The five growth questions for ecommerce brands

Guide · Frameworks · 6 min read · last verified 2026-07-27

An ecommerce growth plan is a map of the shopper's journey with the leaks marked and owned. The journey has three visible stages — discover, compare, checkout — plus a quiet fourth after the order, and every stage now has an AI layer sitting on top of it. The five growth questions that govern any plan (the flagship five-question framework sets them out) still apply to a brand selling products, but the useful way to walk them is not question by question. For ecommerce, loss, creation, reach, execution and proof all attach to journey stages, not to departments — so this piece walks the journey and lets the questions surface where they actually live.

The journey is shorter than it used to be

A shopper who once moved through search results, listicles, and a dozen product tabs can now ask an AI shopping assistant for "a running jacket that packs into its own pocket and copes with rain" and receive named products with reasons attached. Marketplaces run their own recommendation layers doing the same compression. Discovery and comparison, once separate stages a brand could influence separately, increasingly happen inside a single generated answer — and the brand's own site is often the last place visited rather than the first. None of this removes the journey; it compresses it, and compression punishes brands whose presence was thin at any single stage.

Discover: where shoppers first meet the brand, or never do

Loss at the discovery stage is silent by construction. A brand absent from the assistant's answer, invisible in marketplace search for its own category terms, and unmentioned in the social threads where the category gets discussed has not lost a sale it can count — it has lost a shopper it never saw arrive.

What decides presence here is whether the brand's products are described anywhere in the language shoppers use. Assistants and marketplace engines assemble answers from product data, reviews, and content; a jacket whose packability exists only in a campaign video cannot be recommended for packability. The discovery-stage work is therefore an audit: list the questions shoppers ask about your category, check each surface — assistants, marketplace search, community threads — and record where the brand appears and where it does not. That recorded list is the ecommerce version of knowing where buyers slip away unseen, and it doubles as the channel decision, because the surfaces where your category is asked about are, by definition, where your buyers already are.

There is also a structural choice at this stage: marketplaces bring the audience but keep the customer relationship; the brand's own site reverses the trade. Most brands need both, but the growth plan should say which stage of the journey each one is being asked to win, rather than treating them as interchangeable shelves.

Compare: the product feed is the argument

At the comparison stage, the shopper — or the assistant acting for them — weighs attributes. What feeds that weighing is data the brand controls completely and often neglects completely: the product feed. Spec fields, materials, sizing guidance, care instructions, compatibility, availability. Every empty field is a comparison silently lost, because an engine cannot rank a product on a property it was never told about.

This reframes the creation question for ecommerce. The asset that most needs writing is usually not an article; it is enriched product data, buyer questions answered directly on the product page, honest comparison content between the brand's own lines, and visible responses to critical reviews. Reviews deserve particular respect at this stage: they are the one content type shoppers trust more than the brand, and assistants quote them. A brand cannot write its reviews, but it can earn them, answer them, and fix the product truths they keep repeating.

Checkout: the stage you control most and study least

Checkout is where the journey touches operations. On the brand's own site, the levers are friction and confidence: delivery promises stated plainly, payment options shoppers expect, stock truthfully displayed. On marketplaces, the lever is different — retail media, the paid placements that buy presence at the exact moment of decision inside someone else's store. Retail media spend is best understood through the journey lens: it rents position at the checkout end of the funnel, which makes it powerful for capture and useless for creating the demand that discovery feeds.

Execution discipline lives at this stage too, because checkout fixes are operational rather than editorial: a feed correction, a delivery-promise rewrite, a payment integration. Each needs one named owner and a date, exactly as any growth action does — the difference is that the owners sit in operations and engineering, and a plan written only for the marketing team will skip this stage entirely.

After the order: the journey that writes the next one

The quiet fourth stage feeds the first three. Delivered orders generate the reviews that comparison engines quote; post-purchase questions reveal the product-page gaps the next shopper will hit; repeat purchases tell you which discovery promises the product actually kept. Brands that treat the post-order stage as customer service only are discarding the cheapest research they own. The loop is literal: today's fulfilled order writes tomorrow's comparison content.

Where the five questions hide in the journey

Measuring a journey you only partly see

The final question deserves its own stage because ecommerce measurement is deceptively comfortable: orders, sessions and conversion are all countable, which tempts brands into measuring only the end of the journey. The starting stages need a different instrument — a recorded baseline of what shopping assistants and marketplace searches actually surface for the brand's core category questions, taken before the feed work and content work begins, then re-checked on the same prompts after. This is the job a market growth intelligence platform does for the surfaces above; Magrios runs that scan-and-benchmark loop so a brand can see whether the discovery answer changed, not just whether orders did. A useful first pass takes days, not quarters — getting value from AI research in the first week shows the shape.

Walked this way, the five questions stop being an abstract checklist and become a walk down the shopper's path: find the stage that leaks, fix what feeds it, prove the fix moved the answer. The brands that win the compressed journey are the ones who know it stage by stage — because the assistant answering for the shopper certainly does.