Blog
Guides, growth case studies, and marketing playbooks from the Kanso team.
A first-session checklist for Kanso: add your website and profiles, confirm what Kanso read, connect your own Claude and Search Console, pick a first Today card and approve a first draft. Includes a table of what each connection adds and which specialists use it.
Shreyas M VAn AI growth OS is one workspace where specialist AI agents prepare marketing work from one shared view of your business, and a person approves each piece. A plain definition, five tests to check any tool, a comparison with a hire, an agency, a chatbot and a spreadsheet, when it is the wrong choice and an example week built from Kanso's documented features.
Shreyas M VHow Kanso's approval workflow works: a draft is edited, approved by the workspace owner, previewed and confirmed in a separate step before anything goes public. Covers the preview, the owner-only actions, why nothing posts twice, what your own Claude can and cannot do, how code changes stay small and how to disconnect an account.
Shreyas M VHow Kanso works, screen by screen: it reads your website and profiles, specialists and your own Claude prepare drafts and fixes, and you approve each piece. Covers setup, the first-day workspace, which apps to connect first, the Today list, the Content page, the preview-then-confirm publish step and what verified means.
Shreyas M VKanso AI is an AI growth workspace: it reads your website and profiles, writes down your business, and eleven specialist agents prepare search fixes, articles, posts, replies, reel plans and competitor notes for your approval. Includes the setup steps, the eleven specialists, a table of what Kanso does not do and a comparison with a hire, an agency and a generic chatbot.
Shreyas M VWhat AI visibility tools measure, what they cannot, and seven questions to put to any vendor. Includes a table of price and engine disagreements between the pages read, a cost-per-prompt check with the arithmetic shown, and a note on whose list you are reading. Kanso tested no tool.
Shreyas M VHubSpot's Website Grader read from HubSpot's own 2010-era post: a dated timeline, the arithmetic behind its growth figures, how its scoring description changed, and a six-question test for free tools. No source shows the leads or customers it produced.
Shreyas M VA 180-minute weekly budget for founder-led content, split across capture, picking, drafting, publishing, engaging and review, with a stop rule for each block and 120, 90 and 15-minute versions for bad weeks. The time figures come from three guides; the budget is Kanso's own design and untested.
Shreyas M V37signals' founders say they build demand by publishing opinions, products and books under their own names. This page reads their podcast and profit essay, grades five channels by the evidence behind each, and adds a five-test check for your own writing. No source gives traffic, sales or book figures.
Shreyas M VA founder's guide to building in public with a risk checklist for customers and competitors, a measurement plan and an eight-week test with stop rules. The sources report no measured results, so the checklist and plan are Kanso's analysis and the worked example uses invented numbers.
Shreyas M VGoogle's rules on AI content since February 2023, and nine pieces of published evidence on AI-scaled sites graded on origin, method, date and independence. The sources show no penalty for using AI, and no audited before-and-after record of a named site.
Shreyas M VA page-by-page test for deciding which pages of an Indian B2B SaaS site need a version for each market, with a four-structure comparison, an hreflang map and pricing-page rules. Sourced India versus US figures are dated, and the structure rules are our analysis, untested.
Shreyas M VA channel playbook for dev-tool founders: docs, self-serve trials, open source, Hacker News, communities, content and paid, each with a sourced example, a measure and an evidence label. It ranks no channel, because no source measures results.
Shreyas M VA channel plan for agency owners that lines up nine ways to get clients by cash cost, time to a first client, one rating study and spam risk, with the rupee arithmetic shown. All four sources sell outreach or funnels, and none gives India-wide data.
Shreyas M VSearch, opted-in WhatsApp, marketplace listings and creator seeding, tested against the rupee arithmetic: what one order leaves, the click price a brand can afford and the repeat rate that breaks even. Built from three 2026 guides; none gives measured channel costs for India.
Shreyas M VTen founder-reported Product Hunt outcomes from three public sources, sorted by what they produced, with the reasons founders give for the gap between upvotes and customers. No source counts launches, so the share that produce customers is unknown, and the reports are self-reported.
