AnswerThis growth playbook

Pick the current plan set and make every block on the page agree with it: cards, the "Included in Max" section, the comparison table, the FAQ and the Academia / Industry switch. Remove the duplicate AI Writer line on the Pro card. Then check that the in-app upgrade screen, checkout and any ad copy quote the same prices, and add a short "which plan do I need?" row (thesis, lab review, regulated SLR) above the table.

Price confusion is one of the cheapest conversion leaks to fix: visitors who doubt the price leave or pick a plan that misses the feature they came for, and both show up later as refunds and support tickets. It also stops AI answers from quoting the wrong price.

[ ] Plan cards match billing
[ ] "Included in Max" matches the cards
[ ] Comparison table prices and features match
[ ] One AI Writer line per plan
[ ] FAQ answers match
[ ] Academia and Industry switch both checked
[ ] Upgrade screen and checkout match
[ ] Ads and feature pages quoting prices updated

• Changing the table shifts plan mix

  Watch average revenue per new subscriber for a month.

• Academic and industry prices differ on purpose

  Label each set clearly under its own switch.

• Your websitePricing page plan cards (checked October 22, 2026): Free $0, Pro $35 a month per seat ("Every advanced capability in one plan: systematic review, meta-analysis, Trial Landscape and all connectors"), Lab & Team $75 a month (annual, minimum 3 seats) and Max $250 a month ("Site licenses for institutions & libraries"). The Pro card lists both "AI Writer with full manuscript drafting" and "AI Writer — standard models". An Academia / Industry ("HEOR · pharma · CROs") switch sits above the cards.Source
• Your websiteThe same pricing page, further down: a section headed "Included in Max" says "Systematic review, meta-analysis, Trial Landscape and every regulatory connector are all part of one plan, Max"; the comparison table "What is included at every tier" lists Free $0, Pro $14/month, Research $30/month and Max $100/month, with the systematic review pipeline only in Max. The page therefore shows two price lists and two answers to which plan includes systematic review.Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source

Agree one current figure for papers, researchers, institutions and projects, with the date and how it is counted. Update the homepage, meta descriptions, plan cards, researcher profiles, comparison and enterprise pages, and fix the label and spelling slips (including "Plagerism" and the 01, 02, 02, 02 numbering).

Researchers and procurement teams notice small errors, and AI assistants repeat whichever number they find first. Consistent numbers make the verification promise believable.

• Bigger numbers look better

  Use the defensible one and say what it counts.

• Your websiteHomepage: H1 "Scientific AI For All Verified Research Workflows"; body copy says "300M+ papers" while the meta description (the text search results show) and the Free plan say "250M+ papers"; "Trusted by over 1 Million+ researchers", "600,000+ Personal Libraries Created" and a line reading "Trusted by 200,+ institutions and individuals" next to "200,000+"; the customer logo labels read "Standford university" and "McKmsey & Company"; the AI Writer card says "detect AI or Plagerism"; the five workflow stages are numbered 01, 02, 02, 02, 04, 05.Source
• Your websiteResearcher profile for Douglas G. Altman (one of 1,006): title and description are specific ("1634 publications and 1100715 citations") but the page says "Institution information unavailable", the "Research Areas" block is empty, the social preview uses the generic "AnswerThis is the world's most powerful AI for research", and the sign-up prompt says "250+ million papers".Source
• Your websiteThe Elicit comparison page puts AnswerThis at "300M+ papers and clinical trials" against Elicit's "138M+ papers and 545,000 clinical trials", and says AnswerThis writes a "Full review with a citation on every line, built from 150+ sources".Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source
• HiringFounding Engineering Lead (San Francisco, posted September 2, 2026): AnswerThis is "building the research workspace for scientific knowledge inside enterprises"; "More than 200,000 researchers use the platform across universities and Fortune 500 companies"; the work "decides whether new drugs and medical devices reach patients"; the lead will "work directly with enterprise customers" across "Next.js, Python/FastAPI, retrieval, agents".Source

Build three fast landing pages, one per intent (literature review, systematic review, citation check), each under 1 MB with one headline, one proof block and a sign-up button. Compress and lazy-load images and video, load tags after the first paint, and send paid traffic there instead of the homepage. Compare cost per activated sign-up with the homepage.

