Local business website conversion: the complete guide

A local business website’s job is narrower than most website advice assumes: get someone who has already decided you might be worth contacting to actually do it, with the least possible friction, on whatever device they’re using — usually a phone.

Getting found and getting chosen are different problems, and a site that wins visibility but fumbles the moment someone lands on it is throwing away money it already spent to earn that visit. See the FOUND, TRUSTED, CHOSEN framework for how the two connect.

This is the evidence-based version of that argument: what should actually sit above the fold and in what order, how form length trades off against lead quality with real completion-rate data behind it, why a phone call deserves to be treated as a first-class conversion path rather than a consolation prize for a broken form, what a genuine controlled experiment says page speed is worth, and where the mobile-versus-desktop gap is real and where it’s simply assumed.

The underlying numbers come from this site’s own local business conversion rate statistics — platform data at real scale, with every widely-repeated figure that didn’t survive being traced to its origin flagged rather than quietly reused.

Above the fold has two jobs, and neither of them is looking impressive

The most common failure on a local business homepage isn’t a missing element — it’s the wrong ordering of the right ones. A hero image, a slogan, a navigation menu and a “learn more” button that leads to another page before anyone can act. That structure was built for brand awareness, not for a visitor who has already decided you’re worth a look and is deciding, in the next few seconds, whether to bother.

Above the fold has exactly two jobs: prove you’re a real, legitimate business, and remove the first click of friction toward contacting you. Everything else (full service detail, the complete portfolio, an extended FAQ) belongs below it, where it costs a scrolling visitor nothing but doesn’t delay the two things that actually convert.

Figure 1

Local business landing page

1

2

3

4

5

6

1. 1 Sticky call bar A tappable tel: link, and a WhatsApp icon where the audience genuinely uses it, pinned through the whole scroll — not just in the header. This is the fix the click-to-call data below makes hardest to justify skipping.
2. 2 Headline and specific proof What you do, where, and one line establishing this isn't a stranger — a credential, years trading, a review count. Not a slogan.
3. 3 The primary action, doubled A call button and a short enquiry-length form sitting side by side. Neither should be buried under the other.
4. 4 Services and proof, in detail Photos, specifics, what's included. Below the fold is the right place for the depth a research-mode visitor is looking for.
5. 5 The full contact form Longer than the primary short form if the job genuinely needs it — its length should match what's being asked, not a fixed template.
6. 6 FAQ and close The objections a phone call would have surfaced anyway, answered before they become a reason to leave.

Two things in that layout are easy to get wrong even when the intent is right. The sticky element has to survive the whole scroll, not just sit in a header that disappears the moment someone moves past it — a call bar that vanishes after the first screen has quietly reintroduced the exact friction it was built to remove.

The “primary action, doubled” band means genuinely offering two paths at once, not defaulting every visitor into a form because that’s easier to build: someone reaching for their phone shouldn’t have to find and dismiss a form first.

Start with the phone number, literally

A phone number displayed as an image, or as plain text that doesn’t trigger a tap-to-call action on mobile, is the single most common, lowest-effort-to-fix conversion problem on local business websites. Confirm every instance of your phone number is a proper tel: link, visible without scrolling on mobile, ideally in the sticky position shown above rather than only in a header that scrolls away.

It’s worth being specific about why this is worth fixing before anything else: independent call-tracking data covering 70 million-plus calls found 56% of callers reach an actual person, and 38% of those become a qualified lead. A tel: link that doesn’t work, or a number rendered as an unclickable image, doesn’t make that conversion path perform worse — it closes it entirely for anyone who tapped expecting a call to start.

Add a WhatsApp option where it’s genuinely used

In markets where WhatsApp carries significant discovery and communication load (see our WhatsApp Business statistics), a WhatsApp click-to-chat button next to or instead of a phone number can outperform both a phone number and a form, because it matches how customers already communicate.

Match your form’s friction to what you’re actually asking

The advice to “keep forms short” is close to correct and still the wrong level of resolution. The more useful finding, from form-analytics data across 739 forms and more than 93 million form views, is that a form’s stated purpose predicts its completion rate independently of length — two forms with the same number of fields convert very differently depending on what the visitor believes they’re committing to.