Shreyas M VWhat builds trust in content from CA, law and consulting firms, a six-step check to run against your professional body's rules before each post, and an audit of five figures from a CA marketing guide. It states no ICAI or Bar Council rule: that text was not in the research.
Shreyas M VA sourced timeline of Quibi's spend against the signals that followed: funding figures that disagree, a download spike, a fall in the app rankings and a six-month shutdown. No source gives a marketing budget, so the page says what that leaves unproven, and adds seven checks for a founder.
Shreyas M VMarketing OKR examples for awareness, acquisition, conversion and retention, with seven tests that catch a vanity key result and a worked cost-per-lead rewrite. Built from four published guides, one from an OKR software vendor. It contains no data of Kanso's own.
Shreyas M VAn essay with a test-before-commit rule for new marketing channels: a written brief, a spend cap, a kill line and a stepped budget, built from Salesforce, Coursera and Madison Logic. It retells no history of Google+ or Clubhouse and reports no test results.
Shreyas M VSeven core questions for a founder interviewing a Head of Growth, each with what a strong answer contains, what a thin one sounds like and a follow-up probe. Includes a 0 to 2 scoring sheet and a worked reading of the percentages in sample answers. Built from two guides; the rubrics are untested.
Shreyas M VA five-slide outline for the marketing section of a board deck, built from two practitioner guides: pipeline against target, cost per customer won, payback, where new business comes from and the decision. Includes recalculated worked examples and where the guides disagree. It contains no data of its own.
Shreyas M VA labelled opinion with a model: 10% to 30% of the working week, which is 4 to 12 hours in a 40-hour week, set by whether a channel already works and never below four hours. Shows the arithmetic, a split of the hours and rules for changing them. No source supplies the figures.
Shreyas M VEight technical SEO checks for a Next.js site, written as CI assertions: rendering, metadata, canonicals, sitemap, robots and status codes. Seven come with excerpts from Kanso's own test suite and one is a labelled sketch. It reports no ranking or traffic results.
Shreyas M VA six-question decision tree that sends each old page to keep, update, overhaul, merge, redirect or delete, plus a refresh log template with baseline and check dates. Built from three practitioner guides and Google's help page, read in November 2026. None of them measures results, and the tree is untested.
Shreyas M VA nine-check pass/fail protocol for sites made with AI builders, covering direct loads, titles, canonicals, indexing instructions, links and the sitemap, built from Google's SEO Starter Guide and a pre-launch audit method. It reports no tests of any builder.
Shreyas M VA 30-day run-up and a seven-day calendar for launching on Hacker News, Reddit, LinkedIn and X, with the asset due each day and the platform rules behind it. The sequence is Kanso's design, built from sourced rules and timings, and no source tests it against results.
Shreyas M VA one-screen matrix for duplicate, moved, retired and private pages: which of a canonical, a permanent redirect, noindex or robots.txt to use, and the failure each wrong setting causes. Built from two Google Search Central pages and three named secondary sources. It contains no tests of its own.
Shreyas M VSeven naming rules, three naming models and eight governance rules (owner, link log, review, audit) drawn from three vendor guides, with a worked controlled vocabulary and a copyable spreadsheet builder. Kanso's own UTM builder is not built yet. It contains no data of its own.
Shreyas M VA five-question decision tree that sorts a marketing task into a workflow, an assistant or an agent by how often it repeats, whether its steps can be fixed, what data it needs and what a wrong result costs. Includes 14 tasks classified, built from three vendor guides and untested.
Shreyas M VA keyword prioritization scorecard turns a list of candidate keywords into one ranked queue by scoring each on six weighted criteria out of 100. See every weight, the evidence label each score needs, the score bands and a worked example, plus how this differs from the Google Ads rule with the same name.
Shreyas M VSeven weekly marketing numbers for a seed-stage founder, each with its formula, where to read it and the decision it should drive. Includes Waveup's worked formulas recalculated, a small-volume funnel example and a 30-minute review. Built from three secondary guides; it contains no data of its own.
Shreyas M VA twelve-field marketing brief template with the part three template pages leave out: acceptance criteria, the pass-or-fail tests a deliverable is checked against before you accept it. Includes a filled example for an invented company and a rule for work that fails.
