The short version
The price increase needs to happen. It serves four jobs at once — margin lift, premium-brand signal, anchor pricing for the diffuser launch, and rebuilding the MER math without forcing acquisition to do all the work — and aluminum getting deferred took out the only other lever in Move 2.
The marketing team's concern is fair: raising prices into a soft consumer environment is risky. But the data says the risk is smaller than the upside, and the right execution makes it smaller still.
The thesis in five lines:
- The "price test" question has two answers, not one. A/B testing (random visitor split) is the cleanest measurement but slow and limited. A staged before/after rollout — what the Plan already specifies — is what actually moves the business. Run both, in that order.
- Shopify and Amazon are different problems. Shopify supports real price A/B testing via Intelligems or similar; Amazon doesn't (you can A/B content, not price). Different toolkits, different success metrics.
- We already have a natural experiment in the data. Listed prices crept from $14.16 to $15.31 across 2025 (+8%) and same-SKU volume fell ~6%. Naive elasticity is −1.2 but the true number is closer to −0.5 to −0.8 once you strip out the acquisition collapse. That's enough headroom to do the $16 → $20 move with confidence.
- Premium positioning is the dissatisfaction-mitigant. Customers tolerate (and often prefer) a price increase that's narrated. "We're charging more because the bottle is heavier, the certifications are deeper, the diffuser is coming" is a different conversation from a stealth hike.
- Test on Shopify first, then mirror on Amazon. Shopify gives clean read-outs; Amazon doesn't. Don't flip Amazon list prices until the Shopify test has answered the elasticity question.
Where Grow sits today — the numbers that bound the test
Pulled from the parquet store, May 19, 2026. These are the constraints the playbook is reconciled against.
The accidental price experiment we already ran
The data already gave us a quasi-experiment. Listed prices on hero scents stepped up around August 2025 (most went up 6–11%). Same-SKU year-over-year, the evergreen-spray basket moved like this:
| Metric | Q1 2025 | Q1 2026 | Change |
|---|---|---|---|
| Avg gross unit price | $14.50 | $15.22 | +4.95% |
| Units sold (Shopify, evergreen sprays) | 13,591 | 12,800 | −5.82% |
| Discount intensity | 7.01% | 4.63% | −2.4 pts |
| Naive elasticity | −1.18 |
Naive elasticity of −1.18 is the headline. The honest read on it: the volume drop was substantially upstream of price — new-customer acquisition fell 48% YoY in the trailing 11 weeks (Canopy assessment, Turnaround Narrative). Strip that confound and the true elasticity for evergreen sprays at this price band is likely in the −0.5 to −0.8 range. That's normal for a premium-branded fragrance with no good substitute, and it matches the category research: premium candle/fragrance shows low per-unit price sensitivity and high promo sensitivity.
What this means for the $16 → $20 move (+25%): at −0.7 elasticity, expect ~17% volume drop, +13% gross revenue per existing customer, and +38% contribution dollars per unit sold. At −1.0 (pessimistic), volume drops 25%, gross revenue is flat, contribution dollars per unit still up ~25%. The worst plausible outcome of a clean test is contribution-dollar neutral with brand-signal upside.
The other reassuring data point: the 90-day repeat rate for spray cohorts didn't degrade when prices crept up. Sept-Oct 2025 cohorts (acquired at the new $15+ price) posted the year's highest repeat rate at 19%. Premium customers stayed loyal. The customers we lose at $20 are mostly the ones we'd struggle to retain anyway.
Strategic frame — A/B test or roll forward?
The first question to settle, because the marketing team's "are we testing this or are we just doing it?" question is the right one.
What "price A/B testing" actually means
True A/B price testing means showing different prices to different visitors at the same time, randomly assigned, on the same product page. It is the gold standard for measuring elasticity because it controls for everything else — seasonality, traffic source, creative, weather. On Shopify it requires a third-party app and (for Plus stores) checkout scripts; for standard stores it runs on Shopify Functions.
The catch: it's slow and it's controversial. Slow because you need statistical power, which for a brand at Grow's traffic (roughly 60–80K monthly Shopify sessions) means 4–8 weeks per test before you can read a result. Controversial because some customers can detect they're being shown a different price than their friend, and a vocal minority will raise it on social. The exposure risk is real but small in practice — Intelligems and Shoplift have run >$500M of tests across DTC brands without a meaningful PR blowup pattern.
What "roll forward" means
Change the price for everyone on a defined date, monitor the metrics against same-period-prior, and decide at day 30 / 60 / 90 whether to proceed, hold, or roll back. This is what the Plan already specifies ("$16 → $20 listed on three hero scents · 90-day elasticity test"). It's faster, simpler, and politically easier — but it's confounded by everything else moving in the world.
