Public Record
What changed, and when
We log material content updates here so you can see how this site changes over time — corrections, methodology shifts, new sections, and editorial decisions. Typos and styling tweaks don't make the list; anything that could change a recommendation or a number does. See our methodology for how we decide what to update.
- Tools
The calculator now knows the difference between estimated and measured
If you've actually tested your infusion — a home tester or a lab result — the dosage calculator will stop guessing. Switch Step 2 to I measured it, enter the number with its real unit, what you tested and where the number came from, and we drop the label, decarb and extraction assumptions entirely: your figure becomes the potency and we calculate servings from it, with no range attached. Every result now carries a provenance stamp — Calculated when we estimated it, Measured input when the number is yours — so it's always clear who did the measuring. Try it on the edible dosage calculator. - Tools
Machine Finder answers in three questions instead of six
The quiz used to make you answer six questions before it told you anything. Now three — how often you infuse, how big your batches are, and the one thing that matters most to you — get you a straight provisional answer: stay manual, a compact all-in-one, or a large-batch machine, with the reason it fits. If you want a named product, three optional questions turn that class into a winner, a runner-up, and why we didn't pick the third option. Both the short and the full result are shareable links, and the scoring weights behind the new "what matters most" question are published on our methodology page. - Content
Gummy Batch Builder on the tincture gummies recipe
The gummies recipe now does the arithmetic both ways: tell it your tincture strength and it says how strong each gummy will be, or tell it the dose you want per piece and it says how much tincture to pour. It replaced two overlapping calculators that used to sit on that page, and the result is labelled as a calculated estimate — not a measurement. - Site
Honest reporting on which recommendations readers actually see
Behind the scenes we fixed how we measure our own product callouts. A few blocks were logging clicks without logging that anyone had seen them, which made them look worse — or better — than they really are. Links inside articles are now counted separately from blocks you have to scroll past, and we no longer publish a "click rate" for anything we can't honestly measure a view for. Nothing on the page changed for you; it means our own decisions about what to keep, cut, or rewrite are based on real numbers. - Site
One clear next step at the top of every section page
The menu is now just the four things you might be doing — Plan, Make, Verify, Fix — and each one opens a short list of tasks ("Find my setup", "Make cannabutter", "Calculate a dose", "Weak batch") instead of content-type labels. The stage you're currently in is highlighted, so you always know where you are. Section pages open with a single main action plus two quieter alternatives, and the section right below it moves the decision forward with two clear paths rather than repeating the same buttons in a different box. - Site
Section pages now start with a decision instead of a description
Every section page — Infusion, Recipes, Dosage, Tools, Fix It — used to open with a full screen of explanation before offering anything to do. Those headers are now about half as tall and lead with the choice instead: pick what you already have, or what you want to make, and go straight there. Headings say what the page actually covers ("Cannabis Recipes by Infusion Type" rather than "Recipes"), links say what you'll get done ("Make tincture" rather than "Read guide"), and the busiest paths are separated from the long tail so the important ones aren't buried. The menu now groups everything by what you're doing — plan, make, verify, fix — while keeping the plain names for each section. - Site
We fixed how we measure ourselves, so the next changes are the right ones
Nothing you read changed today — how we judge it did. Every gear block, next-step link and email offer on the site now reports itself in one shared vocabulary, so we can finally answer two honest questions per surface: did readers ever actually reach it, and did it earn a click once they did. The old accessory-rail counters, which quietly counted a single page view several times, are retired in favour of one viewport-verified impression per reader. The dosing sheet and the batch tracker now report which one was offered where, so we can tell whether the offer or the placement is the problem. Practical upshot: fewer guesses, and anything we add next has to beat a real number. - Site