Source: Zuko Analytics, Form Type Benchmarking — 739 forms, 93,022,997 form views, view-to-completion rate. “Enquiry” and “Contact” are Zuko’s own separate categories; confirm which one your form actually resembles before treating either gap as automatically meaningful.

Registration forms convert highest not primarily because they’re short, but because someone filling one in has usually already decided to proceed: the form is a formality, not a decision point. Enquiry forms convert lowest because the visitor is still deciding whether you’re even an option worth pursuing, and every additional field reads as a cost being asked of someone who hasn’t committed to anything yet.

The practical implication: a first-contact enquiry form, the one that sits above the fold for a visitor who just arrived, should be the shortest, least demanding form on the site, and a longer, more qualifying form is better placed later in the journey, once the visitor has already decided you’re worth pursuing.

Field length isn’t the only variable, either. HubSpot’s own analysis of the three form fields that most depress landing page conversion, drawn from 40,000-plus customer landing pages, found that age, phone number and street-level location fields specifically correlate with lower conversion, beyond what field count alone predicts. Each of those is worth questioning individually before assuming the fix is simply “remove one field, any field.”

That finding is directional rather than a precise per-field percentage, and it’s worth being explicit about what doesn’t hold up: a table claiming exact conversion rates per field count (18.2% for one field, down to 4.2% for nine) circulates widely under a vague “HubSpot” attribution, but no version traces to an actual published study, and at least two mutually inconsistent versions of the same table exist — a clean sign that neither is real.

Treat the direction as reliable and any precise field-by-field percentage you see quoted as unverifiable.

Device matters here too, and not only for phone numbers. The same 739-form dataset shows desktop forms completing at 54.5%, tablet at 49.3%, and mobile at 47.5% — a real but moderate gap. On mobile specifically: use type="tel" and inputmode="numeric" so the correct keyboard appears automatically, enable autofill rather than fighting it, and check that every field is large enough to tap accurately with a thumb rather than a mouse cursor.

Where a longer form is genuinely unavoidable (a detailed quote request that needs real specifics to be useful), a multi-step layout that reveals length progressively is worth testing against a single long page rather than assumed to help by default: completing step one creates a small commitment toward finishing the rest, which a single intimidating page doesn’t offer.

If a qualifying question (budget range, timeline, service needed) is added deliberately to trade volume for quality, the lead value calculator and the cost per lead calculator are the tools to check whether that trade is actually paying off, rather than assuming a shorter list of leads is automatically worse.

Treat click-to-call as a first-class conversion path, not an afterthought

Every figure in the section above measures a form submission. For a large share of local businesses (most trades, medical and legal searches especially), a phone call is a separate, and often larger, conversion type entirely, with its own funnel that a form-only view of the website simply doesn’t see.

Figure 2

Call placed Converts

1. 01 A person answers 56% of callers reach an actual person rather than voicemail, an automated system, or going unanswered — the ceiling on everything below. Invoca, 70M+ calls
2. 02 The call becomes a lead 38% of answered calls are qualified as a lead. of answered calls
3. 03 The lead converts on the call 42% of qualified leads convert during the call itself, not afterward. of qualified leads

Source: Invoca, Lead Conversion Benchmarks Report, 70 million-plus calls and 600 million-plus minutes of conversation across 10 industries including Home Services, United States, published July 2026.

Google’s own Local Services Ads product (covered in more detail on this site’s local lead generation statistics) already treats an answered call as an equally valid lead event to a form fill. Google is pricing calls as a first-class conversion type, not a fallback. What the Invoca data adds is what happens once the call actually connects, which is a step further down the funnel than Google’s own lead-pricing documentation discloses.

The practical implementation follows directly from where the ceiling sits. Since a call that never reaches a person can’t convert whatever your own team does on the ones that connect, phone number placement and prominence deserve as much design attention as the click-to-call button’s visual design:

• Sticky, not scrollable-away. The number needs to survive the whole page, in the position shown in Figure 1 — not only in a header that disappears after the first screen.
• A real tel: link, not an image or unformatted text. Covered above, and worth repeating here because it’s the precondition for every figure in this section applying to your site at all.
• A button that reads as a button, not a label. High contrast against its background, a large enough tap target for a thumb, and ideally an icon alongside the number rather than the number alone — the visual difference between “here is our number, for reference” and “tap here to call now” matters more than it looks like it should.
• Placed in the thumb zone on mobile, not only at the very top of the page where a one-handed user has to stretch to reach it.

