image of an office collaboration scene (for a mobility and transportation)

Aha-moment roadmap

BookSell's revenue model is 90% driven by revenue sharing from retailers.

Generating the first consistent sales is retailers' aha-moment to stay on a membership.

Retailers showed urgency to install the app yet struggled to generate their first sales.

The task

My goal was to get retailers to their "aha moment" quickly while improving overall revenue and retention metrics. This involved identifying and removing key adoption blockers, and conducting customer discovery to understand retailers needs for product-market fit.

‍

Key Results:
- Median time to first product listing ≤ 30 minutes and time to first sale ≤ 20 days, with 10% of new stores reaching 5 sales within 30 days.
- Improved MAU/total ratio (books synced or sold) to keep churn ≤ 5%.
- 20 average items sold per active store.

Actions taken

Research: Conducted customer discovery with leads and current customers to map JTBD and blockers by segment.
- Most indie booksellers had the intention to sell online, but their stores were in early stages and didn't know how to build an acquisition engine.
- Some of the early performers were not booksellers but added books as upsells.
- Our free tier (10% commission only) made it impossible for retailers to invest in paid acquisition, generating conversion but blocking adoption.

Mapped the path to aha and created retargeting events for each step.

Hypotheses: There were 2 simple routes to explore.

We could help the indie bookseller segment by removing their blockers, including finishing optimizing their store for conversion (creating collections,, organizing navigation, adding cart abandonment retargeting) and building their acquisition engine (ad accounts setup, campaign design, content generation, and performance-based iteration).
- These services would create a new revenue stream and could be potentially turned into a product.
We could change our target audience, focusing on websites with high-traffic and active acquisition engines, but this would require an API-based product.

Experimentation: Although an API-based product made sense long term, we tested our first hypothesis by offering potential services to assess interest.

We had discovery calls with 10% of our free-tier members.
We sent an email campaign announcing new services. The first attempt had a 5% CVR.

Proposed solutions:

Change pricing: Remove the free tier, with only 10% commission, and start with a Basic plan that triples retailers' margins (30%) and allows them to invest in paid acquisition. A 30-day free trial would allow them to reach the aha-moment before spending any money.
Even though book publishing wasn't an issue (retailers averaged ~100 books synced), accelerating this process was necessary for them to start selling as soon as possible:
- Guided onboarding: connect eCommerce store, select the first 3 books (bulk import available) and publish the first product.
- Product tour with quick details about the basic navigation tabs.
- Email retargeting based on key drop-off moments (store, select, publish, process)
- Documentation and video tutorials.

Quick Design & Marketing Services tab with a Calendly CTA.
The first marketing services were executed manually, helping us understand the process and retailers' expectations. Our value proposition was around a cheap and quick way to solve their gaps (7/10 of quality, but allowing retailers to make any tweaks to the deliverables).

‍

Mid-term MVP:

AI design & marketing service automation: I proposed a shared interface that both the retailer and our team could use. We would use Airtable Interfaces for data input and control, and n8n as the workflow engine:
- Retailer connects Meta Ads and eCommerce credentials (stored for API calls), pulling branding.
- Retailer selects the product to promote or picks a brand topic.
- BookSell provides n8n with instructions and best-practice prompts so ads match brand and policy.
- n8n uses OpenAI/Claude/VEO3 to generate copy and assets based on BookSell prompts and retailers input.
- Retailer and BookSell can optionally review creative campaigns.
- n8n calls Meta API to create campaigns and ads. Product data informs guardrails such as max CPC.
- Retailer confirms and publishes the campaign since it involves spend.
- n8n ingests Ads performance data to pause underperformers and shift budget to winners over time.
- Best performing assets and settings are written back to the system to create a lightweight learning loop.