One decision per gear block, and a Green Dragon recipe you can follow at a glance
Three of our busiest guides had a generic wall of gear at the bottom. Those are gone. In their place, each page asks the one question that actually changes your result — is your oven lying about 240°F, is your alcohol strong enough, are your molds the size you think — and names the single tool that answers it. The tincture guide now opens Method 1 with a complete Green Dragon spec card: ratio, proof, decarb, steep time, straining, and realistic finished volume, so you don't have to read six paragraphs to start. The troubleshooting index is now honestly an index — symptom first, with the full answer living on the guide behind it — and "why don't edibles work on me" is the top entry, because it's the question most of you arrive with. Failure pages now offer the printable batch tracker, which is the thing that actually tells you which fix worked. - Tools
The dosage calculator now tells you what it assumed
Every edible calculator online gives you a different number, and none of them tell you why. Ours now asks the three questions that actually move the answer: what you're starting with (raw flower, already decarbed, concentrate, or RSO/FECO), whether your label says Total THC or raw THCA, and which method you're using. Distillate and RSO no longer get charged a decarb loss they already paid at the lab — an 800 mg syringe reads like 800 mg. If your COA lists THCA, we apply the 0.877 conversion and say so instead of letting you overshoot. Hit Show the math under any result for line-by-line arithmetic plus a plain-English list of every factor we applied and why. You can copy a shareable link that carries your whole setup. The page also explains where other calculators disagree with us and what they're hiding. Elsewhere: functional mushrooms moved out of the top nav into the footer, so the header is only the five things most people came for. - Site
Said things once, and put the break-even math where you need it
The stage strip at the top of each page was saying the same thing twice — the stage's job and then the page's job, in near-identical words. It now says your job here, once. The big "is a machine worth it" calculator had also been copied onto four different pages; the full editable version now lives in one place, and everywhere else gets a three-number summary with a link to change the assumptions. On the dosage calculator that summary uses the flower weight you already typed in. The machine reviews — Ardent FX, MB2e and the LEVO 2 guide — now show that break-even too, which they never did before. All of them use our conservative recovery assumption, so no two pages quote you a different payback. A few duplicate product notes and a second cheat-sheet box came out of the calculator page. - Site
Fewer asks per page, one clear next step
Guides had grown up to a dozen competing calls to action — an older hand-placed set that was never cleared out when the Lab handoffs arrived. Each page now carries one dominant next step, at most a couple of genuinely relevant product notes, and one place to save the cheat sheet. Duplicate product mentions that repeated an item already listed in the starter-set box are gone, machine pitches were pulled out of the middle of the how-to guides, and the "what's next" strip quietly steps aside when it is no longer the main handoff. The mobile menu is now grouped by what you are actually doing — plan, make, verify, fix — with bigger tap targets. Nothing instructional was removed. - Editorial
Two product blurbs claimed hands-on use we haven't earned — corrected
The Ardent FX and LEVO II blurbs that appear in callouts across the site said "we've used" and "our pick." Every review page here is stamped specs plus owner reports, not hands-on, so those two lines contradicted our own evidence label. They now say what they actually rest on. In the same pass, machine prices used by the payback math read from one place, so a price change can't leave the break-even arithmetic quietly wrong, and the unused display module for hands-on tests was removed until there is a real test to publish. - Tools
The dosage calculator now answers “is a machine worth it?”
