Disqus
Summer 2007AcquiredAdd comments to your website.
Pressure-test this opportunity
Explore the risks and possibilities with a prompt for ChatGPT, Claude, or your agent.
Add comments to your website.
Explore the risks and possibilities with a prompt for ChatGPT, Claude, or your agent.
Disqus turned blog comments from a page-level feature into a network. Founded by Daniel Ha and Jason Yan in 2007, the service gave publishers an embeddable discussion system and gave readers an identity that traveled across participating sites. By May 2011, the company reported 750,000 websites, about 35 million participants, and nearly 500 million monthly unique visitors.[1][2]
Zeta Global acquired the company on December 5, 2017. The transaction did not end the product: Disqus became a Zeta business unit, and its team, terms, pricing, and publisher focus were initially unchanged. The deeper story is how an embedded community product became an advertising and data asset. That shift financed continued operation, but it also joined comment moderation, identity, consent, security, and cross-site data use in one unusually sensitive product surface.[3]
Ha and Yan decided at the end of 2006, while studying computer science, to work together on tools for forums, group discussions, and online community building. The packet does not establish how they first met, their school, or earlier employers. It does show that they began with a broader set of discussion tools rather than a fixed plan to build a commenting company.[4]
Early testers chose the direction. The blog-comment feature attracted the strongest response, so the founders made it the main product. Y Combinator accepted the company into its Summer 2007 batch and supplied enough money, advice, and connections for both founders to leave school and work full-time. Disqus launched publicly near the end of 2007, then raised more capital to expand.[1][4]
The product thesis joined identity and distribution. A reader should not have to rebuild reputation on every site, and a publisher should not have to create a comment system from scratch. Ha said, "People are more willing to contribute to websites when they feel they are rewarded or have control over their comments."[4] A portable identity could preserve a user's contributions while connecting conversations scattered across source sites, aggregators, and other discussion tools.
That architecture created two constituencies from the start. Publishers installed the software and governed their communities; readers encountered the same identity layer across many properties. Distribution came largely from enthusiastic bloggers and commenters, not a single anchor customer. The research packet contains only one verbatim founder quote, so a second founder quote cannot be supplied without inventing evidence.
The core product was a hosted comment system that a publisher embedded on its own pages. Readers could comment through a shared identity instead of creating a separate account for every site. Publishers received spam controls, moderation tools, analytics, and configurable community rules. The YC profile later described three connected products: the embedded comment network, an advertising network, and an owned-and-operated destination.[1]
The network mattered because each install improved distribution for the identity layer. A reader who met Disqus on one publication could encounter the same profile elsewhere. The publisher kept discussion on its property while outsourcing accounts, comment delivery, and moderation infrastructure. In 2016, Disqus also relaunched its own site around followed content, cross-site comments, and standalone threads, although it soon considered reducing that consumer effort in favor of business data products.[7]
Moderation grew beyond deletion and spam filtering. Current materials describe restricted-word rules, holding linked comments for review, media limits, closing old threads, shadow banning, and a toxicity filter that uses Google's Perspective API to flag likely abuse or low-quality contributions.[12][13][14] These are publisher controls over a community product, not evidence that the service itself makes every moderation judgment.
Advertising became part of the embed. Sponsored comments launched in 2014, with Disqus taking a share of revenue through the free product. Current advertising materials place native Sponsored Stories between publisher content and the comment unit across more than 10,000 sites.[7][15] The current privacy policy also covers public comments, profiles, polls, cookies, browsing information, advertising-partner disclosures, and data sharing with Zeta.[16]
Disqus sold to publishers while serving their readers. In 2011, named customers included BarackObama.com, Time, IGN, NASDAQ, PCMag, ReadWriteWeb, and Wired.[2] Small sites could install the service cheaply or free; large publishers could buy VIP service, add-ons, and later ad or analytics capabilities. That split let the free embed drive distribution while paid and advertising products pursued revenue.
The packet contains network reach, not a reliable market-size series. Disqus reported 750,000 sites, 35 million participants, and almost 500 million monthly unique visitors in May 2011. It said traffic, users, and communities had each grown at least 500 percent over the previous year. A third-party study cited by the company claimed a 75 percent share among sites using third-party commenting systems, but that claim is low-confidence and was not independently audited.[2]
By 2016, TechCrunch reported roughly two billion monthly unique users across the network. That figure described potential reach, not registered or active commenters. The YC profile's later claim of more than four million websites is undated. No current active-site, commenter, comment-volume, or market-share metric appears in the packet, so those historical numbers should not be presented as current scale.[7][1]
Facebook Comments and Adobe-owned Livefyre were direct competitors in 2016. Native publisher tools could replace an external embed, while Facebook, Reddit, and other social destinations competed for the discussion itself. TechCrunch's acquisition coverage argued that conversation was migrating away from on-site comments toward social platforms. The packet does not quantify that movement, but the mechanism is clear: the publisher controlled the page while social platforms controlled concentrated identity, attention, and notification loops.[7][9]
Disqus's defense was the combination of installed distribution, portable identity, moderation infrastructure, and cross-site data. Its vulnerability was that the same cross-site reach carried consent and disclosure obligations no page-level comment widget could ignore. Zeta bought an active publisher network and interest data, not merely a user-interface component.