Faster pages convert more paid clicks and lift ad quality scores, so the same budget buys more sign-ups as spend grows.

• Tracking gaps from deferred tags

  Check conversions in each ad platform before switching all traffic.

• Google PageSpeedHomepage on mobile (October 22, 2026): performance 54/100 (desktop 78); largest content appears after 15.1s in the lab test (target 2.5s); the page downloads about 10.6 MB in 115 requests, 9.3 MB of it from the site builder's CDN; Google Tag Manager blocks the main thread for 420 ms. Real Chrome visitors: LCP 2.6s and INP 215 ms at the 75th percentile, so the site fails Core Web Vitals on both, and LCP got 18% slower over the recent history.Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source
• Tech checkThe homepage loads about 20 third-party services, including Google Tag Manager, Google Analytics, Google Ads, Meta, TikTok, Reddit, X conversion tracking, Bing Ads, Microsoft Clarity and FirstPromoter (affiliate tracking); third parties account for 10.5 MB of the 10.6 MB page.Source

List every tag with its owner and the budget it serves. Remove tags for platforms with no current spend, move conversion events to server-side sending (Meta Conversions API, Google enhanced conversions) where possible, and load the rest after the page is usable. Document who may add a tag.

Fewer, better-owned tags speed the page and give cleaner conversion data, which the new Growth Lead needs before scaling spend.

• Losing attribution

  Run old and new conversion sending side by side for two weeks.

• Tech checkThe homepage loads about 20 third-party services, including Google Tag Manager, Google Analytics, Google Ads, Meta, TikTok, Reddit, X conversion tracking, Bing Ads, Microsoft Clarity and FirstPromoter (affiliate tracking); third parties account for 10.5 MB of the 10.6 MB page.Source
• Google PageSpeedHomepage on mobile (October 22, 2026): performance 54/100 (desktop 78); largest content appears after 15.1s in the lab test (target 2.5s); the page downloads about 10.6 MB in 115 requests, 9.3 MB of it from the site builder's CDN; Google Tag Manager blocks the main thread for 420 ms. Real Chrome visitors: LCP 2.6s and INP 215 ms at the 75th percentile, so the site fails Core Web Vitals on both, and LCP got 18% slower over the recent history.Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source

Keep only shared answers that pass a bar (a real research question, 10+ sources, no duplicates). Rewrite their titles as clean questions, show the user's name only if they chose to, group them by field and language, and mark the rest noindex. Link each kept answer to "Run this question yourself".

A smaller library of strong, cited answers earns more search and AI citations than hundreds of raw prompts, and protects users who shared without expecting to be named.

• Losing long-tail traffic

  Noindex only pages with no search clicks in 90 days.

• Your websiteThe "Shared Questions" library (/questions, described as "the most commonly searched academic and research questions with comprehensive answers"): entries are raw user prompts shown with the user's display name, many in other languages (Indonesian, Arabic, Spanish, Portuguese, French, Thai), some duplicated, some instructions rather than questions ("Make me an research paper...", formatting and slide requests), some with a single source, and one asking to search with a rival product.Source
• Your websiteSitemap (1,506 URLs, October 22, 2026): 1,006 researcher profiles, 220 shared answers, 83 blog posts, 45 feature pages, about 100 free tools (sample-size and effect-size calculators, PRISMA/CONSORT checklists, 30+ citation-style generators, a retraction checker, DOI and BibTeX converters, regulatory-document search) and five comparison pages: Elicit, Jenni AI, SciSpace, Paperpal and Consensus. ResearchRabbit is covered only in a blog post; there are no pages for Undermind or for general assistants such as ChatGPT deep research.Source
• Your websiterobots.txt names OAI-SearchBot, ChatGPT-User, PerplexityBot, Perplexity-User, ClaudeBot and GPTBot and allows them, blocking only /api/ and /s/.Source

Write 40 prompts researchers actually ask (by task, field and plan budget), run them weekly in ChatGPT, Perplexity, Google AI answers and Claude, and record which tools are named, which pages are cited and what is said about AnswerThis. Fix the pages behind wrong answers first.