The calculator is where most people arrive, so the cost question now lives there too: what a batch costs you today, what a sealed-chamber machine changes, and how many batches it takes to earn its price back. Defaults are editable, the arithmetic is expandable, and the Machine Finder is one click away if you'd rather have the comparison done for you. - Tools
The Machine Finder now tells you whether the machine actually pays back
After the winner and runner-up, the result shows a payback check: how many batches it takes to earn the price back at the cadence and batch size you entered, and how many months that is. The arithmetic is expandable — time saved per batch, flower recovered, the hourly rate we used — so you can disagree with an assumption instead of trusting a number. When the math says the machine will never pay back in dollars, it says that plainly and tells you the honest reasons to buy one anyway (consistency and smell), which are not financial ones. - Site
The home page now asks where you're standing, not what you want to read
The front page opens with the four stages of the workflow — plan, make, verify, fix — instead of three reading paths, so people arriving with a broken batch land on diagnosis rather than a recipe. At the same time the last fourteen pages that weren't wired into the workflow (mushroom extractions, honey, sugar, seasonings, drinks, Kool-Aid, savory recipes, flower sourcing, cooking tools, the LĒVO butter walkthrough, and Start Here) now show their stage and hand you a next step like everything else. No page in the lab dead-ends now. - Site
The site now works like one workflow: plan, make, verify, fix
Every guide, recipe, tool, review, and troubleshooting page now tells you which part of the process you're in — planning a setup, making a batch, checking how strong it is, or fixing a result that went wrong — and offers one clear next step plus one alternative if your situation is different. No more pages that stop dead, and no more piles of competing buttons. You can still start anywhere: a weak-butter problem starts at Fix, a machine question starts at Plan, and nothing forces you through steps in order. - Site
Handoffs now match what you came for — and two dead product picks are gone
If you're reading a tincture or gummy page, the next step we offer is dose accuracy — mg per mL, mg per dropper, mg per piece — not machine shopping. Machine comparisons stay on the pages where buying a machine is actually the question (cannabutter, decarb, the finder). On the dosage calculator, the next action is saving your own numbers with the free batch tracker instead of a product pitch. We also removed two placeholder product callouts that never went anywhere (delta-8 flower, hemp sampler) and demoted high-proof grain alcohol from a product callout to plain sourcing advice — it can't ship in most states and we earn nothing on it, so pretending otherwise was noise. Email capture on the free downloads now tracks the same way everywhere, and the slide-in waits for the bottom bar to leave the screen instead of stacking on top of it. - Editorial
Locked the addition checklist and scored the growth menu
Codified the way we've been operating into a permanent gate: every new page, component, CTA, event, or affiliate surface must pass a 7-item checklist — real signal, written hypothesis, reversibility, tracking on day one, honesty rubric, a read date, and kill criteria — before it ships. No "nice to have" launches, no dark experiments. Alongside it, scored 15+ candidate expansions (deeper testing, potency vertical, storage & smell control, digital products, Amazon Influencer, direct partner deals, Reddit, Pinterest, newsletter) against the same rubric we use for buyer opportunities. Cells without real evidence are left TBD instead of guessed. Also wrote down an explicit kill list — grow, CBD wellness, pet cannabis, mushroom expansion, ads, paid placements, grocery affiliate, concentrates hardware — so those pitches never come back without a formal strategy amendment. Nothing on the site changed today; this is a governance ship. - Tools
MagicalButter is a real recommendation now, not a dead link
MagicalButter approved us for direct affiliate. Retired the old Amazon short link (which had been flagged as a duplicate of the LEVO link and was rendering as a "Link coming soon" badge on every buyer page) and wired the direct partner URL into four high-traffic tools pages plus a new MagicalButter MB2e review. The review is an honest owner-report synthesis — not a fabricated hands-on test — and clearly labels its evidence level. The Ardent vs LEVO, LEVO comparison, LEVO cannabutter guide, and cannabutter recipe all now surface MB2e as the value/big-bowl alternative where it actually fits the reader's job. - Tools
Wired Awin products into the pages people actually read