The recommendation: do both, in this order
The 90-day calendar
- Weeks 1–2 (now): Set up Intelligems (or Shoplift) on Shopify. Build the price ladders, configure the test, write the price-page copy variants.
- Weeks 3–8: First A/B test. Test $16 vs $20 on three hero scents simultaneously, 50/50 split. Read profit-per-visitor, not just CVR.
- Weeks 9–12: Decide and roll. If $20 wins on profit-per-visitor (likely), roll $20 to all evergreen sprays. If it doesn't, test $18 vs $20 to find the inflection point.
- Weeks 13–24: Confirmation period. Run the rolled price and watch same-period YoY against the baseline. Layer on the brand voice and packaging story.
- Q4 2026: $22 secondary test on hero scents only, once the diffuser launch has anchored Grow's premium positioning at $100+.
The A/B comes first because it's cheaper to be wrong in a 50/50 split (you lose half a quarter of contribution on the losing arm) than to be wrong on a full rollout (you eat 100% of the downside until you reverse). The roll-forward follows because A/B alone doesn't generalize — Intelligems' own data shows test winners often don't fully repeat at full traffic, mostly because A/B-arm sample is biased toward repeat visitors. The two methods together are how you measure and capture the lift.
Tactical — Shopify
Tooling: which app, and why
| Tool | Monthly cost | Strength | Verdict |
|---|---|---|---|
| Intelligems | $499 (Plus plan; price testing is gated to this tier) | Profit-aware (pulls COGS), the category standard, $500M+ GMV tested, integrates with Shopify Functions for non-Plus stores | Recommended Pay the $499 for one quarter. You earn it back in the first test if it lands. |
| Shoplift | $300+ for price testing add-on | Strong on theme/content; price testing is a paid add-on | Backup Similar profile, no decisive advantage for Grow's situation. |
| Elevate AB | $99/mo | Budget option; 80% of Intelligems' features at 20% of cost | Consider If we want to validate Intelligems before committing to $499/mo, run a smaller test here first. |
| Split | $10/mo (free under 10 orders) | Cheap, basic | Underweight Not enough statistical infrastructure for a real price test at Grow's scale. |
| Native Shopify A/B | — | Doesn't exist for prices. Shopify offers theme/content split testing only. | No |
Recommendation: Intelligems for the formal test, $499/month for 3 months ($1,500 all-in). If the first test lands the expected lift, the tool pays itself back roughly 20x in year one on the spray category alone.
The four Shopify price-testing patterns — when each is right
Random visitor split, same SKU, two prices
What: Intelligems shows visitor X the $16 page and visitor Y the $20 page, 50/50 random, persistent across the session and cart.
Best for: Reading clean elasticity on a single SKU. The first test we should run.
Risk: Customer detection if a household has two devices on the same product. Low risk in practice; mitigate by setting the cookie at the household level and being honest if asked ("we're testing pricing — here's the lower price").
Recommended for Grow: Yes — Test 1. Run on Woodland Sage, Blondewood, Vanilla Peppermint (top three by revenue).
Different prices by state or region
What: Use Shopify Markets to set $16 in some states and $20 in others. Common in DTC; not controversial because customers don't compare across states.
Best for: Validating elasticity in a way that's more durable than A/B (markets don't see different prices on the same page). Also useful if you want to test by income demographics (NY/CA vs MW/SE).
Risk: Customer detection if a buyer ships to a different state than they live in. Coastal-pricing optics if framed as a "blue state tax."
Recommended for Grow: Backup. Only if Pattern A reads inconclusive after 6 weeks.
Grandfather existing customers; raise on new only
What: Klaviyo segment-aware checkout. Returning customers keep their $16 price for the next 60 days; new visitors and lapsed (>9mo) see $20.
Best for: Maximum dissatisfaction mitigation. The marketing team's worry is overwhelmingly about retention; this insulates the retention base.
Risk: Operational complexity. Requires Intelligems Plus + Klaviyo integration. Hard to maintain past 90 days — eventually you have to bring everyone to the new price.
Recommended for Grow: Yes — as the rollout posture, not the test. Once the A/B confirms $20, communicate to the existing list as a "thanks-for-being-with-us" 60-day grace period at $16. The data says repeat customers use discount codes 2x more than new ones, so this is a real benefit, not a fake one.