You cannot improve AI visibility you have not measured, and the baseline tells the new hire which pages to build first.

• Answers vary run to run

  Run each prompt three times and use the share.

• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source
• Your websiterobots.txt names OAI-SearchBot, ChatGPT-User, PerplexityBot, Perplexity-User, ClaudeBot and GPTBot and allows them, blocking only /api/ and /s/.Source
• Your websiteMCP page: "Your research, inside your AI assistant"; connect AnswerThis to a compatible assistant over the Model Context Protocol to search papers, use the library and save cited Writer drafts; "Public paper search is available without connecting an AnswerThis account".Source
• Competitor researchRivals' own homepages: Elicit, "AI for Scientific Research", "Trusted by over 7 million researchers and 8 of the top 20 pharmaceutical companies"; Consensus, "an AI academic search engine for peer-reviewed literature"; Undermind, "your AI co-researcher for the literature" that "finds the papers others miss". AnswerThis's own Elicit comparison credits Elicit with "Published accuracy benchmarks for its pipeline".See the research Source

Add pages for Undermind, ResearchRabbit (move the blog post), "ChatGPT deep research for literature reviews" and "Perplexity for academic research", using the same fair table as the Elicit page: sources searched, how claims are checked, export, systematic review support and price. Date each page and cite the rival's own site.

Comparison searches are the highest-intent organic traffic, and fair tables are what AI assistants quote when asked which tool to use.

• Unfair claims about rivals

  Cite their own pages and date every claim.

• Your websiteSitemap (1,506 URLs, October 22, 2026): 1,006 researcher profiles, 220 shared answers, 83 blog posts, 45 feature pages, about 100 free tools (sample-size and effect-size calculators, PRISMA/CONSORT checklists, 30+ citation-style generators, a retraction checker, DOI and BibTeX converters, regulatory-document search) and five comparison pages: Elicit, Jenni AI, SciSpace, Paperpal and Consensus. ResearchRabbit is covered only in a blog post; there are no pages for Undermind or for general assistants such as ChatGPT deep research.Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source
• Competitor researchRivals' own homepages: Elicit, "AI for Scientific Research", "Trusted by over 7 million researchers and 8 of the top 20 pharmaceutical companies"; Consensus, "an AI academic search engine for peer-reviewed literature"; Undermind, "your AI co-researcher for the literature" that "finds the papers others miss". AnswerThis's own Elicit comparison credits Elicit with "Published accuracy benchmarks for its pipeline".See the research Source
• Your websiteThe Elicit comparison page puts AnswerThis at "300M+ papers and clinical trials" against Elicit's "138M+ papers and 545,000 clinical trials", and says AnswerThis writes a "Full review with a citation on every line, built from 150+ sources".Source

Define activation (for example, a first cited literature review saved to the library), send activation and first payment back to Google, Meta and LinkedIn as offline conversions, and report cost per paying researcher by channel, country and audience. Shift budget weekly.

Ad platforms optimize toward the event you feed them; feeding them payers instead of sign-ups changes who they find.

• Fewer conversions slow learning

  Use activation as the signal and payment as the check.

• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source
• Tech checkThe homepage loads about 20 third-party services, including Google Tag Manager, Google Analytics, Google Ads, Meta, TikTok, Reddit, X conversion tracking, Bing Ads, Microsoft Clarity and FirstPromoter (affiliate tracking); third parties account for 10.5 MB of the 10.6 MB page.Source
• Your websitePricing page plan cards (checked October 22, 2026): Free $0, Pro $35 a month per seat ("Every advanced capability in one plan: systematic review, meta-analysis, Trial Landscape and all connectors"), Lab & Team $75 a month (annual, minimum 3 seats) and Max $250 a month ("Site licenses for institutions & libraries"). The Pro card lists both "AI Writer with full manuscript drafting" and "AI Writer — standard models". An Academia / Industry ("HEOR · pharma · CROs") switch sits above the cards.Source

Add one next step to each tool that carries the result into AnswerThis: save the citation to a library, run a literature search on the calculated topic, or screen papers against the checklist. Track tool-to-sign-up by tool and improve the bottom ten.