Once the call connects, what happens next is measured by call tracking rather than by the website — see our call tracking statistics for how many calls actually go unanswered, which channels produce calls that turn into real appointments rather than just conversations, and how long a call needs to run before it’s likely to be a genuine enquiry rather than a misdial.

A controlled experiment says what page speed is actually worth — not a rule of thumb

Most page-speed advice for local businesses rests on correlation, or on a decade-old rule of thumb repeated without its date attached. Google’s own 2022 case study of Rakuten 24, an online retailer, is different: a genuine controlled A/B test, half of traffic on the original site and half on a Core Web Vitals-optimised version, over one month, holding everything else constant.

Source: Google web.dev, Rakuten 24 case study, published 24 August 2022. One company, one-month controlled A/B test, Japan.

Revenue per visitor moved by more than conversion rate did in the same test, which points to a second effect beyond the single decision to convert or not: visitors on the faster version also spent more per order, suggesting speed shapes browsing behaviour more broadly, not just the final click.

Separately, the same case study reports a correlation between good Core Web Vitals scores and conversion in its real-user monitoring data, a 61.13% lift associated with a good Largest Contentful Paint score and 55.88% with a good First Input Delay score.

First Input Delay is no longer one of Google’s Core Web Vitals — Interaction to Next Paint replaced it as a stable Core Web Vital on 12 March 2024 — so that half of the case study measures a metric that has since been retired.

That second figure is also a correlation from field data, not a second controlled experiment, and it can be confounded by other factors that differ between fast and slow sessions — genuinely supporting evidence, not independent proof.

This is a single ecommerce company’s result in a market outside most local businesses’ own, so treat the precise percentages as a ceiling of what’s possible rather than a number any specific business should expect to replicate.

What travels reliably from ecommerce to a local trades or services site is the mechanism, not the exact figure: a visitor waiting for the largest element on the page to finish rendering is a visitor deciding whether to stay before they’ve seen what you actually offer, and that decision point exists on a plumber’s landing page exactly as it does on an online shop.

For the fuller technical picture (Core Web Vitals pass rates across the mobile web generally, and where local business sites are most likely to be losing the metric that matters most) see our page speed statistics.

Checking your own Core Web Vitals status costs nothing and takes minutes: PageSpeed Insights and the Chrome UX Report both report it free of charge. Where the fix turns out to be structural rather than a single oversized image — a theme that ships four times more JavaScript than the page uses, say — it belongs with the rest of the website design and development work rather than being patched at the margins.

One number worth retiring while we’re here: the claim that “53% of mobile visitors abandon a site that takes longer than three seconds to load” genuinely traces to a real study: Google’s own “the need for mobile speed” post, published on its DoubleClick blog on 8 September 2016, with the underlying data collected in March 2016.

The problem is entirely the date. It predates Core Web Vitals as a measurement framework by four years, and describes a mobile network that has moved through two more generations since it was measured.

It may still be directionally true that slow pages lose mobile visitors (the Rakuten case above is more recent, controlled evidence for that same direction), but the specific “53% at three seconds” figure isn’t current evidence, and most roundups quoting it in 2026 leave the 2016 out of the sentence.

Design decisions that should differ by device, and the ones that shouldn’t

Mobile carries most local search traffic, and desktop still out-converts it on average — both true, and neither tells you what to actually change, because the size of that gap turns out to depend heavily on what the visitor is being asked to do.

Source: Unbounce, Conversion Benchmark Report — the only two industries where Unbounce discloses absolute desktop and mobile conversion rates on the same page, rather than only a relative gap. The all-industry figure sits at an 8% relative gap favouring desktop; no local-services or home-services device breakdown has been published in the sources reviewed for this guide.

Ecommerce shows a real, sizeable gap. SaaS shows almost none, and on a dataset where 79% of SaaS visits happen on mobile, that near-parity isn’t a minor footnote.

A business that assumes “desktop always wins” industry-wide would be wrong for SaaS, and no comparably specific figure exists yet for local trades or services. That’s an honest gap in what’s currently measurable, not a reason to guess: no substantial study isolating a local-services mobile-versus-desktop conversion gap was identified in the sources reviewed on 8 September 2026.

