Google quietly admits not every business needs a website

Do You Still Need a Website in 2026 for Marketing?

Google's Search Relations team publicly questioning whether every business still needs a website is less about websites "dying" and more about which businesses can safely build only on rented platforms versus which still need to own a slice of the open web.

This analysis examines how Google's "it depends" answer re-prices the value of owning a website versus running only on social, apps, or closed ecosystems, and what that means for channel mix, risk, and budgets in 2026.

Key Takeaways

Websites shift from default to strategic asset: For some models (mobile games, hyperlocal sellers, micro-creators), social or app-only setups can work, but for most revenue-focused brands the site becomes a control layer for data, monetization, and resilience - not just a digital brochure.
Channel mix decisions should be ROI- and risk-weighted, not habit-based: Map where demand actually starts (search, social, word-of-mouth, app stores) and where you capture value. For any business with more than one primary acquisition channel, a lightweight but controlled website usually pays off through measurement, remarketing, and stronger negotiation power with platforms.
SEO's role changes from "get traffic" to "own your surface in a fragmented journey": Even if most discovery happens on social or chat, a site still anchors branded search, support content, and long-tail questions that AI summaries and answer engines will draw from.
Platform-only strategies need explicit risk buffers: If you run without a real site, you should intentionally build redundancy - multiple platforms, exportable customer lists, and at least a minimal web presence for legal, trust, and migration scenarios.

Situation Snapshot

A Search Off the Record podcast episode (about 28 minutes) featured Google's Gary Illyes and Martin Splitt discussing whether businesses "still need a website" in 2026. They repeatedly avoided a universal rule and answered "it depends" based on use case and audience [S1][S2].

Key factual points from the episode and SEJ reporting:

Advantages of having a website highlighted by Illyes and Splitt:
- Data control and ownership (analytics, first-party data).
- Control over monetization (direct sales, ad networks, subscriptions).
- Ability to host tools and services (calculators, configurators, gated content).
- Independence from platform content rules and moderation [S1].
Examples where a website was not central:
- Indonesian businesses (around 2015-2016) operating fully on social networks with strong sales and retention [S1].
- Mobile games with only minimal legal pages on the web, but massive revenue via app stores [S1].
- Illyes' own use of WhatsApp groups as the "real" presence for some communities [S1].
Trust and quality angle: Splitt stated he would prefer a strong, credible social profile over a poorly executed website [S1].
Open web stance: Illyes said that if you want maximum reach for information or services, a website is still likely the best option - but stressed this was personal opinion, not official policy [S1].

This comes against a backdrop where users find brands across search engines, social feeds, marketplaces, and chatbots rather than a single discovery channel [S1][S3].

Breakdown & Mechanics

Thesis: Google's "it depends" framing signals that a website is now one node in a broader discovery and conversion system, not an automatic requirement. The decision should weigh cost, control, and risk against how and where customers actually find and interact with you.

1. Discovery mechanics: where journeys begin

Mechanism: User need → opens preferred surface (Google, TikTok, Instagram, Amazon, WhatsApp group, app store) → finds brand → continues the journey across a mix of channels.

Recent global surveys show search engines and social networks now sit very close as top brand research channels, each used by around one-third of internet users [S3]. That means:

For visual, impulse, and culture-driven products (fashion, beauty, food trends), discovery may skew to social.
For information-heavy or high-consideration decisions (B2B software, healthcare, high-value services), discovery often starts on search engines and review sites.

If your discovery is 80-90% inside a closed platform (for example, TikTok Shop or an app store), a site shifts from "must-have" to "strategic nice-to-have" for pure acquisition - but it still matters for control and risk, discussed below.

2. Control and economics: own vs rent

Compare two simplified paths:

Website-led: User finds you (search or social) → lands on your site → you control event tracking, pricing, upsells, and the support experience.
Platform-only: User finds you inside a platform → buys or engages there → the platform controls data, rules, and to some degree pricing and visibility.

Impact:

Margins: Marketplaces and app stores often take 10-30% fees. Selling direct on your site can reclaim some of that spread.
Data: On your site you can build first-party data assets (email, purchase history). Platforms usually limit what you can export.
Negotiation power: With only one platform, you are dependent on its ranking algorithms and policies. With a site and multiple channels, you can reallocate budget if a platform tightens reach or increases fees.

Website ownership is therefore less about vanity presence and more about shaping unit economics and reducing dependency.

3. Risk surface: platform volatility vs open web

Platform-only strategy:

Platform policy shifts or bans can cause organic reach to collapse or accounts to be disabled.
You must appeal or migrate your audience slowly, often with only partial data.

Website-anchored strategy:

Even if one distribution channel shrinks, you still have a canonical brand home that users can search for directly.
You maintain a base for email, owned communities, and future channels.

For risk management, the presence of a site acts as insurance: low expected impact in stable times, high value during shocks.

Community reaction, based on prior debates, is likely to highlight this risk. Many SEOs report clients who were over-reliant on one platform suddenly losing traffic or monetization when rules changed. This is an informed pattern-level observation, not a quote from this episode.

Impact Assessment

Paid Search and Performance Ads

Effect direction: Mixed, with more segmentation by business model.

Who can function without a site:

App-first products using Google App Campaigns and store pages.
Local service providers using call ads, Local Services Ads, and Google Business Profiles as primary surfaces.