People using a tool are already doing the work AnswerThis automates; a relevant hand-off converts better than a generic banner.

• Tool users resent sign-up walls

  Keep the tool free and make the hand-off optional.

• Your websiteSitemap (1,506 URLs, October 22, 2026): 1,006 researcher profiles, 220 shared answers, 83 blog posts, 45 feature pages, about 100 free tools (sample-size and effect-size calculators, PRISMA/CONSORT checklists, 30+ citation-style generators, a retraction checker, DOI and BibTeX converters, regulatory-document search) and five comparison pages: Elicit, Jenni AI, SciSpace, Paperpal and Consensus. ResearchRabbit is covered only in a blog post; there are no pages for Undermind or for general assistants such as ChatGPT deep research.Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source

Build an industry landing page per use (HEOR evidence reviews, medical affairs, regulatory literature searches), run LinkedIn campaigns to those titles with a demo or pilot offer, and route leads to sales. Measure demos and pilots separately from academic sign-ups.

Industry seats are worth many student plans, and a separate funnel keeps cheap academic clicks from hiding them.

• LinkedIn costs per click are high

  Judge on cost per qualified demo.

• Your websitePricing page plan cards (checked October 22, 2026): Free $0, Pro $35 a month per seat ("Every advanced capability in one plan: systematic review, meta-analysis, Trial Landscape and all connectors"), Lab & Team $75 a month (annual, minimum 3 seats) and Max $250 a month ("Site licenses for institutions & libraries"). The Pro card lists both "AI Writer with full manuscript drafting" and "AI Writer — standard models". An Academia / Industry ("HEOR · pharma · CROs") switch sits above the cards.Source
• Your websiteEnterprise page: "AI Systematic Reviews for Life Sciences" for "pharma, biotech, and medical device teams", "50+ Enterprise teams trust AnswerThis globally", "95% Reduction in screening time", "Every output meets FDA, EMA, and HTA requirements out of the box", connectors to "PubMed, Embase, Cochrane", "REST API and SSO", with "Book a demo" and "Contact sales"; no named enterprise case studies.Source
• HiringFounding Engineering Lead (San Francisco, posted September 2, 2026): AnswerThis is "building the research workspace for scientific knowledge inside enterprises"; "More than 200,000 researchers use the platform across universities and Fortune 500 companies"; the work "decides whether new drugs and medical devices reach patients"; the lead will "work directly with enterprise customers" across "Next.js, Python/FastAPI, retrieval, agents".Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source

Publish two or three customer stories (named or anonymized with permission) with measured time saved, and a methods page that shows how screening, extraction and grading are logged and how outputs map to PRISMA and HTA submission needs. Word the requirements claim to match what the method shows, after a regulatory review.

Evidence is what this audience buys; a documented method shortens security and quality reviews and makes the claim defensible.

• Customers decline to be named

  Anonymize with their approval.

• Your websiteEnterprise page: "AI Systematic Reviews for Life Sciences" for "pharma, biotech, and medical device teams", "50+ Enterprise teams trust AnswerThis globally", "95% Reduction in screening time", "Every output meets FDA, EMA, and HTA requirements out of the box", connectors to "PubMed, Embase, Cochrane", "REST API and SSO", with "Book a demo" and "Contact sales"; no named enterprise case studies.Source
• HiringFounding Engineering Lead (San Francisco, posted September 2, 2026): AnswerThis is "building the research workspace for scientific knowledge inside enterprises"; "More than 200,000 researchers use the platform across universities and Fortune 500 companies"; the work "decides whether new drugs and medical devices reach patients"; the lead will "work directly with enterprise customers" across "Next.js, Python/FastAPI, retrieval, agents".Source
• Competitor researchRivals' own homepages: Elicit, "AI for Scientific Research", "Trusted by over 7 million researchers and 8 of the top 20 pharmaceutical companies"; Consensus, "an AI academic search engine for peer-reviewed literature"; Undermind, "your AI co-researcher for the literature" that "finds the papers others miss". AnswerThis's own Elicit comparison credits Elicit with "Published accuracy benchmarks for its pipeline".See the research Source

Fix the caption, add a section for supervisors, lab managers and librarians with a lab discount for their group, and ask happy paid users who run labs to join. Track referred seats by referrer type.