Shreyas M VThe CAC payback formula with each input defined, a spreadsheet layout and 13 worked test cases with the arithmetic shown, built from three published guides. Includes a unit check on one guide's examples and a hypothetical churn table. It is a guide, not an interactive tool.
Shreyas M VA 14-column editorial calendar template for teams of two to five, with seven stages and six stage gates that set out what must be true before an item moves on. Includes sign-off by team size and a dated worked example. The gates are Kanso's design and untested.
Shreyas M VThe sitemap-to-AI-articles case as a sourced timeline: the method, five reported figures with their arithmetic, what Google's guide says, and what is still unverified. Every case figure comes from one secondary source, and none is independently audited.
Shreyas M VA competitor analysis template with a source, a date, an observed-or-inferred flag and a confidence rating for every claim. Includes a four-level scoring rubric and a ledger filled in from the three pages behind it. It contains no market data or survey of its own.
Shreyas M VGoogle says a week or so for a new page, and other published estimates run from hours to eight weeks. See each figure by source with what it measures, why they disagree, and a log template for timing your own posts. It contains no Kanso indexing log.
Shreyas M VTwelve AI CMO tools compared by function, with eight published criteria, vendor prices dated October 2026 and annualised with the arithmetic, the four places the vendor pages disagree, and a seven-day test sheet. Kanso did not test these tools, so there are no screenshots or scores, and no rankings.
Shreyas M VMonthly SEO tool prices from four guides converted to rupees at ₹95.11 to the dollar, with 18% GST worked out and the places the sources disagree. Then a stage-by-stage buying guide from free tools to one all-in-one, and why group-buy plans are not recommended. Prices are asking prices, not tested.
Shreyas M VThe sample rules, metric, windows and decision rule for testing title tag rewrites on Kanso's own blog pages, fixed before any title changes. Includes the sourced evidence on Google's title rewriting and a worked CTR example. No results exist yet.
Shreyas M VPublished AI Overview rates run from 15.69% to about 50% of queries, and the sources disagree on why. See the figures side by side, a manual monthly sampling method with a log template, and the margin of error for a sample of 100. It reports no Kanso sample.
Shreyas M VA ten-section brand voice guide template for AI writers, with each idea credited to the published guide it comes from, a banned-phrases list and a step-by-step drift test with a scored, invented example. The drift test is Kanso's own method and has not been measured.
Shreyas M VThree signals that it is time for a first marketing hire, signs it is too early, generalist versus T-shaped profiles, and a six-line cost method. Includes dated rupee bands for adjacent roles in India. No source we read prices the role itself, and the page says so.
Shreyas M VA five-step workflow for editors: extract claims, match each to a source someone opened, re-derive the numbers, decide, and log it. Includes Kanso's provenance log template with a filled example, built from three sources read in October 2026. No test of our own.
Shreyas M VA competitor-tracking log where every finding carries its evidence type, source, retrieved date and a High, Medium or Low confidence grade. Includes the grading criteria and ten published statistics graded as a worked example, one Medium and nine Low. It contains no Kanso test data.
Shreyas M VExport Search Console queries, cluster them by intent, check each cluster against your existing pages for cannibalization, then score what is left. Includes a worked example with invented figures, and says plainly that the scoring model is untested.
Shreyas M VThree published guides give 12 to 20 percent of revenue for a young company, 6 to 12 percent once sales are rolling and under $1,000 a month before revenue. See each figure side by side, worked out in rupees with the arithmetic, plus a cut-first order. None of the text is India-specific.
Shreyas M VA copyable rubric of twenty yes-or-no checks for AI-written drafts, in five groups: accuracy, originality, intent, voice and sourcing. Each check has an invented pass and fail example, with blocker rules, score bands, a worked scoring and a reviewer calibration method. The bands are untested.
Shreyas M VWhen ChatGPT or Claude is enough for marketing and when a growth OS pays off, judged on time, cost, consistency and context. Seven situations where doing it yourself wins, a break-even calculation with hypothetical figures, and no test of Kanso against either assistant.
