the moment when a person realizes that the page matches the ad, and is not just looking at a blank background, a preloader, or a late-appearing banner.[5]
A good INP is 200ms or less. If a user clicks on a button, opens a calculator, or submits a form, and the page delays response, some applications are lost after viewing the offer.[5]
A good CLS is 0.1 or less. In advertising, this is especially important for forms, buttons and messengers: if the element moves, the user may click the wrong way or lose confidence in the page.[5]
TTFB and FCP help diagnose what exactly is slowing down the landing: server response, rendering, resources, third-party scripts, or a heavy first screen. Web Almanac 2024 shows that mobile FCP and LCP are still noticeably worse than desktop values, so mobile diagnostics cannot be replaced by desktop ones.[7]
Why mobile traffic suffers first
Contextual advertising often has a high share of mobile clicks, especially in local services, e-commerce and urgent demand. But the mobile user is in worse conditions: weaker device, more unstable network, smaller screen, higher intolerance to latency. Therefore, a page that âopens normallyâ on a work computer may fail the campaignâs mobile economy.
Google Ads Help directly links mobile landing speed to mobile advertising results and gives the example that a 1 second delay in mobile retail can impact mobile conversions by up to 20%.[3]In a separate AMP help, Google lays out the same business logic: faster landings typically lead to more conversions because they reduce bounces and improve the post-click experience.[4]
!
Checking an advertising landing page only from a desktop means assessing a different experience than what a business pays for in mobile clicks. What matters for CPL is not the speed in the office, but the speed of campaign users.[6]
Where to look for a problem in Google Ads, Yandex Direct and analytics reports
Speed only becomes manageable when it is linked to advertising economics. A separate PageSpeed report without campaign data does not show where money is being lost. And vice versa: an advertising report without technical metrics forces you to optimize bids where the landing needs to be corrected.
A minimum set of diagnostics: costs, clicks, CTR, CPC, conversions, CPL, devices, landing URLs, download and form submission events. In Google Ads, you need to look at the landing page report and mobile optimization data.[3]PageSpeed Insights provides field data when available and laboratory correction tips.[6]In Yandex Direct, it is important to take into account impressions, bids, CTR, quality factor and possible restrictions related to site accessibility.[9]
Automatic strategies need special attention. Smart Bidding optimizes bids for conversions or the value of conversions in each auction.[10]If a slow page reduces conversions or skews conversion registrations, the algorithm receives a worse learning signal: It sees spend and clicks, but doesn't receive enough reliable bids to confidently find similar auctions.
| Where to watch | What to look for | What conclusion should I draw? |
|---|---|---|
| Google Ads Landing pages | pages with expensive traffic and poor mobile optimization | priority for technical review[3] |
| PageSpeed Insights | LCP, INP, CLS, field and laboratory discrepancies | understand whether users have a real problem[6] |
| GA4 or other analytics | CTA clicks, form start, successful submission | separate download loss from a weak offer |
| CRM | valid leads and quality of requests | do not optimize speed for junk conversions |
| Yandex Direct | CTR, quality factor, bids, availability stops | check whether the technical condition interferes with displays and efficiency[9] |
What to fix first
Speed optimization often fails due to incorrect work order. The team first compresses all images or argues about the framework, although the main brake may be in the server response, a third-party widget, a heavy first screen, duplicate analytics, or a form that is connected later than the main content. For the advertising economy, priority is determined not by the beauty of a technical report, but by the impact on expensive clicks.
Itâs worth starting with landings, where the most expenses go or where the CPL is higher than normal. Then you need to break down the problem into stages: server, first screen, resources, scripts, interactivity, form, analytics events. This ordering is consistent with the logic of PageSpeed Insights: laboratory data helps find causes, but the final assessment must take into account real user data.[6]
Stage 1
Gather the URLs where the majority of your ad spend is spent and correlate them with CPL, devices, and conversions. Don't start with low-traffic pages, even if they have a poor technical score.