Researchers choose tools on their supervisor's or librarian's word, and one referral can bring a whole group.

• Self-referral

  Pay only on new paying accounts.

• Your websiteAffiliate page: "earn 20% referral bonus for every person you refer, each payment period", "Perfect for educators, influencers, and research communities"; influencers with 10,000+ followers are asked to email the outreach team; the big "20 %" figure is captioned "Research papers"; a header link says "Refer and earn".Source
• Your websitePricing page plan cards (checked October 22, 2026): Free $0, Pro $35 a month per seat ("Every advanced capability in one plan: systematic review, meta-analysis, Trial Landscape and all connectors"), Lab & Team $75 a month (annual, minimum 3 seats) and Max $250 a month ("Site licenses for institutions & libraries"). The Pro card lists both "AI Writer with full manuscript drafting" and "AI Writer — standard models". An Academia / Industry ("HEOR · pharma · CROs") switch sits above the cards.Source
• Your websitePricing page testimonials come from a PhD student, a medical doctor and researchers worldwide ("compiling an updated literature review in minutes"; "during my PhD thesis preparation"); the homepage logo row shows MIT, Penn, Stanford, Oxford, NASA and McKinsey.Source

Build a calendar of thesis and term deadlines for the top ten countries by sign-ups, raise budgets and run deadline-specific creative two to four weeks before each, and pull back between peaks.

Spending into demand peaks lowers acquisition cost without new channels.

• Rivals bid up the same peaks

  Start earlier and target long-tail terms.

• Your websitePricing page testimonials come from a PhD student, a medical doctor and researchers worldwide ("compiling an updated literature review in minutes"; "during my PhD thesis preparation"); the homepage logo row shows MIT, Penn, Stanford, Oxford, NASA and McKinsey.Source
• Your websiteThe "Shared Questions" library (/questions, described as "the most commonly searched academic and research questions with comprehensive answers"): entries are raw user prompts shown with the user's display name, many in other languages (Indonesian, Arabic, Spanish, Portuguese, French, Thai), some duplicated, some instructions rather than questions ("Make me an research paper...", formatting and slide requests), some with a single source, and one asking to search with a rival product.Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source

Submit the MCP server to the connector directories of the major assistants and MCP registries, write one setup page per assistant, and add a "connected from" source so sign-ups from assistants can be counted.

Researchers who already work in an assistant can adopt AnswerThis without leaving it, and early directory listings get reviews and ranking before rivals.

• Directory reviews take time

  Submit early and in parallel.

• Your websiteMCP page: "Your research, inside your AI assistant"; connect AnswerThis to a compatible assistant over the Model Context Protocol to search papers, use the library and save cited Writer drafts; "Public paper search is available without connecting an AnswerThis account".Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source
• Your websiterobots.txt names OAI-SearchBot, ChatGPT-User, PerplexityBot, Perplexity-User, ClaudeBot and GPTBot and allows them, blocking only /api/ and /s/.Source

Count paying users by email domain, pick the ten universities with the most, and offer their library a pilot license with usage reports. Add AnswerThis to library research guides and run librarian webinars.

Usage data from their own students is the strongest argument a librarian can take to a budget meeting.

• Long budget cycles

  Time pilots to the next budget year.

• Your websitePricing page plan cards (checked October 22, 2026): Free $0, Pro $35 a month per seat ("Every advanced capability in one plan: systematic review, meta-analysis, Trial Landscape and all connectors"), Lab & Team $75 a month (annual, minimum 3 seats) and Max $250 a month ("Site licenses for institutions & libraries"). The Pro card lists both "AI Writer with full manuscript drafting" and "AI Writer — standard models". An Academia / Industry ("HEOR · pharma · CROs") switch sits above the cards.Source
• Your websitePricing page testimonials come from a PhD student, a medical doctor and researchers worldwide ("compiling an updated literature review in minutes"; "during my PhD thesis preparation"); the homepage logo row shows MIT, Penn, Stanford, Oxford, NASA and McKinsey.Source
• HiringFounding Engineering Lead (San Francisco, posted September 2, 2026): AnswerThis is "building the research workspace for scientific knowledge inside enterprises"; "More than 200,000 researchers use the platform across universities and Fortune 500 companies"; the work "decides whether new drugs and medical devices reach patients"; the lead will "work directly with enterprise customers" across "Next.js, Python/FastAPI, retrieval, agents".Source

