Web Hosting Statistics Updated November 2026

Human Written & Fact Checked · Updated

100 Hosting Provider Statistics for 2026

A hosting-provider ranking changes when its denominator changes. Measured website usage, customer accounts and company revenue are useful observations, but they answer different questions.

This page combines original technology measurements with provider disclosures. Company financial figures explicitly include products beyond web hosting; they are not estimates of a hosting-only market.

Headline figures

  • Global sampled websites: Shopify: 5.4% website usage (9 November 2026). (W3Techs / Q-Success, Nov 2026)
  • Global sampled websites: Hostinger: 5.2% website usage (9 November 2026). (W3Techs / Q-Success, Nov 2026)
  • GoDaddy global business: GoDaddy customers: 20.5 million (Q3 2026). (GoDaddy, Oct 2026)
  • IONOS global business: IONOS customers: 6.91 million (H1 2026). (IONOS Group, Aug 2026)
  • OVHcloud global business: OVHcloud served 1.7 million customers (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • OVHcloud global business: Connected power capacity: 250 MW (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • DigitalOcean global business: Committed datacenter capacity: approximately 155 MW (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • OVHcloud global business: Webcloud revenue: €199.7 million (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • DigitalOcean global business: DigitalOcean revenue: $281 million (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • OVHcloud global business: Group revenue: €1,145.1 million (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • GoDaddy global business: Normalized EBITDA: $434.1 million (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • OVHcloud global business: France: 48% of group revenue (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)

Which hosting providers appear in website measurements?

Website usage measures a technical relationship instead of customer spending. Hosted site builders and infrastructure providers appear together because they can both be detected serving websites.

Which larger hosting groups appear in the snapshot?

This is a website-detection snapshot, not a count of subscriptions. A site builder can appear alongside a general infrastructure provider because both serve websites. Company-group labels may combine multiple brands. The observations do not measure customer spending or reliability.

How visible are specialist hosting providers?

These entries describe the same worldwide measurement population as the larger groups. Smaller observed shares do not establish weaker infrastructure or support: the unit remains a detected website, irrespective of traffic or workload. Keep provider group names intact when comparing snapshots.

Which smaller providers remain visible in the sample?

Regional positioning and commercial branding do not change the denominator of this table. A provider can host multiple sites for one account, and a single business may use several providers. These observations provide technology-footprint evidence instead of a revenue estimate.

How large are providers' customer bases?

A customer account is not a website. Customer additions, average account revenue and revenue retention describe different aspects of a provider's commercial relationship.

What did providers report about customer additions?

  • GoDaddy added 40,000 customers during the quarter (Q3 2026). (GoDaddy, Oct 2026)
  • IONOS added approximately 280,000 customers in the half-year (H1 2026). (IONOS Group, Aug 2026)
  • Customer additions during the quarter: 22,000 (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • Customer additions since December: 35,000 (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)

Read each observation at its own reporting date. A half-year customer change must not be compared as if it covered the same duration as a quarterly change. Provider definitions also differ: an account can buy domains, hosting and other services, while several sites can operate under one account. Revenue retention concerns spending by a defined cohort instead of the percentage of people who renew. Average revenue per user is a company-defined operating metric and is not a quoted monthly hosting plan. These distinctions prevent an apparently simple league table from becoming misleading.

How should customer value and retention be interpreted?

  • IONOS prior-year-half customers: 6.47 million (H1 2025). (IONOS Group, Aug 2026)
  • Average revenue per user: $250 (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • Like-for-like net revenue retention: 103% (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)

Customer counts alone cannot explain the value of the underlying accounts. Some accounts purchase several products or operate multiple websites, while others purchase a domain without a hosting plan. GoDaddy's ARPU uses trailing annual revenue and its defined customer base; it is not a monthly subscription price. OVHcloud's retention measure concerns revenue from a cohort instead of customer headcount. The prior-year IONOS base is a comparator within that company's own reporting framework. These observations should not be pooled into a supposedly standard customer-value ranking across providers.

Where do providers earn revenue and operate capacity?

Customer geography and datacenter geography answer different questions. Revenue earned from a region does not establish where its websites or backups are physically stored.

What geographic revenue and footprint does OVHcloud report?

  • Other Europe: 29% of group revenue (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Rest of world: 23% of revenue (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • OVHcloud operated 46 datacenters (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • OVHcloud customers spanned over 140 countries (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)

A commercial footprint describes the reach of a group, not the location of every customer workload. Country coverage does not guarantee that a datacenter operates in each country. Power capacity is another separate measure: connected capacity, operating capacity and a commitment for future delivery are not interchangeable. None measures the number of websites that can be hosted without assumptions about utilisation, hardware and workload. The figures provide scale and expansion context, but they cannot independently support latency, data-residency or uptime claims. Site-specific requirements need the provider's applicable region and service documentation.

