What Vibition Taught Me About Ecommerce Research

What Building Vibition Taught Me About Product Research

Venture Step E011 captured me trying to restart Vibition, an ecommerce project I had cofounded before graduate school. I was excited to search for products again. I was also carrying a useful failure from our first launch.

We had discovered a product through a paid research service, treated the signal as an early advantage, and committed money to inventory. Then COVID disrupted the order. Delivery took longer than expected, competitors arrived, advertising assumptions changed, and the margin we thought we understood became much less forgiving.

We eventually sold through the inventory. I did not consider the launch a financial success. I did consider it an unusually effective education.

The dashboard was not the business

The original product looked promising because the visible signals pointed in the same direction. The category had activity. The keywords appeared useful. A research tool presented the opportunity in a way that made action feel justified.

The weakness was easy to miss. Other subscribers could see similar data. A shared research product can organize a market, but it does not reserve the opportunity for one seller.

That distinction became obvious after the order was placed. Our outcome depended on matters the discovery screen could not settle: supplier timing, international disruption, competing launches, advertising cost, price pressure, customer response, fulfillment expense, and the amount of cash trapped in inventory.

flowchart LR
    A["Discovery signal"] --> B["Product candidate"]
    B --> C["Customer evidence"]
    C --> D["Economics and differentiation"]
    D --> E["Small reversible test"]
    E --> F["Bounded inventory commitment"]

The diagram captures the process I wish I had used more explicitly. Discovery belongs near the beginning. It does not belong at the commitment gate.

FBA reduced work but did not remove the risk

In the episode, I repeatedly misnamed Fulfillment by Amazon. The correct name is FBA.

Amazon's current FBA overview describes a service that stores eligible seller inventory, picks and packs orders, ships them, and handles designated customer-service and returns work. That is a meaningful operational capability.

It does not turn the seller's list price into profit. Amazon says FBA costs depend on the product and services used. Its fee-estimation workflow asks for details such as dimensions, weight, category, price, shipping, and costs so a seller can compare Amazon fulfillment with seller fulfillment.

The seller still owns the product choice, sourcing, quality, claims, inventory funding, compliance, pricing, advertising, and downside. Convenience changes the operating model. It does not erase the business model.

The episode's discovery method needed a stronger second half

I talked about product-research tools, active advertisements, keywords, competitors, sourcing, seasonality, repeat purchase, and whether a product could grow into a brand. Those were useful questions.

The missing discipline was falsification. What result would make us reject the candidate? What margin floor did we require? How much could advertising rise before the model failed? What would happen if the order arrived late? Could the proposed difference survive a direct comparison? Which public claims could we prove before launch?

The FTC advertising substantiation policy explains that objective advertising claims need a reasonable basis before they are disseminated. That principle changes product research. A feature is not only a design choice. It may become a public claim with an evidence burden.

A brand cannot rescue weak economics

I ended the episode by asking whether the product could become a brand. I still think that is important, but I would state it differently now.

A brand is not a logo placed on a generic product after the order. It is the promise a company can make repeatedly, the evidence behind that promise, and the operating system that delivers it.

Vibition was interested in products with a sustainability orientation. That was an intention, not proof of an environmental benefit. The FTC Green Guides summary warns against broad, unqualified claims such as green or eco-friendly. A public benefit claim needs to be specific, clear, and supported.

The brand question therefore follows the evidence. Can the business deliver a meaningful difference? Can it support the claim? Can it do so at a price and cost structure that survives realistic conditions? Can the same customer trust it with a related problem later?

The commitment rule I carried forward

The most valuable lesson from Vibition is that product research should make the expensive decision harder to fool.

A candidate should earn commitment across customer need, demand, competition, differentiation, sourcing, unit economics, compliance, claim support, brand fit, and downside. The amount of inventory should reflect the quality of that evidence and the loss the business can carry.

That does not guarantee success. It creates a more honest relationship between uncertainty and risk.

The practical continuation is [[How to Evaluate an Ecommerce Product Before Committing Inventory]]. [[Build Ecommerce Unit Economics Before Ordering Inventory]] handles the financial model, while [[Make an Inventory Commitment Decision]] turns the evidence into a test, wait, reject, or commit decision.

This story was developed with AI assistance from the preserved E011 transcript and the linked Amazon and FTC records. Publication remains unauthorized pending founder, company, platform, financial, privacy, accessibility, and editorial review.