Build a benchmark of real research questions with expert-checked answers, measure citation accuracy, claim support and recall, publish the method and results, and rerun each quarter. Invite rivals' results for the same set.

Researchers and regulated buyers need numbers, and a public benchmark is cited by journalists, reviewers and AI assistants.

• Results look weaker than marketing

  Publish honestly and show the trend.

• Competitor researchRivals' own homepages: Elicit, "AI for Scientific Research", "Trusted by over 7 million researchers and 8 of the top 20 pharmaceutical companies"; Consensus, "an AI academic search engine for peer-reviewed literature"; Undermind, "your AI co-researcher for the literature" that "finds the papers others miss". AnswerThis's own Elicit comparison credits Elicit with "Published accuracy benchmarks for its pipeline".See the research Source
• Your websiteThe Elicit comparison page puts AnswerThis at "300M+ papers and clinical trials" against Elicit's "138M+ papers and 545,000 clinical trials", and says AnswerThis writes a "Full review with a citation on every line, built from 150+ sources".Source
• Your websiteHomepage: H1 "Scientific AI For All Verified Research Workflows"; body copy says "300M+ papers" while the meta description (the text search results show) and the Free plan say "250M+ papers"; "Trusted by over 1 Million+ researchers", "600,000+ Personal Libraries Created" and a line reading "Trusted by 200,+ institutions and individuals" next to "200,000+"; the customer logo labels read "Standford university" and "McKmsey & Company"; the AI Writer card says "detect AI or Plagerism"; the five workflow stages are numbered 01, 02, 02, 02, 04, 05.Source
• HiringFounding Engineering Lead (San Francisco, posted September 2, 2026): AnswerThis is "building the research workspace for scientific knowledge inside enterprises"; "More than 200,000 researchers use the platform across universities and Fortune 500 companies"; the work "decides whether new drugs and medical devices reach patients"; the lead will "work directly with enterprise customers" across "Next.js, Python/FastAPI, retrieval, agents".Source

Fill institution, areas and recent work for every profile, give each a specific social preview, then add "Is this you? Claim your profile" so researchers can add details, follow citations to their work and get alerts. Expand to more researchers only when claimed profiles show engagement.

Complete profiles rank better, and a claimed profile turns a famous name into a user who shares their page.

• Wrong data on a profile

  Let researchers correct it and show the source.

• Your websiteResearcher profile for Douglas G. Altman (one of 1,006): title and description are specific ("1634 publications and 1100715 citations") but the page says "Institution information unavailable", the "Research Areas" block is empty, the social preview uses the generic "AnswerThis is the world's most powerful AI for research", and the sign-up prompt says "250+ million papers".Source
• Your websiteSitemap (1,506 URLs, October 22, 2026): 1,006 researcher profiles, 220 shared answers, 83 blog posts, 45 feature pages, about 100 free tools (sample-size and effect-size calculators, PRISMA/CONSORT checklists, 30+ citation-style generators, a retraction checker, DOI and BibTeX converters, regulatory-document search) and five comparison pages: Elicit, Jenni AI, SciSpace, Paperpal and Consensus. ResearchRabbit is covered only in a blog post; there are no pages for Undermind or for general assistants such as ChatGPT deep research.Source
• Your websiteHomepage: H1 "Scientific AI For All Verified Research Workflows"; body copy says "300M+ papers" while the meta description (the text search results show) and the Free plan say "250M+ papers"; "Trusted by over 1 Million+ researchers", "600,000+ Personal Libraries Created" and a line reading "Trusted by 200,+ institutions and individuals" next to "200,000+"; the customer logo labels read "Standford university" and "McKmsey & Company"; the AI Writer card says "detect AI or Plagerism"; the five workflow stages are numbered 01, 02, 02, 02, 04, 05.Source

Use sign-up data to pick the top three languages, localize the homepage, pricing, top tools and the best shared answers, set hreflang, and run local ads with local prices where purchasing power differs.

