{"id":29191,"date":"2026-05-29T14:30:04","date_gmt":"2026-05-29T19:30:04","guid":{"rendered":"https:\/\/www.jasminedirectory.com\/blog\/?p=29191"},"modified":"2026-05-29T14:57:55","modified_gmt":"2026-05-29T19:57:55","slug":"how-site-speed-affects-whether-visitors-stay-and-buy","status":"publish","type":"post","link":"https:\/\/www.jasminedirectory.com\/blog\/how-site-speed-affects-whether-visitors-stay-and-buy\/","title":{"rendered":"How site speed affects whether visitors stay and buy"},"content":{"rendered":"

Site speed is usually filed under technical SEO, and the previous articles in this series have treated it there. This article takes it out of that file and places it where, by its effects, it also belongs: among the things that decide whether a business gets customers.<\/p>\n

The argument is that speed is not only a technical property of a site but a commercial one. How quickly a page loads affects whether a visitor stays long enough to read it, and whether a visitor who stays goes on to act \u2014 and those are not technical outcomes but the outcomes a business exists for. This article sets out what the evidence shows about that relationship, why a slow site costs more than it appears to, and what a small business<\/a> can do.<\/p>\n

A note on sources is in order. Peer-reviewed research is cited by author and year and listed at the end; Google’s own published guidance is cited as a primary source and identified as such; industry<\/a> research that is not peer-reviewed is cited and explicitly labelled, since some of the most-cited figures on this topic come from industry analyses rather than academic studies.<\/p>\n

What this article covers<\/h2>\n

This article covers site speed as a commercial issue \u2014 its effect on whether visitors stay on a site and whether they buy. It is an analytical article: it examines the evidence for that effect, rather than only instructing.<\/p>\n

It treats what site speed actually means, what a visitor experiences during the seconds a page loads, and what the evidence shows about speed’s effect on both staying and buying. It then explains why a slow site costs more than the visitors it visibly loses, treats speed’s smaller role in search ranking, addresses how fast is fast enough, and sets out the common causes of slowness and what a small business can do about them.<\/p>\n

Why speed is a marketing issue, not just a technical one<\/h2>\n

Speed earns a place in a series about marketing<\/a>, and it is worth being explicit about why, because the placement is not obvious.<\/p>\n

A business naturally thinks of marketing as the work of attracting visitors \u2014 the search visibility, the content, the channels that bring a person to the site. Speed seems to belong elsewhere, to the machinery of the site, a matter for whoever built it. But marketing does not end when a visitor arrives; the visitor still has to stay, and read, and act, and a slow site fails the visitor at exactly that point \u2014 after the marketing has done its work and spent its budget to bring the person in.<\/p>\n

Seen this way, speed is the marketing issue that sits between attracting a visitor and converting one. A business can do everything else well \u2014 rank, write, persuade \u2014 and a slow site will quietly discard a share of the visitors all that work delivered, before they have seen any of it. That is a marketing loss, whatever file it is kept in, and it is the reason speed is treated here rather than left to the technical articles alone.<\/p>\n

There is a practical consequence to filing speed correctly. A business that thinks of speed as purely technical tends to treat it as someone else’s concern and to leave it unexamined for years; a business that recognises speed as a marketing issue gives it the same attention it gives the other things that win or lose customers. The filing is not a pedantic point — where a business files a problem largely determines whether the problem ever gets looked at.<\/p>\n

What “site speed” actually means<\/h2>\n

Before the evidence, the term itself needs a little precision, because “site speed” is looser than it sounds.<\/p>\n

At its simplest, site speed is how long a page takes to load \u2014 the time between a visitor requesting the page and the page being there. But “being there” is itself not a single moment. There is the moment the page begins to appear, when the first content shows; there is the moment enough of it has appeared to be read; and there is the moment it is fully loaded and usable, with nothing still shifting or arriving. A visitor’s experience of speed is shaped by all three.<\/p>\n

Search engines<\/a>, for their part, assess speed through a set of specific measures of the loading experience \u2014 how quickly the main content appears, how soon the page can be interacted with, how much it shifts about as it loads \u2014 and treat that experience as part of how a page is judged (Google Search Essentials, 2022). A small business does not need to master the technical detail of these measures. It needs to hold the plain version: site speed is how quickly a page becomes there and usable for the visitor, and both the visitor and the search engine are paying attention to it.<\/p>\n