A 60-day tracking audit showed zero Awin clicks — not because the products underperform, but because every Awin product (grinder, storage jars, hemp/CBD flower) lived on a single low-traffic sourcing page. Fixed the surface, not the copy: decarb guide now recommends a coarse-grind herb grinder and points readers without flower to a legal CBD hemp option; cannabutter and tincture guides now recommend airtight UV-blocking storage jars in their Storage sections. Also flagged the two Awin flower products that shared one attribution URL as placeholders so their clicks stop conflating with CBD flower until we generate unique tracking URLs. - Trust
Moat sprint: Batch Tracker PDF, Buyer Judgment block, hands-on framework
The email magnet is now a two-file bundle: the dosing cheat sheet plus a new printable Batch Tracker — one page for every batch (flower / decarb / infusion / yield / mg-per-serving). Every cheat-sheet surface (main capture, inline compact, slide-in) now offers both PDFs. On the buyer side, LEVO cannabutter guide, LEVO machines comparison, Ardent FX review, and Ardent vs LEVO now all carry a standardized Buyer Judgment block — Best for / Skip if / Cheaper path / Evidence level — so every machine page honestly says what it can't do and what to buy instead. Alongside it we published the testing protocol itself — the "How we test" section on /methodology documenting the Manual Cannabutter Setup Test. (Update, 2026-08-05: the unused display module for hands-on tests has been removed. We will publish the module when there is a real first-person test to put in it, not before.) No new routes, no cluster expansion, no new taxonomy events. - Content
GSC query-match polish (decarb + LĒVO 1 vs 2)
The 28-day GSC read confirmed Google is now testing the mason-jar decarb cluster (six query variants, 16 impressions) and the LĒVO 1 vs LĒVO 2 query (position 17). We held the content freeze and made two small text-only polishes: the decarb guide's mason-jar intro now naturally covers the tail-query variants Google is showing us ("mason jar decarboxylation", "decarbing in a mason jar", "decarboxylation in mason jar"), and the LĒVO 1 vs LĒVO 2 section expanded from one paragraph into a straight feature-by-feature table for used-market buyers. No new pages, no new components, no CTA changes. - Site
Tracking honesty pass + legacy 404 sweep
Two quiet bugs were hiding real activity from our snapshot: the cheat-sheet forms fired alead_magnet_signupevent that wasn't on the DB whitelist, so 16 real subscribers looked like zero; and no page emitted the whitelistedtool_startevent the dashboard uses to measure calculator/finder abandonment. Both are fixed — the calculator and Machine Finder now emittool_startonce per session, and every cheat-sheet surface now mirrorsemail_signupto the database alongside the existing GA4/Clarity signals. Also captured the top legacy 404s still bleeding traffic: WP Recipe Maker print URLs (sugar, tincture, honey) now redirect to the current guides; the old free-dosage-calculator, Infuzium machine page, and/pages/about-usredirect to their modern equivalents; off-topic slugs (nutrients, mold, CBD types, teen vaping, seed buying, high-protein breakfasts) return 410 Gone. Finally, the two surfaces where money is actually forming got a small push: the Everclear callout on the tincture guide moved up to where readers are actively picking a bottle, and the decarb guide now hands off to the cannabutter guide directly under the accessory rail — reinforcing the decarb → butter → affiliate loop. - Content
GSC-aligned polish: gummies, mason-jar decarb, LĒVO buyer pages
Search Console shows Google testing three areas: gummies with tincture, mason-jar decarb, and LĒVO buyer terms. No new pages — we densified the three existing pages against the exact queries: the gummies recipe gained direct-answer sections ("How to Make Gummies with Tincture", "Can You Use Tincture in Gummies?", "How Much Tincture to Add", "Why Tincture Gummies Won't Set", "Best Tools"), and the LĒVO comparison gained "LĒVO 1 vs LĒVO 2", "Is LĒVO Worth It for Cannabutter?", "Best Machine If You Only Make Butter", and "When to Skip the Machine Entirely". We also cleaned up legacy URLs still showing in GSC: the old Amazon-machines archive now points at the LĒVO comparison, the WordPress/page-sitemap.xmland/cannabis_strains-sitemap.xmlpaths redirect to our real sitemap, and the CBD-pets / bare strains archive return 410 Gone so Google removes them from index. - Site
Cannabutter revenue loop + honest tracking upgrades
Two changes. First, our cannabutter guide is now the center of the maker workflow — it has a side-by-side handoff at the end (skip the machine with a manual starter kit, or compare LEVO/Ardent/MagicalButter) and stronger reverse feeders from the "why is my cannabutter weak?" troubleshooter. Our LEVO 2 operating guide also now shows the LEVO product card in the first screen instead of only at the bottom, so buyer intent has somewhere to click. Second, internal reporting became more honest: the Machine Finder no longer inflates completion rate when someone lands on a shared result URL (they now count as a separate "shared result" entry, not a fresh start), the retired "calculator_result" event is called out as stale, and there’s a new money-join panel showing exactly how many affiliate clicks reconcile 1:1 between our click log and the analytics mirror. Nothing reader-facing changed in recommendations. - Site
