Product case study · Chris Duty
Changes to Wild Alaskan's paid signup flow
Thank you for taking the time to review this case study. I spent about 5 hours in total researching, thinking, and producing this artifact. I look forward to discussing my process in more detail. I can be reached at chrisduty@gmail.com or 512-550-4669.
This case study was completed without knowing key details, including key metrics, unit economics and operational constraints.
This review covers the paid landing page and the signup flow through checkout. The replica shows each screen as it is live (Before) and with the changes below (After); a toggle on every screen switches between them.
This does not include a review of the member app or the post-purchase experience.
1.What was looked at
3,128 public records, collected Nov 1 to 3, 2026 with Apify scrapers and analyzed in Python. Wild Alaskan: 716 Trustpilot reviews (2020 to Nov 2026), 349 App Store reviews (Apr 2025 to Oct 2026), 195 Reddit posts and comments. Comparables: ButcherBox (600 Trustpilot, 298 Reddit), Vital Choice (600 Trustpilot), Sitka Seafood Market (137 Reddit), Maui Nui Venison (119 Reddit plus on-site reviews). Plus 29 active Meta ads, the 7 landing pages behind them, and a read-only walk of the signup flow.
Trustpilot samples are stratified by star, so shares are reported within low-star and within high-star reviews, not across all reviews.
2.What the reviews say
- Detractors complain about things software controls. Wild Alaskan's one- and two-star reviews over-index, against ButcherBox and Vital Choice, on box contents and variety (18% vs 8% and 14%), renewal and cancellation control (16% vs 13% and 3%), portion weight (11% vs 2% and 9%), and reminders and notifications (6% vs 4% and 3%). They under-index on damaged packaging (11% vs 19%) and carrier problems (7% vs 12%).
- The cancel problem is being solved; the contents problem is not. Cancel and renewal complaints in low-star reviews fell from 9 of 39 (2024) and 10 of 60 (2025) to 2 of 26 so far in 2026, which lines up with the members-only app and in-app plan editing. Contents complaints did not fall: 6 of 39, 11 of 60, 7 of 26.
- Promoters praise the fish and the habit. Reweighted, 54% of all Trustpilot reviews praise quality or freshness; the App Store set is 331 five-stars of 349. In order: taste and freshness vs grocery fish, individually sealed portions, human service by text, recipes, provenance, and convenience for people far from good fish.
- Price is argued on both sides. Promoters say “pricey but worth it.” Detractors do the math: “$32 a pound,” “$14 a serving.” Price is 8% of low-star reviews. The value case is won or lost on whether the box matches what the member pictured.
“Since they won't disclose the ratio, I was worried the expensive sablefish would be underrepresented while the cheaper cod and pollock did most of the heavy lifting. Nothing here gave me confidence otherwise.”
“I changed the frequency three days before my product shipped. Apparently it didn't take.”
“Wild caught salmon really does taste much, much better than farm raised... and more nutritious too!”
3.What the ads say
All 29 active ads lead with a discount or a free item. The value proposition underneath is three lines: wild-caught and never farmed (83% of ads), delivered to your door (62%), processed in the USA (28%). Recipes and cooking confidence appear in none. No ad and no landing page shows a price.
The seven landing pages on get.wildalaskancompany.com test offers and heroes; none tests price per portion, box contents, household fit, recipes or the app.
4.What the comparables do differently
- Sitka Seafood Market sells the fisherman. “Real fish. Real fishermen.” One landing page with a grocery-store comparison table, six verified-buyer testimonials, and the offer as the closer, not the headline. Members can see and swap items in each box before it ships.
- Maui Nui Venison sells its supply limits instead of hiding them. Membership is framed as early access to limited restocks that often sell out. Its landing pages show exactly what is in the box (2 sirloin, 2 medallions, 1 loin filet, about 5 lb), the price, and the member perks. Only 12% of its ads mention a discount. Wild Alaskan's boxes are also supply-driven; the difference is that Maui Nui says so on the way in, while Wild Alaskan says “selections vary based on availability” three times on one screen.
- ButcherBox puts the price in the ad and defends it: “Why is ButcherBox worth $179 per month?” Its loudest 2026 thread after packaging is about removing weight transparency: “they no longer tell you the weight of the meat you are selecting.” Four-year members cancelled over it.
5.Suggestions, by screen
Each change is documented below and on the prototype. The After screen shows a coral bar to the left of all suggested changes.
Landing page (Free Salmon for a Year)
- Price in the hero
One line under the free-salmon sentence: "Boxes start at $139. Shipping included."