Shreyas M VNo public source holds a dated pre-launch record of a named top launch, so this page builds a timeline from three sources instead. It places a 30-day warm-up, launch emails and maker comments on one line, shows the hours arithmetic and labels each step prescribed, observed or inferred.
Shreyas M VA seven-check manual checklist for vetting subreddits: three gates (rules, activity, comment depth) and four fit checks, with verdict rules and a hypothetical worked example. It merges three vendor guides, and no source measures how well the method works.
Shreyas M VMonthly SEO prices in rupees from three Indian agency pages and one US guide, read on 27 October 2026, lined up by tier with the overlap worked out. Covers what each tier buys, GST on top, the red flags in a quote and a payback check. All sources are sellers, so these are asking prices.
Shreyas M VThe design of a test of whether refreshing old posts moves rankings, published before any results exist: hypothesis, random split, one primary metric, 12-week windows and a written decision rule. It also reads the published refresh figures for what they cannot show. No results exist yet.
Shreyas M VDated rupee rates for content writers in India from eight pages, by the word, by the 1,000-word article, by the hour and by the month. Includes the overlap between three sources, a worked eight-article budget with GST, and what a lower quote tends to leave out. No source is a survey of prices paid.
Shreyas M VWhich AI crawlers train models and which power search, what blocking GPTBot, ClaudeBot or Google-Extended changes, and a block-or-allow matrix for eight business models. Includes a worked robots.txt example showing why a partial block can fail. It contains no survey data of its own.
Shreyas M VOne published guide puts a fractional CMO in India at ₹1,50,000 to ₹4,00,000 a month, against ₹8,00,000 to ₹20,00,000 for a full-time CMO. See every sourced figure with its arithmetic, where sources disagree, and the gaps for an AI CMO and a first hire.
Shreyas M VA five-day customer research sprint for founders: read threads, reviews and support tickets first, then run scripted interviews. Includes an eight-question interview script and four fill-in templates. The method is built from three published guides, and the schedule is our own design, untested.
Shreyas M VFourteen startup communities scored 0 to 4 for four founder audiences on four stated checks: open, free, a place to talk and a named audience. See the arithmetic, what the sources say about Slack, why Discord is not scored, and the cost of one gated community worked out.
Shreyas M VA six-step voice of customer copywriting method for reviews and threads, with a worked mining example: nine phrases from three hypothetical statements, counted by voice and not by row, then turned into draft lines. It shows what a small set supports, and no source here measures conversion.
Shreyas M VChargebee's public record shows a content hub, glossary pages, events and a free entry plan, but no free tool. See each free offer graded by source, plus a price-tier table built from Chargebee's own GTM figures that shows which free offer fits which price point.
Shreyas M VboAt sold on Amazon, Flipkart and Myntra first and backed it with Bollywood and cricket endorsers. See the revenue figures with the arithmetic, a channel-mix reading of the 2018 numbers, the figures that conflict, and what no source here covers.
Shreyas M VA 25-week calendar for launching a B2B SaaS in India and the US on one date, from 12 weeks before to 12 weeks after. Built from three published launch guides, with the time-zone arithmetic shown and our own analysis labelled. No source here has India or US market data.
Shreyas M VWhat llms.txt is, what Google's guidance says about it, which tools and agents read it, and a ship-or-skip rule. Includes a table that grades each claim by who made it, the adoption arithmetic, and a plain statement of what no source has measured.
Shreyas M VEvery popular GEO tactic graded as official guidance, research, practitioner opinion or unknown, with Google's guidance read first. Includes the tactics Google says to skip, a five-check framework for GEO vendors, and a plain account of what no source here can show.
Shreyas M VForty SEO audit checks for small sites in five tiers, from the blockers that keep a page out of search to conditional checks most sites can skip. Each row says how to run the check and when to skip it, built from Moz's and Serpstat's audit guides.
Shreyas M VWhat public sources say about Show HN titles, first comments and timing: the rules, the survival figures from a 10,000-post sample, the top-10 scores with the arithmetic shown, and a method for building your own dataset. It says plainly that no first-hand dataset is included.