Stage 2
View LCP, INP, CLS and field data availability in PageSpeed Insights. If field data is not available, use the laboratory test as diagnostic, but not as the sole evidence of the problem.[5]
Stage 3
Remove heavy images, late critical CSS, blocking scripts and unnecessary elements before the offer. The goal is for the user to quickly see the match between the request, ad and offer.[5]
Stage 4
The form, phone, messengers and buttons should be accessible without delays, shifts and conflicts of third-party scripts. Otherwise, the page may quickly look loaded, but not convert well.[5]
Stage 5
Check the events: viewing the landing page, clicking on the CTA, starting to fill out the form, successful submission, transfer of the lead to the CRM. Without this, it is impossible to prove that the acceleration reduced the CPL and did not simply improve the technical report.[10]
How to check that the speed has actually reduced the CPL
Website speedup shouldn't be judged on the basis of "it's greener." The proper test is to compare the advertising economics before and after the changes, taking into account seasonality, bids, semantics, budget and lead quality. If you changed your website, offer, campaigns and CRM at the same time, it will be difficult to isolate the impact of speed.
A practical approach: choose 2-3 expensive landing pages, fix the base period, implement a limited set of high-speed changes, do not change the campaign structure unless necessary, and compare user behavior. Key indicators: share of users reaching the useful screen, CTA events, form conversion, CPL, share of mobile leads, lead validity in CRM. For automated strategies, you need to give the system time to accumulate new conversions because Smart Bidding works on conversion probability and value in each auction.[10]
- Record basic CPC, CR, CPL, devices, landing URLs and lead quality.
- Remove PageSpeed Insights for mobile and desktop versions, and separately note field data.
- Make sure that form events and CTAs are correctly transmitted to analytics and advertising accounts.
- Implement changes to priority landings, without mixing them with a complete rework of the offer.
- Compare CPL and conversions for the same campaigns, devices and URLs.
- Check not only the number of leads, but also valid requests in CRM.
Wrong
Optimization for score
The team improves the overall PageSpeed of the homepage, although ad traffic goes to other URLs. The report shows an increase in scores, but the CPL for campaigns remains almost unchanged.
Right
Optimization for leads
The team selects landing pages with the most expensive traffic, speeds up the first screen and form, checks the mobile experience and compares CPL across the same campaigns and devices.[8]
How Ontop solves this
At Ontop, we view ad landing speed as part of a performance system, rather than as a separate technical service. First, we connect advertising costs, landing pages, devices, analytics events and requests in CRM. This allows you to see exactly where speed is impacting CPL: before the page view, on the first screen, in the form, in third-party widgets, or in conversion transfers.
Next, we will prioritize economic edits. If a page gets expensive traffic and has poor mobile performance, it is more important than a technically weak but barely used page. For evaluation we use Core Web Vitals, PageSpeed Insights, data from advertising offices and real funnel events, and not just the overall technical score.[5]
After implementing the changes, we check the result through advertising metrics: CPL, landing conversion, share of mobile applications, quality of requests and stability of data transfer to analytics. This approach is especially important for campaigns with automated strategies, where the quality and completeness of conversion signals directly influence further optimization.[10]
If PPC is generating clicks, but the cost per lead is rising, Ontop can conduct a related audit of advertising, landing pages, speed, analytics and CRM to find technical and marketing reasons for post-click losses.
Frequently asked questions
Does site speed really affect the cost per lead or just user behavior?
Influences the cost per lead through user behavior and advertising economics. If some of the paid clicks do not reach the page view or convert worse due to delays, the CPL increases even if the CPC remains the same. Google Ads also takes landing page quality into account in its Quality Score diagnostics, and ad and landing page quality is linked to CPC and performance.[1]
Is getting 90+ in PageSpeed Insights enough?
No. A green lab score is useful, but it does not prove a reduction in CPL. PageSpeed Insights shows laboratory and field data; field data better reflects real-world user experience, and the 75th percentile helps rank the majority of users, including complex devices and networks.[6]
What speed indicators should you look at first?