Being the only research assistant with pages in a language wins the search and AI answers in that language.

• Low revenue per user

  Judge on payers, not sign-ups.

• Your websiteThe "Shared Questions" library (/questions, described as "the most commonly searched academic and research questions with comprehensive answers"): entries are raw user prompts shown with the user's display name, many in other languages (Indonesian, Arabic, Spanish, Portuguese, French, Thai), some duplicated, some instructions rather than questions ("Make me an research paper...", formatting and slide requests), some with a single source, and one asking to search with a rival product.Source
• Your websiteSitemap (1,506 URLs, October 22, 2026): 1,006 researcher profiles, 220 shared answers, 83 blog posts, 45 feature pages, about 100 free tools (sample-size and effect-size calculators, PRISMA/CONSORT checklists, 30+ citation-style generators, a retraction checker, DOI and BibTeX converters, regulatory-document search) and five comparison pages: Elicit, Jenni AI, SciSpace, Paperpal and Consensus. ResearchRabbit is covered only in a blog post; there are no pages for Undermind or for general assistants such as ChatGPT deep research.Source
• HiringGrowth Lead | Paid Acquisition + SEO/AEO (posted September 2, 2026, US, SF preferred, $150,000 base + commission + equity): "AnswerThis has helped over 600,000 researchers"; wants someone who has "personally managed more than $500,000/month in ad spend" across "Google, Meta, LinkedIn"; will "Own creative tests, targeting, landing pages, tracking and budget decisions", "Ship useful content, comparison pages, tools and technical SEO improvements", "Build AEO early" across "ChatGPT, Perplexity, Google AI answers", and "Follow people from first visit through signup, activation and paid conversion". The team is "resource-constrained".Source

Pick five topics with high search interest, run living reviews, publish them as dated pages with methods and update logs, and invite domain experts to review them.

Always-current, cited reviews earn links and AI citations and demonstrate the paid feature in public.

• Medical topics carry risk

  Expert review and clear disclaimers.

• Your websitePricing page plan cards (checked October 22, 2026): Free $0, Pro $35 a month per seat ("Every advanced capability in one plan: systematic review, meta-analysis, Trial Landscape and all connectors"), Lab & Team $75 a month (annual, minimum 3 seats) and Max $250 a month ("Site licenses for institutions & libraries"). The Pro card lists both "AI Writer with full manuscript drafting" and "AI Writer — standard models". An Academia / Industry ("HEOR · pharma · CROs") switch sits above the cards.Source
• Your websiteThe same pricing page, further down: a section headed "Included in Max" says "Systematic review, meta-analysis, Trial Landscape and every regulatory connector are all part of one plan, Max"; the comparison table "What is included at every tier" lists Free $0, Pro $14/month, Research $30/month and Max $100/month, with the systematic review pipeline only in Max. The page therefore shows two price lists and two answers to which plan includes systematic review.Source
• Your websiteThe Elicit comparison page puts AnswerThis at "300M+ papers and clinical trials" against Elicit's "138M+ papers and 545,000 clinical trials", and says AnswerThis writes a "Full review with a citation on every line, built from 150+ sources".Source
• Your websiterobots.txt names OAI-SearchBot, ChatGPT-User, PerplexityBot, Perplexity-User, ClaudeBot and GPTBot and allows them, blocking only /api/ and /s/.Source

Check every library weekly against retraction data and email the user when a saved or cited paper is retracted or corrected, with a link to replacements. Offer it free and highlight it on the paid plans.

A useful alert brings users back without marketing and fits the verification promise.

• False matches

  Match by DOI only.