One distinction within this is worth keeping, because it explains a common confusion. A page can begin to appear quickly and still not be usable for some time afterward — the text is visible, but a button does not yet respond, or the layout is still shifting as late elements arrive. A visitor experiences that gap as slowness even though the page “loaded” fast, which is why speed is better thought of as how soon the page is genuinely ready to be used than as a single moment of arrival.<\/p>\n

Speed on a phone is the speed that counts<\/h2>\n

A distinction matters enough to state on its own: a site does not have one speed, it has two, and they can differ sharply. A site can load quickly on a desktop computer and slowly on a phone, and for most small businesses it is the second of these that decides things.<\/p>\n

The reasons a site is slower on a phone are ordinary ones. A phone is, generally, a less powerful device than a computer, with less capacity to process a heavy page; and a phone is often on a mobile network, which can be slower and less steady than the connection a desktop enjoys. The same page, with the same content, can therefore be a brisk experience on a laptop and a sluggish one on a phone held in a customer’s hand.<\/p>\n

This matters because of where the customers are. Most small businesses now receive a large share, often a majority, of their visitors on phones — and a search engine, as the technical articles noted, judges a site largely in its mobile form. A business that has tested its speed only on the computer it works at has measured the version of its site that fewer of its customers see, and not the version the search engine weighs.<\/p>\n

The practical instruction follows plainly. When a business tests its speed, it should test the mobile result, and when it judges whether the site is fast enough, it should judge by the phone. A site that is fast on a desktop and slow on a phone has not solved its speed problem; it has measured around it.<\/p>\n

What happens in the seconds a page loads<\/h2>\n

To understand why speed matters, it helps to consider the thing itself \u2014 what a visitor is actually doing during the seconds a page takes to load. What they are doing is waiting, and waiting is not neutral.<\/p>\n

A visitor who has clicked a result and is watching a blank or half-formed page is in a small state of friction. Their attention, freely given a moment ago, is now being spent on nothing, and attention spent on nothing is quickly withdrawn. The longer the wait, the more of the visitor’s patience it consumes, and patience, once consumed, takes the visitor with it.<\/p>\n

Research into this is older than the modern web’s anxiety about it. A study of how long web users will tolerate waiting found that the tolerable waiting time for retrieving information is, on average, only about two seconds \u2014 and that beyond a point, users abandon the page rather than continue waiting (Nah, 2004). The same work found something a business can use: visible feedback, a sign that something is happening, lengthens the time a user will tolerate. People will wait a little longer when they can see that the wait is not in vain.<\/p>\n

This finding has a small, practical use for a business whose site cannot instantly be made fast. If visible feedback that something is happening lengthens the patience a visitor extends, then a site that shows the visitor it is working — content appearing progressively rather than a blank screen resolving all at once — buys itself a little of the tolerance a wholly blank wait would spend. It is not a substitute for genuine speed, but it is a reason that how a page loads, and not only how fast, affects whether the visitor stays.<\/p>\n

The implication is that speed is a matter of human attention before it is a matter of technical measurement. A slow page is not merely a slow page; it is a page that asks more patience of the visitor than the visitor has agreed to give, and a share of visitors respond to that request by leaving. The figure below shows how that share grows as the wait lengthens.<\/p>\n

\n\n <\/rect>\n <\/line>\n <\/line>\n <\/line>\n <\/line>\n <\/line>\n <\/line>\n <\/line>\n 0%<\/text>\n 25%<\/text>\n 50%<\/text>\n 75%<\/text>\n 100%<\/text>\n 125%<\/text>\n <\/rect>\n <\/rect>\n <\/rect>\n <\/rect>\n +32%<\/text>\n +90%<\/text>\n +106%<\/text>\n +123%<\/text>\n 1 → 3 seconds<\/text>\n 1 → 5 seconds<\/text>\n 1 → 6 seconds<\/text>\n 1 → 10 seconds<\/text>\n As page load time lengthens from one second →<\/text>\n Source: Google\/SOASTA Research, 2017 — industry research, not peer-reviewed.<\/text>\n<\/svg>
Figure 1.<\/strong> The increase in the probability of a mobile visitor leaving, as a page’s load time lengthens from one second. The relationship is steep: a wait that grows from one second to five roughly doubles the chance the visitor abandons the page. Source:<\/strong> Google\/SOASTA Research, 2017 (industry research, not peer-reviewed).<\/figcaption><\/figure>\n