For advertising landing, LCP, INP and CLS are usually important. LCP shows how quickly the user saw the main content; INP - how quickly the page responds to an action; CLS - how stable the interface is. Recommended thresholds: LCP up to 2.5 seconds, INP up to 200 ms, CLS up to 0.1.[5]
Can a slow website make auto strategies work worse?
Yes, if the speed reduces the number of conversions or breaks their registration. Smart Bidding optimizes bids for conversions or value in each auction, so it needs reliable signals. When the landing process loses applications or analytics records them incompletely, the algorithm works with lower-quality data.[10]
What is more important for CPL: speed or quality of the offer?
You cannot replace one with the other. A quick page with a weak offer will not make a good landing page. But a strong offer on a slow page also loses money, because the user may not wait for the arguments, price, form or contact. Google Ads specifically emphasizes that the landing page must match the ad and help turn clicks into customers.[4]
Do I need to speed up the entire site or just the advertising landing pages?
Ideally, the entire site should be fast, but to reduce the CPL, the first thing you need to do is work with landing pages that receive paid traffic. If your advertising budget is spent on specific URLs, these should be the priority: they are easier to relate to speed, conversion, CPL and lead quality.[3]
Why can the mobile version spoil the CPL if the desktop works fine?
Mobile users are more likely to find themselves in environments with weak networks and less powerful devices. In Web Almanac 2024, mobile sites are noticeably inferior to desktop sites in terms of the share of good LCP, and Google Ads directly recommends improving the speed of mobile landing pages for better mobile advertising results.[7]
Conclusion
Speed is not just a technical KPI; it is part of the unit economics of paid acquisition.
When delays reduce post-click reach and landing-page trust, CPL rises even if media buying stays flat. That is why performance work belongs inside commercial optimisation, not outside it.
Sources
- Quality Score - Google Ads Help - Confirms that the Quality Score is a diagnostic metric that includes expected CTR, ad relevance, and landing page quality, but is not used during the auction:https://support.google.com/google-ads/answer/6167118?hl=ru
- Ad Quality - Google Ads Help - Validates how ad and landing page quality relates to positioning, cost-per-click, clicks, conversions, and overall ad performance:https://support.google.com/google-ads/answer/156066?hl=ru
- Evaluate the performance of your landing pages - Google Ads Help - Confirms the importance of landing pages for Google Ads conversions, the role of mobile landing speed, and an example of the impact of a 1 second delay on mobile conversions:https://support.google.com/google-ads/answer/7543502/evaluate-the-performance-of-your-landing-pages
- Optimize your ads and landing pages - Google Ads Help - Confirms that the landing page should be relevant to the ad and keywords, quickly provide the expected information and help convert clicks into customers.:https://support.google.com/google-ads/answer/6238826?hl=en-EN
- Web Vitals - web.dev - Confirms Core Web Vitals composition, LCP, INP, CLS thresholds and need for 75th percentile scoring:https://web.dev/articles/vitals
- About PageSpeed Insights - Google for Developers - Validates the difference between lab and field data, PageSpeed Insights thresholds, 75th percentile value, and lab test limits:https://developers.google.com/speed/docs/insights/v5/about
- Performance | 2024 | The Web Almanac by HTTP Archive - Confirms market context on Core Web Vitals and mobile performance, including the percentage of mobile sites with good LCP:https://almanac.httparchive.org/en/2024/performance
- Site Speed is Still Impacting Your Conversion Rate - Portent - Confirms the relationship between loading speed and conversion, including data on the 3x difference in lead generation conversion between pages that load in 1 and 5 seconds:https://portent.com/blog/analytics/research-site-speed-hurting-everyones-revenue.htm
- Displaying ads | Yandex Direct - Confirms that Direct impressions take into account CTR, quality factor and bid, and that site accessibility monitoring can pause impressions:https://yandex.ru/support/direct/ru/troubleshooting/shows
- About Smart Bidding - Google Ads Help - Confirms that Smart Bidding uses Google AI to optimize for conversions or conversion value in each auction:https://support.google.com/google-ads/answer/7065882?hl=en