• Your websiteSitemap (1,506 URLs, October 22, 2026): 1,006 researcher profiles, 220 shared answers, 83 blog posts, 45 feature pages, about 100 free tools (sample-size and effect-size calculators, PRISMA/CONSORT checklists, 30+ citation-style generators, a retraction checker, DOI and BibTeX converters, regulatory-document search) and five comparison pages: Elicit, Jenni AI, SciSpace, Paperpal and Consensus. ResearchRabbit is covered only in a blog post; there are no pages for Undermind or for general assistants such as ChatGPT deep research.Source
• Your websiteHomepage: H1 "Scientific AI For All Verified Research Workflows"; body copy says "300M+ papers" while the meta description (the text search results show) and the Free plan say "250M+ papers"; "Trusted by over 1 Million+ researchers", "600,000+ Personal Libraries Created" and a line reading "Trusted by 200,+ institutions and individuals" next to "200,000+"; the customer logo labels read "Standford university" and "McKmsey & Company"; the AI Writer card says "detect AI or Plagerism"; the five workflow stages are numbered 01, 02, 02, 02, 04, 05.Source

Add "Generate AI-use statement" to each project: which steps used AnswerThis, which sources were verified, and a citation format. Users save time and each paper mentions AnswerThis.

A compliance helper becomes word of mouth inside the papers themselves.

• Policies differ by journal

  Make the statement editable.

• HiringFounding Engineering Lead (San Francisco, posted September 2, 2026): AnswerThis is "building the research workspace for scientific knowledge inside enterprises"; "More than 200,000 researchers use the platform across universities and Fortune 500 companies"; the work "decides whether new drugs and medical devices reach patients"; the lead will "work directly with enterprise customers" across "Next.js, Python/FastAPI, retrieval, agents".Source
• Your websiteEnterprise page: "AI Systematic Reviews for Life Sciences" for "pharma, biotech, and medical device teams", "50+ Enterprise teams trust AnswerThis globally", "95% Reduction in screening time", "Every output meets FDA, EMA, and HTA requirements out of the box", connectors to "PubMed, Embase, Cochrane", "REST API and SSO", with "Book a demo" and "Contact sales"; no named enterprise case studies.Source
• Your websiteHomepage: H1 "Scientific AI For All Verified Research Workflows"; body copy says "300M+ papers" while the meta description (the text search results show) and the Free plan say "250M+ papers"; "Trusted by over 1 Million+ researchers", "600,000+ Personal Libraries Created" and a line reading "Trusted by 200,+ institutions and individuals" next to "200,000+"; the customer logo labels read "Standford university" and "McKmsey & Company"; the AI Writer card says "detect AI or Plagerism"; the five workflow stages are numbered 01, 02, 02, 02, 04, 05.Source

Publish a free, limited Trial Landscape page per indication (top 50 indications), refreshed weekly, with a "full landscape and evidence review" demo offer.

A useful free view brings industry buyers in on their own task and qualifies them by indication.

• Giving away the paid feature

  Show phase counts, keep endpoints and exports paid.

• Your websitePricing page plan cards (checked October 22, 2026): Free $0, Pro $35 a month per seat ("Every advanced capability in one plan: systematic review, meta-analysis, Trial Landscape and all connectors"), Lab & Team $75 a month (annual, minimum 3 seats) and Max $250 a month ("Site licenses for institutions & libraries"). The Pro card lists both "AI Writer with full manuscript drafting" and "AI Writer — standard models". An Academia / Industry ("HEOR · pharma · CROs") switch sits above the cards.Source
• Your websiteEnterprise page: "AI Systematic Reviews for Life Sciences" for "pharma, biotech, and medical device teams", "50+ Enterprise teams trust AnswerThis globally", "95% Reduction in screening time", "Every output meets FDA, EMA, and HTA requirements out of the box", connectors to "PubMed, Embase, Cochrane", "REST API and SSO", with "Book a demo" and "Contact sales"; no named enterprise case studies.Source
• HiringFounding Engineering Lead (San Francisco, posted September 2, 2026): AnswerThis is "building the research workspace for scientific knowledge inside enterprises"; "More than 200,000 researchers use the platform across universities and Fortune 500 companies"; the work "decides whether new drugs and medical devices reach patients"; the lead will "work directly with enterprise customers" across "Next.js, Python/FastAPI, retrieval, agents".Source