Admin reporting fixed past 1,000 rows
Internal reporting now reads paginated analytics data instead of stopping at the first 1,000 rows. The snapshot also shows the latest view, event, and affiliate-click timestamps so we can tell whether numbers are moving before making a build decision. No reader-facing recommendations changed. - Site
Admin snapshot simplified — raw stats and a traffic-light health check
Our internal reporting page got noisy with self-monitoring tiles that tracked whether our tracking worked. We trimmed it back to the numbers that matter — reach, engagement, monetization — plus a single green / amber / red light per area. If counts are moving, the wiring works. Nothing reader-facing changed. - Site
Tracking truth, part 3 — every hub and spoke now reports, plus we can see what's leaking
Two changes. First, the remaining hubs and spokes (infusion, recipes, dosage, tools, troubleshooting, mushrooms, and 17 leaf guides) now use the same tracked links as our flagship guides — 70 more links flipped in one pass. Second, we added three quiet signals so we can spot problems instead of only confirming what already works: which guides readers finish vs. abandon halfway, which tools people start but never finish, and which searches land on a missing page so we know what to write next. Internal reporting now answers “what should we work on this week?” with a single sentence on the admin dashboard instead of a SQL dive. None of this changes what we recommend to readers — only our ability to tell, honestly, what's working and what isn't. - Site
Tracking truth, part 2 — the seven highest-traffic guides now report their own funnel
The decarb guide, cannabutter guide, tincture guide, gummies recipe, dosage calculator, machine finder, and starter kit page now use the same tracked links as the home page. That’s 45 in-content links flipped over in one pass. We also added three honesty tiles to our internal reporting: which instrumented pages are actually emitting clicks, what percentage of affiliate clicks join back to a funnel event (so we can spot any buyer module quietly bypassing the wrapper), and the new calculator-success metric next to the old one so we can watch the dedupe land before retiring the old series. Nothing about what we recommend changed — only our ability to tell which recommendations are working. - Site
Tracking truth pass — homepage now uses the same tracked links as the rest of the site
Every link on the home page — the three intent tiles (Make / Dose / Machine), the Start Here button, the Quick Start recipe steps, the four pillars, and the featured guides — now flows through the same tracked-link wrapper we use elsewhere, with the same fields (surface, placement, label, destination). This closes a measurement gap where homepage clicks looked low only because they weren’t being counted the new way. Our internal reporting also now reads both the legacy and new events during the transition, so we don’t double-count and don’t lose history. Nothing about what we recommend changed. - Site
Tracking consistency pass — every affiliate click and buyer-module view now logs the same way
We unified how clicks are measured across the site. Every recommended product, every rail card, and every “next step” button now records the same set of fields — which page, which component, which product, and where the visit came from — so we can tell honestly which guides actually help readers find the right tool. Also added written standing rules so future updates inherit the same treatment. Nothing about what we recommend changed; only how we keep ourselves accountable to it. - Trust
Published methodology, changelog, and corrections process
Added a public methodology page explaining our source hierarchy, kitchen-testing protocol, affiliate policy, and what we will not claim. Added this changelog and a corrections form. Added Last reviewed stamps and inline affiliate disclosures at every product callout. - Editorial
Honesty pass on product recommendations
Softened "best decarboxylator on the market" superlatives across Ardent FX coverage to match our actual testing voice. Reviewing every product page for honest cons and "skip this if…" notes. - Content
Removed fabricated review ratings from recipe schema
RemovedaggregateRatingJSON-LD from cannabis Kool-Aid, drinks, and gummy recipes — we don't collect on-site reviews, so emitting them was structured-data spam. - Content
Page-level cohesion and hierarchy audit
Locked one consistent layout shape per page type (hubs, flagships, recipes, troubleshooting), de-duplicated stacked callouts, and standardized where the dose-handoff and safety floor render. - Site
Migrated to bethehippy.com with full 301/410 redirect map
Moved canonical to bethehippy.com, preserved every legacy URL with a true 301, and retired stale pages with proper 410s. - Content
Rewrote flagship titles in problem-language
Switched flagship titles from generic "X Guide" framings to the questions readers actually type — e.g. "How to Decarb Weed Without Killing Your THC," "How Many Drops of Tincture?"
Spotted something we should add or correct? File a correction — we read every one.