Shreyas M VA playbook for small teams: a ten-field experiment template, then the low-traffic statistics in plain terms. Worked sample-size and duration tables, five options when traffic is short, and a decision rule to write before launch. No experiment results are reported.
Shreyas M VA 25-field content brief template in four parts, with information gain and source fields that the pages read for it do not list. Includes a filled-in brief for this page's own topic, a field-coverage table for two ranking pages, and the fields a solo writer can skip.
Shreyas M VX can work for B2B SaaS when the buyer is a developer or founder, and the evidence does not support it as a demo engine. See every figure in three ranking guides graded by date and basis, a five-question decision table and a $1,000 test worked through.
Shreyas M VA six-check fairness rubric applied to ten SaaS 'X vs Y' pages as three published write-ups describe them. See which show figures on both sides, dates and the rival's strengths, and why the pages themselves were not opened.
Shreyas M VA risk-by-task matrix for 17 marketing tasks, scored on reach, undo cost and stakes: which need approval before AI output ships, which a spot-check, which only a log. Built from three vendor guides, with the scoring shown, no tests of our own and a clear line between documented and analysis.
Shreyas M VA four-step SaaS positioning method (alternatives, unique attributes, value, segment) worked through two invented companies, with a statement template, five tests and a two-week schedule. The examples are illustrations, and no source here measures results.
Shreyas M VTwelve SaaS pricing pages scored on a five-doubt rubric (cost, fit, limits, worth, navigation) from two published teardowns, with the arithmetic shown, where the guides disagree and steps to score your own page. No screenshots, and no source measures conversion.
Shreyas M VPostman's path from a free Chrome Web Store tool to a team standard, read channel by channel. Dated user and funding claims with the arithmetic shown, a workspace breakdown, and a stage-by-stage check of which moves a dev-tool founder can reuse.
Shreyas M VZapier's integration pages pair a searched task with data only Zapier holds, published for every app and app pair. See the page families, the templates side by side, traffic figures that disagree, and a six-check usefulness gate.
Shreyas M VSkip AI when briefing it, reviewing it and redoing the rejected part takes longer than doing the task yourself. A break-even formula for founders and seven tasks to hold back, each tied to a published source, with hypothetical hour figures and no tests of our own.
Shreyas M VA task-by-task map of 19 marketing jobs: which an AI CMO can run alone, which it can only assist, which stay human, and where approvals belong. Includes what Kanso itself does not do and why the sources disagree on autonomy.
Shreyas M VA one-page marketing strategy fits nine fields on one sheet, from situation and objective to what you are not doing. Get the template, a worked example for an invented company, a table showing which of three published templates each field comes from, and what to leave out.
Shreyas M VTen AI company homepages read on 17 October 2026, with each claim scored on a four-level evidence rubric. See which claims show a source, a named person or only a link, which show nothing beside them, and how to run the same check on your own page.
Shreyas M VA content distribution plan for small teams that maps one Kanso article to twelve assets across LinkedIn, X, Reddit and owned channels. See the selection checks, what changes by channel, a time budget, and what the page cannot show: no results are reported.
Shreyas M VCanva's template pages share one layout, carry a working tool and sit in a parent and child hierarchy, but three write-ups put its organic traffic anywhere from 73.2 million to 270 million a month. The figures side by side, the arithmetic behind the 1 million new users claim, and what a small team can borrow.
Shreyas M VA five-step, 30-minute weekly Reddit routine for B2B founders: rule check, search, a 10-point thread qualification rubric, replies and a log. See the rubric scored on three example threads, a four-week stop rule, and what no source could show: no results are reported.
Shreyas M VPull the queries ranking around positions 8 to 20 out of Search Console, check the live results, and sort each into improve, leave or new page. Includes eight do-not-touch rules and a worked example with invented figures, and says plainly that no source here measures results.
Shreyas M VBrowserStack's growth read from its own company page and three profiles: a dated timeline, the arithmetic behind its funding and customer figures, where the sources conflict, and a sitemap method for the search traffic no source measures.
Shreyas M VA technical founder can cover the first 90 days of marketing in 71 hours over 13 weeks. See the hours, the output and the stop rule for every week, how the estimates were built, and what two sourced founder accounts do and do not show.