Who still gains from a site:

Any brand wanting granular conversion tracking, remarketing lists, and controlled landing pages for Performance Max and search campaigns.

Actions:

Audit current campaigns: how many rely on site-based conversions versus app or platform conversions?
For app-first or marketplace-first products, model whether adding a simple, fast site would enable profitable search or Shopping campaigns that are impossible today.

Organic Search and Content

Effect direction: Clear downside for platform-only approaches, especially outside pure entertainment or hyperlocal cases.

Without a site, you:

Surrender informational queries ("how to...", "compare...", troubleshooting) that could otherwise bring high-intent users.
Rely on platform profiles and possibly AI summaries that may not fully reflect your message or offers.

With a site, you:

Control canonical information that AI answer engines and knowledge panels can pull from.
Own branded search results beyond social profiles and third-party listings.

Actions:

Map "must-own" topics: brand terms, high-intent questions, FAQs, support queries.
Ensure there is at least a clean, fast, well-structured site section covering these, even if most marketing energy lives on social or in communities.

Social, Community, and Creator-Led Growth

Effect direction: More validation for social-first strategies - but with guardrails.

Social-only setups can:

Achieve rapid traction where community and content matter more than corporate polish.
Reduce friction for users who already live inside a given app.

But:

You inherit the platform's design, ad formats, and data limits.
Moving to another platform or adding a site later can be harder if you have not prepared basic assets (domain, brand system, email collection).

Actions:

If you choose social-first, document an explicit "escape plan": domain reserved, basic landing page in place, regular export of subscriber or customer lists where allowed.

Apps, App Stores, and ASO

Effect direction: App-dominant businesses have the strongest case for minimal sites, but still benefit from a solid web anchor.

App store pages can handle discovery, install, and payment, especially with search ads inside stores.

A web presence still helps for:

Support, FAQs, and longer-form education that is awkward inside an app listing.
SEO for brand terms and "app + use case" queries.

Actions:

Treat the site as an extended documentation and support hub feeding both users and search engines, even if acquisition is mostly via app stores.

Operations, Analytics, and Governance

Effect direction: Platform-only stacks are cheaper to maintain but weaker for measurement and governance.

Platform-only:

Faster to set up, with fewer technical maintenance tasks.
Fragmented measurement dashboards across social, marketplaces, and ad managers.

Website present:

More work (hosting, security, performance) but a more unified view of behavior across sources.
Easier compliance with privacy rules by managing consent and data flows centrally.

Actions:

For any business spending serious money on ads, treat a site plus analytics stack as infrastructure for attribution and experimentation, even if short-term sales are dominated by one marketplace or social platform.

Scenarios & Probabilities

Base case - Hybrid presence becomes standard (Likely, ~60%)

Most serious businesses maintain:
- One owned site (often streamlined).
- Two or three priority platforms (social, marketplace, app store).
Websites handle information depth, brand search, legal content, and data. Platforms handle discovery and engagement.
SEO budgets persist but shift from pure traffic goals to support for brand, content reuse, and AI visibility.

Upside for the open web - Regulation and user behavior favor sites (Possible, ~25%)

Privacy rules tighten data sharing inside walled gardens.
Ad costs in major platforms rise, pushing brands to look for cheaper, intent-rich channels.
In this path, owning a site and first-party data becomes a stronger commercial advantage, and SEO regains share as a cost-efficient acquisition driver.

Downside for sites - Platforms dominate discovery and commerce (Edge, ~15%)

Social commerce, marketplaces, and app ecosystems absorb most search-like behavior.
AI chat and assistants fetch answers without much direct site traffic.
Many small businesses keep only bare-minimum sites for compliance and basic info, while performance budgets flow mainly to closed ecosystems.

These probabilities are informed estimates based on current trajectories and may change as platforms, regulation, and user behavior evolve.

Risks, Unknowns, Limitations

Podcast is commentary, not policy: Search Off the Record often reflects internal thinking but does not set ranking rules or product roadmaps [S2]. Google can acknowledge successful platform-only models while still tuning algorithms around the open web.
Vertical and regional variation: The Indonesian social-only success examples may not generalize to markets with different infrastructure, regulations, or consumer behavior [S1]. Local payment and messaging habits matter.
Lack of fresh, comparable cross-channel data for 2026: Public reports typically lag by a year, so using 2023-2024 numbers to forecast 2026 introduces uncertainty [S3].
AI impact still fluid: How much AI assistants will depend on and reward traditional websites versus APIs and structured feeds is still uncertain.
Scope of this analysis: It assumes profit-focused, longer-term brands. Ultra-short-term arbitrage plays or hobby projects might rationally skip a site and rely entirely on whichever channel currently converts.

What could significantly change this reading:

New Google policies that strongly prioritize web content in AI results or search features.
Regulatory moves that either break up or further empower large platforms in how they handle commerce and discovery.
Reliable new data showing a sharp shift of brand discovery either back toward search or further into social and app ecosystems.

Sources

[S1] Matt G. Southern, 2026, News article, "Google's Search Relations Team Debates If You Still Need A Website."
[S2] Google, 2026, Podcast, "Search Off the Record" episode discussing whether businesses still need websites.
[S3] DataReportal / We Are Social / Meltwater, 2023, Report, "Digital 2023: Global Overview Report" - sections on channels people use for brand discovery.