Raise the single, hold the bundle (or vice versa)
What: Single 5oz spray goes to $20. The 3-pack stays at $48 (effective $16/unit), making it the obvious value. Or invert — raise the bundle so the single looks better.
Best for: Increasing AOV. Already at $58 (Q4 2025). At $20/unit single and $48 bundle, the math nudges bundle-buying.
Risk: The discount code data shows existing customers already use bundle pricing heavily. Doubling up on bundle benefit risks training customers to wait for bundle pricing.
Recommended for Grow: Yes — layer on, not lead with. After single goes to $20, restructure 3-pack at $48 (3×$16, framed as a bundle save of $12). Increases AOV and softens the per-unit price perception.
What to measure, in priority order
- Profit per visitor (contribution-dollar terms). Not CVR. A 12% CVR at $16 making $13 contribution vs a 10% CVR at $20 making $17 contribution — the second wins. Intelligems reports this natively when COGS is loaded; load the $3.23 spray COGS before running.
- First-order AOV. Did the higher single price push customers into bundles or out of the order entirely? Watch for unit-count-per-order, not just dollars.
- Repeat rate at 30 / 60 / 90 days. The retention question. The 2025 natural experiment says this didn't move; confirm.
- Return rate. Premium-priced products sometimes see lower return rates (buyer's commitment); sometimes higher (regret). Read both directions.
- Brand sentiment surrogate. Klaviyo unsubscribe rate, organic review velocity, support ticket volume mentioning price. These are leading indicators of dissatisfaction.
Do not measure top-line revenue alone for the first 30 days. Same-store revenue can swing on traffic; the right metric is contribution per visitor or per session.
Tactical — Amazon
Amazon is a different problem. The short version: you cannot A/B-test prices on Amazon. Amazon's "Manage Your Experiments" tool supports A/B testing on title, main image, bullets, description, and A+ content — but not on price, because Amazon's marketplace doctrine is one-price-per-buyer. The actual mechanisms available are list-price changes, coupons, Prime Exclusive Discounts, Subscribe & Save, and Brand Tailored Promotions. Each does something different.
The Amazon situation today, from our data
The Q1 2026 Amazon spray numbers tell an uncomfortable story.
| Metric | Q1 2025 | Q1 2026 | Change |
|---|---|---|---|
| Amazon spray revenue | $199K | $143K | −28% |
| Amazon spray units | 13,344 | 22,210 | +66% |
| Effective unit price | $14.92 | $6.43 | −57% |
Amazon has been trading dollars for volume. Heavy coupon and promo usage on multi-pack ASINs is the most likely driver. Volume is up dramatically, revenue is down. This is the opposite of where we want to go on Amazon — and it's why Amazon price strategy needs more attention than Shopify, not less.
The Amazon playbook
Stop the coupon bleed
Audit every active coupon, Subscribe & Save discount above 10%, and Prime Exclusive Discount on spray ASINs. Per Amazon's March 2026 fee changes, coupon redemption now incurs a per-redemption fee on top of the discount — making heavy couponing materially less efficient than it was in 2024. Target: pull aggressive discounts down to ≤10% on all single ASINs by end of Q2.
Raise list price on hero ASINs to $19.99
Amazon's algorithm prices the Buy Box based on listed price, not coupon price. Raising list while leaving a small Subscribe & Save discount intact gets you:
- $19.99 list displayed (premium signal)
- $17.99 effective via 10% S&S (matches Shopify $20 net of typical loyalty)
- Stronger Buy Box position vs the prior $14.99 list (lower listed prices get lower Featured-Offer weight in 2025+)
The recommended cadence: change list price for 14 days, measure unit velocity vs the prior 14-day baseline, repeat at 28 days. This is the closest legitimate substitute for A/B on Amazon.
Replace coupons with Brand Tailored Promotions
BTPs are free to run (no per-redemption fee), they target specific customer segments (cart abandoners, repeat buyers, brand followers), and they don't show up in the public listing. This shifts Amazon discount spend from "anyone on the platform" to "people who specifically engaged with Grow." Better economics, better targeting, no Buy Box damage.
Build Subscribe & Save as the anchor
S&S subscribers convert at 5–15% Amazon-funded discount with much lower churn than coupon-driven first-time buyers. The diffuser refill plan (Move 3 of the Plan) makes this even more important — S&S on sprays is the conditioning that makes S&S on diffuser refills feel natural. Grow's current S&S penetration is essentially zero in our data; this is unworked headroom.
What we are not doing on Amazon
- Not running price A/B. The tooling isn't there; the closest substitute is the time-window before/after described above.