If your own site converts noticeably worse on mobile than desktop, that’s worth diagnosing directly rather than accepting as structural: a phone number that isn’t a working tel: link, a form that’s genuinely difficult to complete one-handed, or text small enough to require pinch-zooming are all fixable faults that would produce exactly this pattern, and are more likely explanations than an inherent mobile disadvantage.

Given how much local search happens on a phone in the first place (per our local business website statistics, 16% of US adults are smartphone-only internet users), a wide gap on a local site is usually more diagnostic than the same gap on a desktop-heavy B2B product.

What’s reasonable to build differently by device, based on what actually differs (attention span, input method and context) rather than an assumed conversion penalty:

• Mobile: design for the thumb, not the cursor. Sticky CTAs in reach of a one-handed grip, tap targets sized for a fingertip rather than a pointer, no interaction that depends on hover.
• Mobile: remove typing wherever a tap will do. Numeric keypads for phone fields, autofill enabled rather than disabled, dropdowns and toggles over free-text where the answer set is small.
• Desktop: the extra screen real estate is for proof, not for a longer form. A desktop visitor is often in research mode with more patience for detail — use that for photos, credentials and specifics, not as licence to lengthen the primary form beyond what the mobile version asks for.
• Both: never let the CTA depend on scroll position guesswork. Test what’s actually visible without scrolling on the real devices your visitors use, not a single desktop browser window.

Build trust signals that are specific, not generic

Stock photography, vague claims of quality, and unverifiable superlatives don’t move a sceptical visitor. What tends to work better:

• Specific proof — real project photos, a genuine before-and-after, a named credential
• Reviews surfaced directly on the page, not just linked to an external profile — how to get those reviews in the first place is a separate operational problem from displaying them
• A visible, specific response-time promise (“we respond within one business hour”), which does double duty: it reassures the visitor and creates internal accountability to actually meet it

Decide deliberately on pricing transparency

There’s no universally correct answer here, and treating it as settled is a mistake. Showing prices can filter enquiries toward people who are already qualified on budget — fewer leads, likely better quality. Hiding them keeps every visitor in the funnel but risks a conversation that starts with sticker shock. Test rather than assume, and if you do test, measure the change in lead quality, not only volume.

A worked example: three changes to a landing page that reads like most local business sites

The following is a deliberately generic, illustrative example — not a real client audit, and not a real business. It’s built to show how the reasoning above applies to a page, not to report a result.

Picture a boiler repair and servicing company’s homepage, built two years ago and untouched since. Desktop-first, then squeezed to fit mobile. It performs reasonably by the owner’s own sense of things, but nobody has looked at why since it went live.

None of these four changes requires a redesign. They’re a link attribute, a shortened field list with a second form added elsewhere, an image export setting, and a reordering of what’s already on the page — which is roughly the pattern across most of what actually moves conversion on a local business site: less new content, more correcting what’s already there to match how people are actually deciding.

A conversion checklist, in leverage order

• Phone number is a real, tappable tel: link, sticky and visible through the whole scroll on mobile — not just in a header that disappears
• A WhatsApp option added where the audience genuinely uses it
• The above-the-fold form is the shortest one on the site, matched to an enquiry-level ask; longer, more qualifying forms sit further down the journey
• Phone number fields use type="tel" and inputmode="numeric"; age, precise location and other high-friction fields are questioned individually, not assumed necessary
• Core Web Vitals checked directly via PageSpeed Insights, not assumed adequate — with particular attention to the largest visible element’s render time
• Mobile layout genuinely designed for mobile, not the desktop layout shrunk to fit — CTAs in the thumb zone, tap targets sized for a finger
• Reviews and specific proof surfaced directly on the page
• A visible, specific response-time promise
• A deliberate decision, tested rather than assumed, on pricing transparency

What conversion work can’t fix

A well-converting website still depends on what happens after the enquiry lands. See our lead response time statistics — the evidence there suggests the largest single leak in local business lead generation happens after the click, not on the page itself. Conversion optimisation gets someone to raise their hand, whether by call or by form. It doesn’t answer the phone for you.

It also cannot fix a site whose template will not allow the changes. Past a certain point the conversion work stops being edits and becomes a rebuild, which is website development work rather than optimisation. Where this sits in the wider argument is covered in the local lead generation guide.