How does operating capacity differ from committed capacity?

  • Operating power capacity: 140 MW (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • OVHcloud operated over 500,000 servers (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Additional committed future capacity: 20 MW (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)

Connected power provides a different view of infrastructure from the amount already operating. Likewise, a signed datacenter-capacity commitment is not proof that the associated systems are currently available to customers. DigitalOcean describes its additional capacity as expected to arrive in future periods. OVHcloud's physical server count describes another unit entirely. Converting any of these measures into hosted websites would require assumptions about utilisation and workload. A small content site and a resource-intensive application have very different needs, even when they use the same provider and geographic region.

Which products and customer segments generate revenue?

Provider revenue is broader than web hosting. Segment disclosures help identify that boundary instead of treating every cloud dollar as a hosting-market observation.

How does OVHcloud separate customer segments?

  • Digital Starters revenue: €563.9 million (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Digital Starters: 49% of group revenue (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Digital Starters like-for-like growth: 5.4% (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Digital Scalers revenue: €359.1 million (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Digital Scalers like-for-like growth: 12.4% (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Corporates revenue: €221.7 million (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Corporates growth: 2.6% (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)

These are customer segments, not alternative names for the company's product divisions. Their revenue should not be added to Private Cloud, Public Cloud or Webcloud revenue, which presents another view of the same business. Retaining this distinction makes the figures useful for understanding the provider's customer mix without double counting its activity.

How do private cloud and public cloud differ commercially?

  • Private Cloud revenue: €690.4 million (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Private Cloud: 60.3% of revenue (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Public Cloud revenue: €255.1 million (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Public Cloud: 22.3% of revenue (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)

OVHcloud reports these as separate product divisions, so their revenue describes the company's mix of products instead of the whole cloud market. Private Cloud is not a synonym for shared web hosting, and Public Cloud can support many applications other than websites. Revenue shares express the division's contribution to the reporting group. They say nothing by themselves about the number of customers, website traffic or physical infrastructure allocated to that division. Cross-company comparisons need compatible definitions before treating similarly named segments as equivalent.

What do web-focused and broader product divisions report?

  • GoDaddy Applications and Commerce revenue: $524.3 million (Q3 2026). (GoDaddy, Oct 2026)
  • GoDaddy Core Platform revenue: $817.5 million (Q3 2026). (GoDaddy, Oct 2026)
  • Webcloud: 17.4% of revenue (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)
  • Webcloud like-for-like growth: 3.2% (Fiscal year ended 31 August 2026). (OVHcloud, Oct 2026)

The web-focused label helps narrow the interpretation, but the original OVHcloud table explicitly includes Other alongside Web Cloud. GoDaddy uses a different structure, separating Applications and Commerce from Core Platform. Those two providers therefore cannot be compared as though their divisions contain identical products. In particular, domain registrations and complementary business services can sit alongside website services. The figures remain useful for understanding how each provider organises its business. They do not supply a clean, directly comparable measure of shared-hosting revenue or the average price paid for a website.

How do provider revenues and contractual demand compare?

Revenue, bookings and contractual obligations describe different stages of a commercial relationship. Adding them would count activity more than once.

What do current-period group revenues show?

  • GoDaddy revenue: $1,341.8 million (Q3 2026). (GoDaddy, Oct 2026)
  • IONOS revenue: €701.1 million (H1 2026). (IONOS Group, Aug 2026)
  • IONOS reported revenue growth: 6.9% (H1 2026). (IONOS Group, Aug 2026)
  • IONOS currency-adjusted revenue growth: 8.2% (H1 2026). (IONOS Group, Aug 2026)
  • DigitalOcean revenue growth: 29% (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)

Recognised revenue relates to services delivered under the applicable accounting rules. Bookings can reflect orders or renewals before the corresponding revenue is earned. An annualised operating metric projects a current measure over a year and should retain the issuer's exact definition. Remaining performance obligations relate to contracted future delivery; they are not cash already received or a guarantee of next-quarter revenue. The reporting periods also vary across companies. These figures show business scale and demand context, not a controlled comparison of hosting prices, growth quality or the size of the pure web-hosting market.

How do bookings and annualised metrics differ from revenue?