The evidence: speed and whether visitors stay<\/h2>\n

The clearest evidence on speed and staying comes from a large industry analysis carried out by Google with the performance firm SOASTA, which examined a great many mobile pages and the behaviour of their visitors.<\/p>\n

That analysis found a steep relationship between load time and the likelihood of a visitor leaving. As a page’s load time lengthened from one second to three, the probability of a mobile visitor bouncing rose by about 32 percent; as it lengthened to five seconds, the probability rose by about 90 percent; and by ten seconds it had risen by some 123 percent (Google\/SOASTA Research, 2017; industry research, not peer-reviewed). The same body of work reported, in a figure that has been widely repeated since, that more than half of mobile visitors will abandon a page that takes longer than about three seconds to load.<\/p>\n

These figures should be read with appropriate care. They come from an industry analysis rather than a peer-reviewed study, they describe mobile pages in particular, and they describe probabilities across many sites rather than a guaranteed outcome for any one. What they establish, though, is not in serious doubt and matches the older academic finding on tolerable waiting time: the relationship between a slower page and a lost visitor is real, and it is steep \u2014 each additional second of waiting costs a meaningful share of the visitors who were waiting.<\/p>\n

For a small business the lesson is concrete. A site that loads in two seconds and a site that loads in six are not two versions of an acceptable experience; the slower one is shedding a substantial fraction of its visitors before they have read a word, and doing so silently, since a visitor who leaves during the load leaves no trace a casual look at the site would show.<\/p>\n

The invisibility of this loss is worth dwelling on, because it is what lets the problem persist. A visitor who abandons a page during the load never appears as a complaint, never sends a message, never registers as anything a business would notice — they simply are not there. A business can therefore be losing a steady share of its hard-won visitors to slowness and have no felt sense that anything is wrong, which is precisely why the measurement this article urges is necessary: the loss does not announce itself, and only a deliberate look will find it.<\/p>\n

The evidence: speed and whether visitors buy<\/h2>\n

Losing a visitor before the page loads is the first cost of slowness; the second is what happens to the visitors who do stay. Here the evidence concerns conversion \u2014 whether a visitor goes on to do the thing the page exists for.<\/p>\n

The relationship runs in the same direction as the relationship with bouncing, and for a connected reason. Speed shapes the whole of a visitor’s experience, not only its first moment, and a site that is slow to load is often slow at every subsequent step too \u2014 slow between pages, slow to respond to a click, slow at the point of enquiry or purchase. A visitor moving through such a site meets friction repeatedly, and friction at each step costs a share of the visitors at that step.<\/p>\n

Industry analyses of commercial sites have consistently reported that conversion rates fall as pages slow, and that even improvements of a single second can move conversion measurably; the same Google and SOASTA work found that, as the number of elements on a page grew, the probability of conversion fell sharply (Google\/SOASTA Research, 2017; industry research, not peer-reviewed). The precise figures vary between studies and sectors, and a business should treat any single number with caution \u2014 but the direction is consistent across them, and the direction is what matters: a faster site converts better than a slower one, other things being equal.<\/p>\n

The mechanism is worth holding because it is not mysterious. A visitor’s willingness to continue is a finite resource, spent down by every wait and every point of friction; a fast site spends little of it and leaves the visitor with patience intact at the moment of decision, and a slow site arrives at that same moment having already spent much of the visitor’s goodwill on loading screens.<\/p>\n

It is worth being clear about which visitors this costs a business. The visitors lost to slow conversion are not idle browsers; they are, by definition, the visitors who stayed, read, and were interested enough to move toward acting. They are the most valuable visitors the site had — the ones nearest to becoming customers — and a slow site loses a share of them at the very last steps, after every other part of the marketing has succeeded in producing them. That is the most expensive place a business can lose a visitor.<\/p>\n