- Not raising price beyond $19.99 until Shopify is at $20. Amazon list price should track Shopify retail with no premium in either direction, per Amazon's price-parity surveillance. Diverging risks suppression.
- Not testing on Amazon before Shopify. Lower data quality, slower read, no controllable variant. Shopify reads first.
The dissatisfaction question — what mitigates the marketing team's concern
The marketing team is right to be cautious. Tariff-driven price increases across DTC since 2024 have produced real customer fatigue. The question isn't whether to worry; it's whether the worry justifies not raising prices, given the alternative is continued structural losses on a 58% COGS base.
The dissatisfaction mitigants are well-documented in DTC literature and confirmed by Grow's own behavioral data.
Narrate the price increase, don't stealth it
The brands that absorb price increases worst do it silently. The brands that absorb them well — Native, MALK, Graza in recent memory — pair the increase with a visible story about why. Grow has three story tracks ready: deeper certifications (USDA + ISO + ASTM, with EWG/MADE SAFE/B Corp coming), the diffuser launch raising the brand's overall price ladder, and the ongoing investment in clean-ingredient sourcing. None of these are inventions — they're things we're already doing. The increase is the visible signal that the brand is moving up-market.
Required: One on-site banner, one welcome-flow update, one short founder note from Dan to the existing list explaining the move. Katelyn owns. Estimated effort: 6 hours.
Grandfather existing customers for 60 days
Pattern C above. Anyone who has bought in the prior 12 months gets a one-time code that holds the $16 price for their next two orders, valid for 60 days. Costs us 2–4% of contribution dollars during the window, returns the goodwill in full, and the data shows the grandfathered cohort is the one that converts to repeat at the highest rate anyway. Returns far more than it costs.
Pair the increase with a visible product upgrade
The aluminum bottle has been deferred — but the bottle isn't the only premium signal. A higher-quality cap, a thicker glass spray bottle, a refreshed label, or a sealed gift box on first orders all communicate "the product got better, not just more expensive." Even a $0.30–$0.50 packaging upgrade signals premium far more than the dollar cost suggests. Whitney and the packaging vendor can land this in 60 days without aluminum's 9-month lead time.
Anchor against a worse alternative
The bundle becomes the deal. At $20 single / $48 three-pack ($16/unit effective), the message becomes "save $12 when you stock up" rather than "we raised prices." Customers who bundle don't perceive the increase at all because their effective per-unit cost is unchanged. AOV up, perceived value up.
Hold the entry-tier sample pack
The 2oz discovery pack stays at its current price. New customers still have a cheap way in, which protects the acquisition funnel. The price increase happens at the 5oz tier where commitment is already implied. This is the standard premium-brand structure (Aesop sample sizes, Le Labo discovery sets, Branch Basics starter kits).
What we do not do
- Apologize in the email. The brand voice imperative (Strategic Plan, "bolder but still quiet") explicitly calls for a confident posture. The note from Dan is direct, not apologetic.
- Roll back the move at the first negative review. Expect 5–10 negative comments per 1,000 customers. They're disproportionately visible but not representative. Pre-commit to a 90-day window and read the aggregate signal.
- Lean on the meta ads team to spin this. The performance channel should not carry the brand narrative. The narrative is on-site, in email, and in founder content.
The 60-day recommended plan
What the data and the research support, made concrete.
Get the tooling and the story ready
- Install Intelligems on Shopify ($499/mo Plus plan). Load COGS for every spray SKU. Owner: Luis / Katelyn.
- Audit Amazon coupons + S&S discounts. Pull anything >10% on single ASINs. Owner: whoever runs Amazon.
- Draft three short customer-facing assets: on-site banner, welcome-flow update, Dan's founder note. Owner: Katelyn.
- Identify three hero scents for the test (recommended: Woodland Sage, Blondewood, Vanilla Peppermint — top three by Shopify revenue, all evergreen).
Run the $16 vs $20 test
- 50/50 random visitor split across three hero scents. 4-week minimum.
- Primary metric: profit-per-visitor (contribution-margin terms).
- Secondary: CVR, AOV, 30-day repeat rate.
- Check-in at day 28: if profit-per-visitor delta is ≥10% in favor of $20 with p<0.05, the test wins and we go to rollout. If inconclusive, extend to day 42.
Move the price, ship the story, mirror to Amazon
- If $20 wins: roll $20 to all evergreen spray SKUs. Restructure 3-pack at $48. Issue 60-day grandfather codes to the prior-12-month customer list.
- Publish the on-site banner, send Dan's note, update welcome flow copy.