  • GoDaddy bookings: $1.4 billion (Q3 2026). (GoDaddy, Oct 2026)
  • GoDaddy Airo annualized bookings run rate exceeded $100 million (Q3 2026). (GoDaddy, Oct 2026)
  • Annual run-rate revenue: $1,125 million (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • Incremental annual run-rate revenue: $93 million (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)

GoDaddy defines bookings through the value of customer contracts entered into during the reporting period, whereas recognised revenue follows delivery over the contract term. Its Airo run rate concerns a particular product initiative and cannot be added to group revenue as a separate market. DigitalOcean's annual run-rate revenue describes a different company-defined annualisation. Incremental run-rate revenue is a change in that metric, not another independent stream of sales. These distinctions matter because superficially similar dollar figures can describe orders, current delivery or an annualised snapshot of activity.

What do contractual demand and overseas revenue show?

  • Remaining performance obligations: $894 million (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • RPO expected within twelve months: $366 million (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • GoDaddy international revenue: $435.5 million (Q3 2026). (GoDaddy, Oct 2026)
  • International revenue: $427.1 million (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • International revenue growth: 7.9% (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)

The contractual observations belong to DigitalOcean; the overseas-revenue observations belong to GoDaddy. They should remain separate instead of being interpreted as one regional pipeline. DigitalOcean changed its RPO presentation to cover contracts regardless of original term, an important qualification for comparisons with older releases. The amount expected within the following year is part of the total obligation, not additional demand. GoDaddy's international figures describe recognised revenue and reported growth for its own global business, with product scope extending beyond hosting. Neither disclosure establishes where customers' servers are physically located.

What do provider operating results reveal?

Financial resilience can matter to a long-lived hosting relationship, but profitability does not establish service reliability. Keep accounting profits, adjusted measures and cash flow separate.

How do reported operating income and margins compare?

  • Operating income: $29 million (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • Operating income margin: 10% (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • Operating income: $342.5 million (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • Operating income margin: 26.4% (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • Net income: $240.1 million (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)

Each issuer defines its adjusted measures and publishes its own reconciliation. EBITDA-based figures omit costs that remain economically important, while an operating margin and a cash-flow margin use different numerators. Segment results can also exclude group-level costs. These disclosures should therefore be read within a company's reporting framework before any cross-company comparison. A dollar amount and a euro amount cannot be ranked without an explicit exchange-rate convention. The figures are company-level business evidence and do not predict uptime, security performance, support quality or the cost of migrating an individual website.

What do adjusted profit measures describe?

  • IONOS adjusted EBITDA: €245.2 million (H1 2026). (IONOS Group, Aug 2026)
  • IONOS adjusted EBITDA margin: 35.0% (H1 2026). (IONOS Group, Aug 2026)
  • Adjusted EBITDA: $114 million (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • Adjusted EBITDA margin: 40% (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • Applications and Commerce segment EBITDA: $241.1 million (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • Applications and Commerce EBITDA margin: 46.8% (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)

IONOS and DigitalOcean publish adjusted EBITDA, while GoDaddy reports profitability for its own business segments. These measures should not be treated as identical simply because they share an EBITDA label. Each company determines its adjustments and segment-cost allocation in the accompanying reporting framework. Margins express the measure relative to the relevant revenue base, so segment and group margins have different denominators. The disclosures provide context about the provider's business model, but are not direct evidence of infrastructure efficiency, service reliability or the customer support experienced by an individual subscriber.

How do segment profitability and cash generation differ?

  • Operating cash flow: $110 million (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • Adjusted free cash flow: $61 million (Quarter ended 30 June 2026). (DigitalOcean, Aug 2026)
  • Operating cash flow: $442.5 million (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • Core Platform segment EBITDA: $261.9 million (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • Core Platform EBITDA margin: 33.4% (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)
  • Free cash flow: $443.5 million (Quarter ended 30 June 2026). (GoDaddy, Jul 2026)

Cash generation and segment profitability answer different questions. Segment EBITDA can exclude costs borne centrally, while free cash flow reflects the issuer's own adjustments. Compare each amount within its original reporting period and currency. None is a direct measure of hosting reliability, support quality or the expected uptime of an individual website.

Glossary

Methodology and sources

Every figure on this page comes from the organisation that produced it: regulators, statistics offices, public bodies, research institutes, and companies reporting their own data or surveys. Each one links to the original report or dataset and carries its release date. Survey results keep their population and fieldwork period, and financial figures keep the measure the producer used. Figures from different producers are not added together or ranked against each other. The page is reviewed monthly and updated when a producer releases newer data.