100 Questions, Nineteenth Category: Marketing Automation, Where HubSpot Leads a Second Category and Its Own Website Is Cited in Half the Answers

Two brands, and HubSpot leads a second category

HubSpot is named in 64 of 100 answers, recommended in 63 and picked in 57. ActiveCampaign is named in 50, recommended in 49 and picked in 45. Both are picked in 31 answers, and at least one of them in 71. On the 20 direct advice questions, 18 produced a pick: HubSpot in 14, ActiveCampaign in 10, then Klaviyo, Marketo and Salesforce Marketing Cloud in 4 each. Below the two leaders the field splits by buyer: Klaviyo (named 28, picked 25) for ecommerce, Marketo (26 and 23) for enterprise B2B, Brevo (26 and 23) on price, Customer.io (24 and 20) and Braze (14 and 14) for product and mobile teams. HubSpot is the first brand in the series to be the most-named in two categories. In CRM it was named in 75 answers and picked in 69; in email marketing it was named in 37 and picked in 30, behind Mailchimp.

Every head-to-head picks every brand it names

All 30 head-to-head questions picked every brand the question named, usually with a decision rule that gives each brand its own case. Only project management has done that so far. It holds even where an answer leans one way: ActiveCampaign versus Mailchimp ends with a rule of thumb that sends complicated needs to ActiveCampaign and simple emails to Mailchimp, and Salesforce Marketing Cloud or Braze for a large retailer closes with a decision frame that assigns each one a case. Marketo versus Eloqua versus Salesforce Marketing Cloud also picks Pardot, under its new name, for the B2B Salesforce case.

Industry questions add a specialist

Twenty-two questions name an industry or buyer type. At least one brand is picked in 19 of them, and HubSpot in 14. In most, a specialist is picked beside HubSpot or instead of it: Klaviyo in all four ecommerce and Shopify questions and Omnisend in three, HighLevel in both agency questions, Lawmatics in both law firm questions, Follow Up Boss in both real estate questions, Kit in both coaching questions, and Braze and CleverTap in both mobile app questions. Financial advisors go outside the set. One answer is a single sentence announcing a comparison; the other puts Snappy Kraken, which is not in the dictionary, "near the top of your shortlist" and offers HubSpot and ActiveCampaign only as more flexible options for advisors who build their own automations. Healthcare splits: the practice question picks HubSpot and Salesforce Marketing Cloud, while the general healthcare question gives a shortlist with fit notes and asks what kind of organization the buyer is before narrowing it.

Alternative questions are where the model holds back

Nine answers pick no brand. Three of them are the single-sentence answers (financial advisors, a self-hosted alternative to ActiveCampaign, best value for money). Five of the nine are alternative questions: 15 of the 20 alternative questions produced a pick, against all 30 head-to-heads, 28 of 30 category questions and 18 of 20 advice questions. Alternative answers also hold 24 of the 46 recommendations that stop short of a pick. The Pardot and Braze alternative answers list six and eight options with a fit note each, then ask for company size and budget. HubSpot alternatives are answered with CRMs: of the six questions that ask for one, Zoho CRM is named in five answers, Pipedrive in four and Freshsales in three, all sales CRMs outside this dictionary. "Cheaper alternative to HubSpot" shortlists Freshsales, Pipedrive and Zoho, and "HubSpot alternative for B2B SaaS" lists Attio, Salesforce, Pipedrive and Close next to ActiveCampaign and Customer.io, so neither answer has a dictionary pick.

Pardot is picked under its new name

Pardot is named in 14 answers and picked in all 14. Nineteen answers use "Account Engagement", Salesforce's current name for the product, and all 14 answers that say Pardot also give the new name. Eight explain the rename, and seven bring up Marketing Cloud Next. Salesforce Marketing Cloud is named in 18 answers and picked in 15, including two where it appears only as part of a bundle with Salesforce's industry clouds for higher education and healthcare.

HubSpot's site is in half the answers

Across 100 answers there were 301 citation events to 103 domains. Vendors' own sites took 190 of them, 63 percent. hubspot.com is cited in 50 answers, 37 of them to questions that do not mention HubSpot, and HubSpot is picked in 33 of those 37. Seven of the 37 ask for an alternative to Marketo, Pardot, ActiveCampaign, Salesforce Marketing Cloud or Eloqua; six of the seven cite a HubSpot blog post, and HubSpot is picked in six. activecampaign.com is cited in 25 answers, and ActiveCampaign is picked in 24 of them. The most-cited independent pages are single buyer's guides on ciopages.com (5 answers), cmomag.com (4) and taqtics.com (3), all cited on category questions that ask for the best or top-rated tools, and none of the three sites is in Google's top ten for any of the 100 queries. Reddit and Wikipedia have zero citations, for the nineteenth teardown running.

