with the skill
**Step 3 is probably turning an easy signup into a meal-planning chore.** The jump from entering a ZIP code and choosing a household size to comparing 40 meals is substantial—especially when people must make five decisions before they can proceed.
That’s a strong hypothesis, not a diagnosis: check whether the drop-off also reflects mobile usability, slow loading, dietary mismatches, or unclear pricing.
### The likely psychology—and what to change
| Mechanism | Why it fits your flow | What to try |
|---|---|---|
| **Choice overload / Hick’s Law** | Forty meals means comparing ingredients, preferences, prep times, and more. More variety can make the decision harder rather than more appealing. | Start with **5 recommended meals**, with an obvious “Swap” action and “Browse all 40” option. |
| **Activation energy** | Choosing the first meal starts a task that looks like it will take work. The effort arrives before customers have experienced the service’s value. | Offer **“Start with our recommended menu”** alongside “Choose my own.” Make the starting point easy, not obligatory. |
| **Regret aversion** | Customers may worry about wasting money on meals they—or someone in their household—won’t enjoy. | Show useful decision cues: ingredients, prep time, dietary fit, and genuine customer ratings. If changes are allowed later, explain **exactly until when**. |
| **Blocked progress** | “You can’t continue until you pick five” makes customization feel like a gate rather than a benefit. | Show **“3 of 5 meals selected”**, preserve selections, and provide **“Fill remaining spots with recommendations.”** |
| **Jobs to Be Done mismatch** | Customers may be hiring you to *reduce meal-planning work*, but signup asks them to do meal planning immediately. | Position customization as control they can use—not work they must complete. Test a curated-menu path. |
### My first test
**Keep step 3, but change it from “build a menu” to “review a menu.”**
- Display five recommended meals
Show the rest (20 more lines)
prominently.
- Let people swap any meal or browse the full menu.
- Use a clear CTA: **“Continue with these 5 meals.”**
- Make the selected meals and total price clear before payment.
- Don’t silently assume dietary suitability; offer easy filters and show ingredient/allergen information.
This combines **choice reduction, lower starting effort, and less fear of choosing badly**, while retaining customer control.
### What to measure
**Don’t judge the test only by step-3 completion.** Track completed paid signups, plus first-box cancellations, refunds, and meal satisfaction. An easier signup isn’t a win if people end up with meals they don’t want.
Before testing, inspect:
- **No meals selected:** suggests difficulty starting, weak relevance, or a usability issue.
- **One to four selected:** suggests difficulty finishing, insufficient suitable options, or the five-meal requirement.
- **Repeated swaps or long browsing:** suggests comparison difficulty or uncertainty.
A few observed signup sessions can help distinguish these explanations quickly.
**Bottom line:** sell the relief of having dinner handled; don’t require customers to prove they can plan five dinners first.