Questions

What's a good conversion rate for a local business website?

The cross-industry median for a dedicated landing page is 6.6%, based on genuine platform data (41,000+ pages, 464 million visitors, 57 million conversions), and home-services lead-generation pages specifically run higher, around 8.5%. See our local business conversion rate statistics for the full breakdown and a caveat about who's actually in that sample — businesses that built a dedicated landing page are more conversion-conscious than the average local business site.

Should my phone number be clickable on mobile?

Yes, without exception. This is one of the lowest-effort, highest-leverage fixes available on a local business website, and its absence is still common enough on real sites to be worth checking specifically. Independent call-tracking data shows 56% of callers reach an actual person and 38% of those become a qualified lead — a phone number that isn't a proper tel: link closes off a conversion path that's already working, before it can even start.

How many fields should a contact form have?

Fewer fields generally increase completion rate, but the more useful finding is that purpose matters more than length alone: form-analytics data across 739 forms shows registration forms completing at 60.7% and enquiry forms at 42.7%, even though both can be roughly the same length. Match the form's friction to what it's actually asking — a first-contact enquiry form should be the shortest thing on the site.

What's the difference between a contact form and an enquiry form, and does it matter for conversion?

In the largest form-benchmarking dataset publicly available, they're measured as separate categories and complete at meaningfully different rates — contact forms at 47.4%, enquiry forms at 42.7%, both below purchase (54.4%) and registration (60.7%) forms. The practical implication is that a low-commitment, top-of-funnel form (someone just finding out if you're an option) needs to ask for less than a form someone fills in once they've already decided to proceed.

How much does page speed actually affect conversion rate, with real evidence?

In a controlled one-month A/B test, Google's own case study of an online retailer found that fixing Core Web Vitals issues increased conversion rate by 33.13% and revenue per visitor by 53.37%, against an unoptimised control on the same site. That's a single company in a different industry to most local businesses, so treat the exact percentages as a ceiling rather than a promise — but it's genuine controlled evidence, not a rule of thumb, that a slow page is actively costing conversions rather than merely being a minor inconvenience.

Should click-to-call be treated as more important than the contact form?

For most trades, medical and legal local searches, treat it as at least equally important, not as a fallback for people who couldn't find the form. Call-tracking data shows 56% of callers reach a person and 38% of those become a qualified lead — a website's form conversion rate alone systematically understates total conversion wherever a meaningful share of mobile visitors are tapping to call instead of filling anything in.

Does mobile really convert worse than desktop for every type of business?

No — the direction is common but the size varies enough that it shouldn't be assumed. Platform data shows an 8% average gap favouring desktop, but the actual per-industry figures range from a 4.2-point gap in ecommerce to virtually no gap, or a slight mobile lead, in SaaS. No comparable local-services breakdown has been published, so a wide mobile-versus-desktop gap on your own site is worth diagnosing directly — a broken tel: link or an unusable mobile form is a more likely explanation than an inherent mobile disadvantage.

What should go above the fold on a local business website?

Two things doing distinct jobs in the same viewport: something that proves you're a legitimate, real business (a credential, a review count, a specific claim — not a slogan) and a frictionless way to act immediately, ideally a call option and a short form offered together rather than one buried under the other. Everything else — full service detail, photos, extended proof — belongs below the fold, where it costs a visitor nothing to scroll to but doesn't delay the two things that actually convert.

Is the '53% of visitors abandon after 3 seconds' statistic still true?

That specific figure traces to a real 2016 DoubleClick mobile-network study, published via Google's own blog in 2017 — but it predates Core Web Vitals as a measurement framework by four years and describes a mobile network that no longer exists in most markets. It may still be directionally true that slow pages lose visitors; the more current evidence for that is Google's own 2022 Rakuten case study, a controlled experiment showing a 33.13% conversion lift from fixing Core Web Vitals issues. Quote that instead of the 2016 figure with a 2026 byline.

Does showing prices on my website help or hurt conversion?

The honest answer is that it isn't settled by strong evidence either way. Pricing transparency can filter out unqualified enquiries — fewer leads, but better quality — or it can deter price-sensitive prospects before they reach a conversation where value could be explained. Testing on your own site, and measuring the change in lead quality rather than only volume, is the only way to know for your specific business and price point.