The company moved through three overlapping models. In 2011 it described VIP service and paid add-ons as its business. In 2014 it introduced sponsored-comment advertising and took a share of ad revenue. In 2017 it said more than 95 percent of sites could keep advertising optional, while larger commercial publishers on the free tier would receive configurable ads and could share revenue through Reveal.[2][7][8]
Current pricing still combines a comment plugin, spam filtering, moderation, analytics, configurable advertising, social login, shadow banning, and paid branding controls.[17] The current privacy policy says advertising is the predominant revenue source.[16]
Economics remain opaque. TechCrunch reported about $10.5 million in total funding by 2016, but the packet supplies no revenue, margin, publisher revenue-share, ad take-rate, customer-acquisition cost, or current contribution to Zeta results.[7] The 16-person team in 2011 and roughly 55-person pre-layoff team in 2016 show operating growth, but they do not support a credible burn estimate without compensation and infrastructure costs.
The strongest dated snapshot is May 2011: 750,000 sites, about 35 million participating users, nearly 500 million monthly unique visitors, a 16-person team, and a $10 million financing from North Bridge and Union Square Ventures.[2] Those company-reported figures show distribution, but only the participant count approximates product engagement.
The later two-billion-monthly-unique figure and undated four-million-site claim expand the reach story without clarifying active use. At acquisition, Ha said a small team supported millions of websites and hundreds of millions of users.[3] Current product, advertising, moderation, pricing, and privacy pages establish continuity, not current scale.
Zeta acquired Disqus on December 5, 2017. Consideration was not disclosed. TechCrunch cited an anonymous source who put the price close to $90 million, so that figure is a report, not a confirmed transaction value. No packet source supplies cash-versus-stock composition, earnouts, retention terms, founder proceeds, or an acquisition agreement.[9]
The company and product outcomes must be separated. Disqus became a Zeta business unit; Ha said the team, products, terms, pricing, and publisher vision were initially unchanged.[3] Zeta's 2025 privacy policy still identifies Disqus as a separately branded affiliate, while live product pages show that comments, moderation, ads, and paid plans remain available.[18][17] The product remains active, not closed. This is product continuity under new ownership, not a failed-company closure.
The structural change came before the sale. Disqus built reach through a free publisher embed, then had to monetize users who were usually visiting someone else's site. Sponsored comments and publisher revenue sharing were one answer. The December 2016 layoff cut 11 employees, about 20 percent of staff, as the company focused more deeply on reading and discussion data for publisher analytics and advertiser marketing.[7]
Zeta's rationale followed that path: publisher reach plus interest and engagement data for audience targeting. CEO David Steinberg said, "Disqus strengthens Zeta’s ability to offer the best of both worlds."[9] The quote is broad, but the transaction logic was specific. Zeta promised publisher analytics, personalization, messaging, and machine-learning-assisted moderation.
The non-obvious mechanism is that the free embed created both the distribution asset and the governance burden. Each publisher installation expanded the identity and advertising network. It also created another context in which a visitor might be observed, profiled, moderated, or disclosed to an affiliate. Monetization did not sit beside the community product; it changed the custody obligations around the community product.
The July 2012 breach exposed a database snapshot with usernames, email addresses, sign-up dates, and last-login dates for about 17.5 million users. Disqus did not discover and disclose it until October 2017, after an independent researcher surfaced the data.[5][6] The packet does not establish the intrusion method, dwell time, or full remediation record, so it cannot support a broader claim about the company's security posture.
Privacy enforcement after the acquisition shows a related custody problem. Norway's regulator proposed a NOK 25 million fine in a 2021 draft concerning tracking, profiling, advertising disclosures, and legal basis.[11] The final 2024 decision was a reprimand, not that fine. It found invalid consent for disclosure of Norwegian user data to Zeta between July 20, 2018 and December 12, 2019, and said the improper sharing stopped when the discrepancy was discovered.[10]
Current policy language makes the ongoing tradeoff explicit. The service says it collects browsing interactions through cookies even when a visitor does not comment, shares identifiers and behavioral data with advertising partners, and combines data with Zeta's other online and offline information for cross-context advertising.[16] Product survival therefore does not close the retrospective. Ownership changes, monetization changes, and consent semantics must remain traceable for as long as the embed remains live.