Reason: No ad and no landing page shows a price. Price is 8% of low-star reviews.
- Trustpilot rating in the hero
Stars, 4.8, 11,240 reviews and the Trustpilot mark under the hero button, linked to Wild Alaskan's Trustpilot page.
Reason: 4.8 on 11,240 reviews is the strongest asset the brand has, and the live hero shows none of it. The review carousel is five screens down.
AssumptionBuilt as a static copy of Trustpilot's micro-combo layout. The real TrustBox widget needs the company's business unit ID.
- Why wild salmon
A new section after How It Works with three sourced facts: leaner with more protein, no added color, and fisheries managed to last. A member line closes it.
Reason: Wild-caught is the #1 claim in the ads (83%) and the page never says why it matters. Wild vs farmed is the only one of the three ad claims members repeat unprompted in reviews. Protein and fat figures are from USDA FoodData Central (FDC 2684440 and 2684441).
- Captain Cuts weight
Under "Free in your first 12 boxes": "Each pack is about [X] oz, [N] portions." Both values are shown as placeholders until confirmed.
Reason: Portion weight is 11% of low-star reviews vs 2% at ButcherBox. The free item is the ad's headline and the page never says what it weighs.
AssumptionPack weight and portion count are unknown and shown as placeholders.
Step 1: Build Box
- Step names say what each step does
The step indicator on every funnel page reads Build Box, Preferences, Checkout instead of Box Type, Build Box, Checkout.
Reason: Box and size are now chosen together on step 1, and step 2 is about schedule and reminders.
AssumptionNo data behind the labels themselves.
- Starting price on every card
Under each box name: "Starting at $139". Prices are flat across the three boxes on the live site, so the line is the same on all three.
Reason: The live step 1 shows no price; the only price is in the cart icon. Price is 8% of low-star reviews.
- Sizes and prices once a box is picked
After a box is picked, the three sizes appear under the cards: 8 portions $139, 12 portions $169, 24 portions $299, each with its price per portion, and "5 to 7 oz portions, shipping included." 12 is the default and marked Most Popular, as on the live site. Before a pick, only the cards and the button show.
Reason: The 8-pack is 40% worse per portion than the 24-pack ($17.38 vs $12.46) and nothing on the live flow says so. The species list stays behind Learn More, as on the live site.
- How often, stated up front
One line under the sizes: boxes come every month, every 6 weeks or every 8 weeks, chosen on the next step and changeable any time.
Reason: Renewal and cadence control is 16% of low-star reviews. The live flow defaults to monthly behind a text link.
- Button and cart say where you are going
The button reads "Continue to Delivery Preferences". The cart drawer shows the box, size, price and this week's list for the chosen box.
Reason: Box and size are both chosen here, so the next step is preferences, not size.
AssumptionNo data behind the button wording.
- Phones: the pinned button says what you picked
The live phone flow pins the Continue button to the bottom of the screen. Here the pinned bar adds one line above the button: box, size and price, e.g. "Wild Combo Box · 12 portions · $169". Before a box is picked, the button is off and reads "Choose a box to continue".
Reason: On a phone the box cards and sizes scroll out of view while the button stays on screen, so the bar is the one place the choice can stay visible.
AssumptionNo data on how often phone visitors scroll back to check their choice.
- More of each step above the fold
Smaller step headings and tighter spacing on every /join page. At 1440 x 900, step 1 shows all three cards and their names, and Preferences shows the box strip, the three schedule options and the reminder fields without scrolling.
Reason: Layout only.
AssumptionNo data; the aim is that the choice on each step is visible without scrolling.
Step 2: Delivery preferences
- What you chose, at the top
Title becomes "Delivery preferences", with a strip under it: Wild Combo Box · 12 portions · $169, and an Edit link back to step 1. The size cards move to step 1.
Reason: The live page shows the size cards again with no memory of the box chosen on step 1.
AssumptionNo data on how often members lose track of their choice.
- How often, visible, with the math
Every month, every 6 weeks and every 8 weeks shown as three options, no Change Schedule link. Under each: what the chosen size works out to, e.g. "12 portions every month is about 3 fish meals a week for one person, or 1 to 2 for two." Monthly stays the default.
Reason: Cadence control is 16% of low-star reviews. 8% of low-star reviews mention being one person, a couple, or a full freezer vs 3% of high-star. The live page's own FAQ says a monthly 12-pack is 3 to 4 meals a week per person.
AssumptionThe live labels are Every 1.5 Months and Every 2 Months. 6 and 8 weeks are the same intervals in words people use.