- Amazon: raise list to $19.99 on three hero ASINs. Replace any coupon ≥10% with a Brand Tailored Promotion. Hold for 14 days, then mirror across the rest of the catalog.
- Set up the 90-day post-rollout watch: same-period YoY revenue, repeat rate by cohort, return rate, support ticket volume.
What "success" looks like at day 90
- Contribution dollars per Shopify visitor up 15–25% versus the pre-test baseline.
- Unit volume down 10–20% (within the elasticity envelope). Anything beyond −25% is a warning.
- Repeat rate (90-day) within ±2 points of the 14–19% baseline. Anything below 12% triggers a comms review, not a price reversal.
- Customer support volume mentioning price < 1% of tickets after week 4.
- Amazon contribution dollars reversing the Q1 2026 negative trend by week 8.
What triggers a rollback
Rollback should be rare and expensive — the willingness to rollback is itself a signal that the brand is not confident in its position. But there are two genuine triggers:
- Volume drops >30% with no offsetting profit-per-visitor lift. This indicates elasticity is much worse than the data suggests; either the price point is wrong or the story is wrong.
- Repeat rate drops >5 points within 60 days of the rollout. This is the durable retention signal — the only signal worth reacting to.
Everything else — social commentary, individual review complaints, a meta CPM spike — is noise. The 90-day window is held.
Open questions for Dan + marketing team
- Subscribe & Save on Shopify. Currently essentially zero. Worth standing up before or after the price test? The recommendation is after — let the price be the primary signal, then add the loyalty mechanism.
- The 60-day grandfather code. Reactive to email list only, or also to prior Shopify customers who never opted in? The wider net costs more but signals more goodwill. Recommendation: wider net.
- Founder note tone. The brand voice imperative says "bolder, not louder." The note from Dan should be 200 words, direct, no apology. Katelyn drafts; Dan revises.
- Amazon Subscribe & Save target. What's the right S&S penetration goal for the spray category by Q4? Recommended: 15% of Amazon unit volume.
- Diffuser launch interaction. Should the spray price increase land before, during, or after the diffuser launch? Recommendation: before (Q3), so the diffuser at $100+ has the spray at $20 as its on-brand anchor.
Appendix — sources and verification
Internal data
- fact_orders, dim_products parquet tables — May 19, 2026 pull, 5oz spray category, Shopify and Amazon channels.
- Plan, Move 2 (
dashboards/Plan.html): committed direction of $16 → $20 → $22. - Strategic Plan, Workstream C (
Strategic_Plan.md): aluminum decision date originally July 2026; per Dan's May 19 update, deprioritized in favor of price action. - Turnaround Narrative (
Grow_Turnaround_Narrative.md): the "price increase is not about margin" framing — premium positioning is the primary asset.
External research
- Intelligems pricing, plan tiers, and Shopify Functions integration (intelligems.io, Shopify App Store).
- Shopify price-testing tool comparison (Shogun database, Convert.com Shopify price-testing guide, LAUNCHTIP 2026 A/B testing roundup).
- Shopify Markets B2B + multi-market pricing (Shopify Help, Absolute Web 2026 brief).
- Amazon Manage Your Experiments scope (Amazon Seller Central) — content A/B only, not price.
- Amazon Pricing Strategy 2026 — Buy Box dynamics, S&S, coupon fee changes (Trellis, Feedvisor, NivoAds March 2026 coupon-fee update).
- DTC price-increase research — eComAmplify, ATTN Agency tariff-impact briefs, L.E.K. Consulting on rising CAC, Beauty Independent on McKinsey's "price-fueled growth is over" finding.
- Premium fragrance/candle elasticity context — IndexBox US scented candles market analysis; Retail Brew premium-vs-mass beauty convergence (Aug 2025).
Self-review
The weakest assumption in this playbook is the elasticity estimate (−0.5 to −0.8). The naive read from the 2025 natural experiment is −1.2, which is genuinely worse than what I'm carrying forward. The reason I'm comfortable: the trailing 11-week new-customer collapse (-48%) is the dominant driver of the volume drop, not the price step. But if the A/B comes back at −1.0+, we should revise to a more graduated rollout ($16 → $18 first, then $18 → $20) instead of the single step. The 90-day calendar is built to surface this before broad rollout, which is the whole point of doing A/B before roll-forward.
The second weakness is the Amazon recommendation. We've never run a controlled list-price change on Amazon, so the "raise list to $19.99 with S&S anchor" plan is more directional than tested. The Step 1 (audit coupons) is a no-regret action regardless. Step 2 (raise list) should be done on one hero ASIN first, watched for 14 days, then mirrored.