Of the 306 picks, 259 went to brands whose own website did not rank in Google's top ten for that query, 85 percent; only business intelligence, at 87 percent, has been higher so far. HubSpot's site ranked on 11 of the 57 questions where HubSpot was picked, and Brevo's on none of its 23. Google's top ten is 86 percent third-party pages, and 56 of the 301 citation events (19 percent) involved a domain that also ranked for the question.

Your blog answers your competitors' questions. HubSpot blog posts about Marketo and Pardot alternatives were cited on questions that never mention HubSpot, and HubSpot was picked in six of the seven answers about a competitor's alternatives that cite hubspot.com. If you publish the most useful page about switching away from a rival, the model reads it when buyers ask exactly that.

Say your current name, and your old one. Pardot was picked in all 14 answers that named it, and every answer that said Pardot also gave its new name. When a product is renamed, pages that state both names plainly let the model connect them and keep the product in the answer.

Industry buyers get a specialist. Ecommerce went to Klaviyo, agencies to HighLevel, law firms to Lawmatics, real estate to Follow Up Boss, coaches to Kit, usually next to HubSpot. A general platform that wants those buyers needs pages that speak to each one directly, or the specialist takes the pick alone.

Ranking is not the entry ticket. 85 percent of picks went to brands whose own site did not rank for the question. What the model reads, mostly vendor pages and a handful of buyer's guides, decides the answer more than who holds the top ten.

One run per query means answer variance is unmeasured; Benchmark 01 runs each query three times across three surfaces. The teardowns are collected on different days, so differences between categories mix the category with the date. The brand dictionary covers the general marketing automation software market and deliberately excludes vertical tools, so their appearances are described in prose and not counted. The readers are AI models, not people: they follow a written protocol and agree with each other closely, but a shared blind spot would not show up as disagreement. The line between picked and recommended is a judgment, documented in the protocol with examples. Sentiment is not coded. No vendor-tool cross-check was read for this category. Google's control counts a brand as ranking only when its own domain is in the top ten; a listicle that features the brand does not count. One run per question, one day, United States, English, logged out. Answers vary between runs, so these are frequencies for this sample. The coding is by AI readers: two agreed on 96.9 percent of brand codes and a third settled the rest, and a different reader could place some brands differently between picked and recommended. Three of the 100 answers are single sentences with no answer body and count as answered with no pick. The dictionary includes four vertical tools (HighLevel, Lawmatics, Follow Up Boss, FMG Suite) but not sales CRMs such as Zoho CRM and Pipedrive or specialists such as Snappy Kraken and MoxiWorks, so HubSpot-alternative and financial advisor results undercount them. Vendors named only as a link to salesforce.com or business.adobe.com are not credited to a product. The review-and-media list is the codebook's fixed list, so buyer's guides such as ciopages.com count as other third-party pages.

Every number above can be recomputed from these files. CC BY 4.0: use them, cite the page.

• queries.csv: the 100 queries with intent labels.
• brands.csv: the 53-brand dictionary with aliases and canonical domains.
• mentions.csv: 386 coded brand mentions with position, type, a 0/1 picked column and whether the brand's domain was in Google's top ten.
• citations.csv: 301 citation events with domain class and Google overlap.
• observations.csv: one row per query with brand, recommendation, citation and control counts.
• dictionary-changes.md: the one dictionary change made after collection, before the final coding run, and the known alias overlaps
• brands-original.csv: the dictionary as first drafted, before that change
• coder-notes.md: how the answers were coded, reader agreement, the judgment calls that matter in this category, and dictionary notes
• codebook.md: the rulebook, with dated amendments through v2.0.
• reader-protocol.md: the written protocol both readers coded against (codebook v2.0).
• picked_stats.json: the picked-level figures.

The earlier teardowns: Teardown 01, project management, Teardown 02, crm, Teardown 03, email marketing, Teardown 04, help desk, Teardown 05, accounting, Teardown 06, payment processing, Teardown 07, payroll, Teardown 08, hr software, Teardown 09, applicant tracking, Teardown 10, password managers, Teardown 11, endpoint security, Teardown 12, business intelligence, Teardown 13, data warehouse and etl, Teardown 14, ecommerce platforms, Teardown 15, website builders, Teardown 16, scheduling, Teardown 17, video conferencing, Teardown 18, e-signature.

In marketing automation the model's picks track the vendors' own websites: vendor pages are 63 percent of what it cites, and one vendor's blog is read for questions about its competitors. Finding out what the model is reading for your category is the first thing a Category Salience Brief does, with a query set you approve first.