Shreyas M VA founder needs four Search Console reports on a monthly schedule: Performance, page indexing, URL Inspection and Sitemaps. See what decision each report supports in a table built from Google's documentation, how the metrics are defined, and which reports to skip.
Shreyas M VZoho's growth strategy checked claim by claim: the product-suite flywheel, the no-ads and bootstrapped story, and what the sources can and cannot back up.
Shreyas M VSearch Console's Generative AI performance report counts impressions in Google's AI features by page, country, device and date, with no clicks or queries. See what it shows, what it leaves out, why you may not see it, and a monthly review routine tied to landing pages.
Shreyas M VMatch each Search Console status to the stage that is failing: discovered, crawled, indexed, ranked or clicked. A six-question decision tree with causes, fixes, Google's own rules for each status and the Next.js settings that stop a page at the index stage.
Shreyas M VHacker News can work for B2B founders who sell to technical buyers, but only as participation, not broadcast. See what its guidelines allow, a post, comment or stay out call for seven situations, and what no source here could verify.
Shreyas M VA LinkedIn posting system for technical founders starts with a log of what you build, decide and hear. See the source map for commits, customer calls and decisions, the screening checks, five post shapes, a worked example and a weekly routine.
Shreyas M VZerodha grew mainly through low pricing, referral and affiliate payouts, and free education in Varsity rather than paid advertising. No source gives Varsity's traffic, sign-ups or cost, so this separates the documented from the inferred and reconciles customer figures that disagree.
Shreyas M VStripe's docs work as marketing by letting a developer run a real payment request before talking to anyone, but no source measures the effect. See what docs.stripe.com shows, a dated search check, how the top pages treat docs, and an eight-line audit for API products.
Shreyas M VMost organic traffic drops come from a small set of causes, and you find yours by segmenting the drop before you change anything. Seven steps, a segmentation worksheet and a change-log template, with AI Overviews tested last instead of blamed first.
Shreyas M VAnswer first, say who you are, mention the product only when it fits and keep the link out by default. A five-step Reddit reply framework built from three top guides, with annotated good and bad example replies, where the guides contradict each other, and a short note on Hacker News.
Shreyas M VFreshworks' reported marketing channels, from comparison pages and expert webinars to a Dreamforce blimp, graded by how well each is sourced. See what its Nasdaq listing and 10-Ks confirm, and what is still unverified.
Shreyas M VDuolingo had 58.7M daily and 140.6M monthly learners in Q2 2026 and targets 100M daily users by the end of 2028. See the timeline, how free lessons, mobile, streaks and weekly testing built the base, why growth slowed in 2025, and how it plans to speed up.
Shreyas M VRazorpay's early traction came from developers who could test the product on their own, but no source measures what its docs added. See a table of every docs claim by source, the acquisitions behind its product suite, and what is known about its startup programs.
Shreyas M VReddit users say they trust the platform 90% of the time to learn about new products, and 51% of global purchase-discussion mentions happen there. See which real subreddits founders use, the organic and paid routes, real case studies and a first-30-days plan.
Shreyas M VX ranks posts in three stages and, according to three 2026 guides, rewards replies, early engagement and X Premium accounts while holding back posts that link off the platform. See what each guide claims, where they disagree and how to test the rest on your own account.
Shreyas M VA five-point checklist to run every week: SEO, AI search, content, competitors and distribution. Covers six SEO terms to check, what Google says about AI Overviews, which AI crawlers to allow, and a 60-minute review plan.
Shreyas M VVercel crossed a $500M annualized revenue run rate in early July 2026, up from $200M+ ARR in September 2025. See the revenue timeline, the funding rounds, and how Next.js, preview deployments, a free plan, v0 and AI agents drove it.
Shreyas M VShopify merchants processed $378.4B in GMV in 2025 and Shopify earned $11.56B in revenue, on a business that began as software for a snowboard store. See the timeline, the App Store, partner and payments flywheel, Shopify Plus, and the logistics reversal.
Shreyas M VEnter your website and let Kanso handle the rest.
Get started free. No credit card required.