Why a slow site costs more than the visitors it loses directly<\/h2>\n

The direct cost of a slow site \u2014 the visitors it loses and the conversions it forgoes \u2014 is the visible cost. There are several further costs that are less visible and that, taken together, make slowness more expensive than it first appears.<\/p>\n

The first is wasted marketing spend. Every visitor a slow site loses during the load was delivered to the site by some marketing effort \u2014 a search ranking earned, an advertisement paid for, content written. When the visitor leaves before the page appears, that effort and that spend are spent on nothing; the slowness does not only lose a visitor, it discards the cost of having attracted them.<\/p>\n

The second is the effect on ranking, treated in the next section. The third is reputational: a visitor who experiences a business’s site as sluggish forms an impression of the business itself, and a slow site can read, fairly or not, as a sign of a business that is careless or behind. The fourth is compounding over time \u2014 a site that is slow today is, through the lost visitors and the weaker ranking, attracting fewer customers tomorrow than it would have. A slow site is not a fixed, one-off cost; it is a small tax levied on every other thing the business does to attract customers.<\/p>\n

The phrase is worth taking literally. A tax is paid on activity regardless of how good the activity is, and slowness behaves the same way: it takes its percentage of the visitors a good ranking earns, the visitors an advertisement pays for, the visitors a piece of content attracts, without regard to how well any of that work was done. Fixing the slowness does not improve the marketing; it stops the marketing from being quietly skimmed. For a small business with little budget to spare, ending a tax on all of its effort is itself a substantial gain.<\/p>\n

Speed and search ranking<\/h2>\n

Speed affects not only the visitors a site receives but, to a degree, whether the site is found at all, because search engines treat the speed and quality of a page’s loading experience as a genuine factor in ranking.<\/p>\n

The effect should be described accurately, without overstatement. Speed is a real ranking factor, but it is not a large one, and it will not lift a poor page above a good one \u2014 a fast page with a weak answer does not outrank a slower page with an excellent answer. What speed does is act as a contributing signal and, in effect, a tie-breaker: between two pages that are otherwise comparable in how well they answer a query, the faster one has an advantage.<\/p>\n

For a small business this means speed should be understood as part of the ranking picture but not mistaken for the whole of it. A business cannot rank a thin or poorly written page by making it fast; it can, however, lose ranking it had earned by leaving a genuinely good page slow. Speed will not substitute for the on-page and off-page work, and a slow site quietly undermines that work \u2014 which is reason enough to attend to it.<\/p>\n

One misconception is worth correcting directly. A business sometimes hopes that making its site fast will, on its own, lift it up the rankings noticeably — and it will not, because speed is not that kind of factor. The honest expectation is the reverse: speed is far more likely to be quietly costing a good page some of the rank it deserves than to be capable of winning rank a page has not earned. A business should fix speed to stop a loss, not in the hope of a gain.<\/p>\n

What speed alone cannot do<\/h2>\n

Because the evidence on speed is striking, it is worth a paragraph of caution, so that a business does not expect of speed more than speed can deliver.<\/p>\n

Speed removes a barrier; it does not create demand. A fast site loses fewer of the visitors it attracts and converts more of the ones who stay — but it does not, by being fast, attract visitors who were not coming, or persuade visitors whom the offer itself does not persuade. A business with a weak offer, an unclear service page, or no visibility will not be rescued by load time; it will simply lose its few visitors slightly less quickly.<\/p>\n

This places speed correctly among a business’s priorities. Speed is necessary — a slow site undermines everything else, as this article has argued — but it is not sufficient, and it sits alongside the on-page work, the off-page work, and the quality of the offer rather than above them. A business whose pages do not answer their questions well should not make those pages fast and consider the job done.<\/p>\n

The right way to hold it is this: speed is a barrier to clear, not an engine to drive growth. Clearing it lets the business’s other work reach the visitor unobstructed, which is genuinely valuable — but the other work still has to be good. Speed makes a good site perform to its potential; it cannot make a weak site good.<\/p>\n