- Reminders, asked for here
"How do you want to be reminded?" Email and mobile number, with "Email me" and "Text me" both checked, "We recommend both." under them, and one sentence: a reminder 7 days before each box with what's in it, and changes allowed up to 3 days before it ships.
Reason: Reminders and notifications are 6% of low-star reviews vs 4% and 3% at the comparables. "Changes that did not take" and "too late to cancel" are the two most common cadence complaints. Capturing email and phone here also means a visitor who leaves before payment can be followed up, which the live flow cannot do until checkout.
Assumption7 days and 3 days are assumptions; the live lock time is unknown.
- A reminder needs somewhere to go
If Email me is checked and the email is empty, or Text me is checked and the number is empty, Continue does not advance. The page scrolls to the reminder section, outlines the empty field in red, and says under it: "Add an email to be reminded, or uncheck Email me." (or the same for text). Filling the field or unchecking the box clears it. With both unchecked, the visitor continues with no error. Works at every screen size.
Reason: Reminders and notifications are 6% of low-star reviews. A reminder promise with no address behind it is the "changes that did not take" complaint in another form.
AssumptionChecks only that the field is filled, not that the address or number is valid.
Step 3: Add-ons
- Add-ons default to one time
"Add to this box" is the main button and "Add to every box" the second. The subhead states the rule.
Reason: The live default makes every add-on recurring, with a one-line notice. Recurring surprise charges recur in the low-star reviews and the Reddit threads.
Checkout: delivery info
- Email and phone carried over
Email and mobile number from the Preferences step appear here filled in and editable.
Reason: They were already given on step 2.
AssumptionNo data on drop-off from re-entry.
- One clear reminder promise instead of one bundled checkbox
The reminder block from Preferences, restated: email and text 7 days before each box, change or skip up to 3 days before it ships, delivery updates only. One optional box: "Also text me offers." Terms acceptance is its own sentence under the button.
Reason: The live page bundles order updates, promotions, calls, terms and arbitration into one checkbox and one paragraph. Communication noise is 6% of low-star reviews.
Assumption7 days and 3 days are assumptions; the live lock time is unknown.
- The next box, on the summary
The order summary adds a Next box block: the next charge date for the chosen schedule, the price, and "Reminder 7 days before."
Reason: The live summary shows the first ship week and nothing about the second charge, which is the one people complain about: "I was sent a box even though I pushed it back."
AssumptionThe next charge date assumes the first charge is today.
- Phones: the order summary first, in one line
On phones the Order Summary sits above the delivery form, collapsed to one line, "Order summary · $169.00", that opens on tap to show the plan, the total, the ship week and the next box.
Reason: The summary now carries the next charge and the reminder promise. Stacked in full above the form it would push the form down about a screen; one line keeps both in reach.
AssumptionThe live phone checkout was not captured; Before shows the desktop order stacked, form first.
6.Questions I would ask the team specific to this use-case
- The box photo on step 1 is the same closed box on all three cards. How did the team land on these images, and what other visuals have been tested?
- What is the lock time before a box ships, and where is it shown to members today?
- What does a Captain Cuts pack weigh? The free item is the headline of the longest-running ad and the page never says.
- Which operational constraints rule out showing the species ratio before the box ships?
7.Assumptions and unknowns
Every number on the After screens that is not from Wild Alaskan's own pages or the research:
- Reminder timing, 7 days before each box. Assumption.
- Change window, up to 3 days before a box ships. Assumption; the live lock time is unknown.
- Captain Cuts pack weight and portion count. Unknown; shown as placeholders on the landing page.
- Every 6 weeks and every 8 weeks. Assumption that these match the live Every 1.5 Months and Every 2 Months. Strictly, 1.5 months is about 6.5 weeks and 2 months about 8.7.
- Fish meals a week. Assumption: one portion is one person's meal, and a month is 4.33 weeks.
- Next charge date. Assumption: the first charge is the day of signup.
- Checkout fields are not validated in this prototype, so the flow can be tested end to end without entering real information.
- Trustpilot rating in the hero. A static copy of the micro-combo layout. 4.8 and 11,240 are from the live page; the real TrustBox widget needs the company's business unit ID.
8.Method
Apify actors collected Trustpilot, App Store, Reddit and Meta Ads Library data. Python did the theme tagging and the reweighting by star. Claude helped with the analysis, the design and the build. The replica is Next.js on Vercel. The full process log is 02_process_and_tools_log.md, available on request.
Replica built for a product case study. Not affiliated with Wild Alaskan Company. Photography, logos and copy in the Before state are the live site's, used here for a private case study. Fonts are open substitutes.