How fast is fast enough?<\/h2>\n

If speed matters, a business reasonably asks how fast its site needs to be \u2014 and the answer is best given as a range and a principle rather than a single number.<\/p>\n

The principle is that speed is a matter of diminishing returns. The large gains come from moving a genuinely slow site \u2014 one that takes many seconds to load \u2014 into the range of a few seconds or less. The older research on tolerable waiting time put the threshold of comfortable patience at around two seconds (Nah, 2004), and the steep part of the bounce relationship lies in those first several seconds; a site that loads within two to three seconds has cleared the part of the curve where most of the damage is done.<\/p>\n

Beyond that point, the returns shrink. The difference between a site that loads in two seconds and one that loads in one and a half is real but small, and rarely worth the effort that closing it would take. A small business should therefore aim to be comfortably fast \u2014 within a few seconds, on a phone, on its important pages \u2014 and should not pursue the last fractions of a second on a site that is already quick. The goal is to be off the steep part of the curve, not to win a speed contest.<\/p>\n

This is a genuinely reassuring conclusion, and worth stating as such, because speed is a topic that invites perfectionism. A small business does not need the fastest site in its field, and chasing that goal would take effort better spent on its pages and its offer. It needs a site that is not slow — a site comfortably within a few seconds on a phone — and that is a finite, reachable target rather than an endless pursuit. Past that target, a business should feel free to stop.<\/p>\n

What makes a small business site slow<\/h2>\n

Slowness has causes, and for a small business site the causes are usually a short and familiar list rather than a deep mystery.<\/p>\n

The most common cause by far is oversized images \u2014 photographs uploaded at a far larger size than the page needs, each one a heavy file the visitor’s device must download before the page is complete. Close behind is the page that simply carries too much: too many elements, too many separate pieces of content and decoration, each adding to what must be loaded. The Google and SOASTA analysis noted exactly this, finding the probability of conversion falling sharply as the number of elements on a page rose.<\/p>\n

Other causes lie deeper in how the site is built. A heavy or poorly built theme can make every page slow; an accumulation of add-ons and plugins, each contributing its own weight, can do the same; and slow or overcrowded hosting can leave even a lean site sluggish. The absence of basic performance measures, such as the caching that lets a site avoid rebuilding a page from scratch for every visitor, belongs on the list as well.<\/p>\n

What unites most of these causes is that they are matters of accumulation and neglect rather than of anything deliberate. No business sets out to build a slow site; the slowness arrives gradually, as images are uploaded at whatever size the camera produced, as plugins are added and never removed, as the site grows heavier year by year without anyone watching the weight. This is encouraging, in a way, because a problem that accumulated through inattention can usually be reduced by attention.<\/p>\n

What a small business can do about it<\/h2>\n

The causes above sort, conveniently, into the ones an owner can address and the ones that need a developer \u2014 and the first step, before any fixing, is measurement.<\/p>\n

A business should run a free page-speed tool on its important pages, as the audit article described, because the tool does not only score the page but names the specific things slowing it. That report tells the business which of its speed problems are the simple kind. The most common of those, oversized images, is genuinely an owner-level fix: images can be resized to what the page actually needs and saved in efficient formats before being uploaded, and on many sites this single change is the largest speed gain available.<\/p>\n

Other owner-level steps include removing plugins and add-ons that are not genuinely used, and enabling caching where the site’s system offers it simply. The deeper causes \u2014 a heavy theme, slow hosting, slowness rooted in construction \u2014 are developer or provider work, and a business that has run the speed tool can brief that work accurately, describing the symptom and showing the report. The order is the order the audit article urged: measure first, fix the simple named causes yourself, and hand the rest on with evidence rather than guesswork.<\/p>\n

It is worth setting a realistic expectation of the result. Reducing oversized images and removing unused weight will, on a typical neglected small business site, produce a noticeable improvement — often the difference between a sluggish site and an acceptably quick one. It will not, by itself, produce the fastest site imaginable, and it does not need to. The owner-level fixes are aimed at the largest, cheapest gains; they move a site off the steep part of the curve, and that is the part of the work that matters most.<\/p>\n

Causes of slowness and what to do<\/h2>\n

The table below gathers the common causes of a slow small business site, with what each is and who can usually address it.<\/p>\n

The table is worth reading with one point in mind. The causes are listed separately, but a slow site usually suffers from several of them at once, and the owner-fixable ones — oversized images, unused plugins, missing caching — are precisely the ones a business can address without waiting for anyone. A business need not diagnose which single cause is dominant before acting; clearing the owner-level causes it can see is the right first move regardless, and the speed tool’s report will already have named them.<\/p>\n\n\n\n\n\n\n\n\n\n\n
Cause<\/th>\nWhat it is<\/th>\nUsually addressed by<\/th>\n<\/tr>\n<\/thead>\n
Oversized images<\/td>\nPhotographs far larger than the page needs<\/td>\nOwner — resize and compress before uploading<\/td>\n<\/tr>\n
Too many page elements<\/td>\nMore content and decoration than the page needs to load<\/td>\nOwner, with judgement; sometimes a developer<\/td>\n<\/tr>\n
A heavy theme<\/td>\nA site design<\/a> that is slow by construction<\/td>\nDeveloper<\/td>\n<\/tr>\n
Too many plugins or add-ons<\/td>\nAccumulated extras, each adding weight<\/td>\nOwner — remove what is not genuinely used<\/td>\n<\/tr>\n
Slow hosting<\/td>\nThe service the site runs on is underpowered<\/td>\nOwner’s decision; provider’s responsibility<\/td>\n<\/tr>\n
No caching<\/td>\nThe site rebuilds each page from scratch every time<\/td>\nOwner if the system makes it simple; otherwise a developer<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Concluding remarks<\/h2>\n

Site speed is usually filed as a technical matter, but its effects are commercial: how quickly a page loads helps decide whether a visitor stays to read it and whether a visitor who stays goes on to act. That places speed among the things a marketing-minded business should attend to, not away from them.<\/p>\n

The evidence is consistent in direction even where it varies in detail. The likelihood of a visitor leaving rises steeply as a page slows \u2014 roughly doubling as a wait grows from one second to five \u2014 and conversion falls as a site grows slower, because a visitor’s willingness to continue is a finite resource that every wait spends down. The older research on tolerable waiting time, putting comfortable patience at around two seconds, points the same way.<\/p>\n

A slow site also costs more than the visitors it visibly loses: it wastes the marketing spend that brought them, it modestly weakens ranking, and it quietly taxes every other effort the business makes to attract customers. The remedy does not require perfection \u2014 a site comfortably within a few seconds has cleared the part of the curve where the damage is done \u2014 and it begins with measurement, after which the most common cause, oversized images, is an owner-level fix.<\/p>\n

The next articles in this series move from technical SEO<\/a> to off-page SEO \u2014 the authority a site earns from beyond itself \u2014 beginning with what a backlink is worth and how a small business can earn links without a budget.<\/p>\n

Future developments<\/h2>\n

Speed is likely to grow more important rather than less, for two connected reasons. The first is that visitor expectations only move in one direction: as fast sites become the norm, the patience a visitor extends to a slow one shrinks, and a load time that seemed acceptable a few years ago seems sluggish now.<\/p>\n

The second is that search engines have, over years, given the loading experience a steadily more defined place in how they assess pages, and there is no sign of that reversing. A business investing in speed is investing in something whose value the prevailing direction of both user expectation and search engine policy is raising.<\/p>\n

There is also a quieter trend worth noting. As the tools for measuring and improving speed become better and more automated, the gap between knowing a site is slow and being able to do something about it narrows — much of what once required a developer is becoming an owner-level fix that a good tool will identify and a modern website system will apply. Speed is one of the few technical matters that is, slowly, getting easier for a small business rather than harder.<\/p>\n

The arrival of AI-driven search does not change this. Whatever composes the answer a customer sees, the customer still, at some point, arrives on a page \u2014 and a slow page fails them at that point exactly as it always has. If anything, a visitor arriving from an AI-generated answer arrives with a precise expectation already formed, and a slow page is a sharper disappointment against a precise expectation. Speed is one of the parts of this series’ subject that the next phase of search will not make less important; a business that has made its site genuinely fast has made a durable improvement.<\/p>\n

Related reading<\/h2>\n