CSD Matrix — E-Commerce Dashboard

Certainties, Suppositions, and Doubts from discovery research

Certainties — we know that…
Users & behavior
Users primarily access the dashboard on desktop devices
Users create their own dashboards because the default is not usable
The dashboard is the most-frequented UI in the product — it's the first thing customers see
The high configuration burden leads to abandonment and underutilization
Product & data
MyView is being used as an unintended workaround, not for its designed purpose
The default dashboard cannot be edited — but this is not clearly communicated
Data discrepancies exist between the old and new dashboards, and users trust the old one more
Item-level and marketplace-level insights are not currently available
Empty and missing channels appear in the revenue graph, causing confusion
Business context
Competitors like Sellerboard offer faster setup and more actionable product insights
Some users have already moved to external tools to fill the gap
Creating even a basic KPI widget is extremely difficult for non-technical users
Suppositions — we assume that…
Users & behavior
Users check the dashboard in the morning as part of a daily routine
Users want fewer configuration options — the current breadth adds complexity without value
A well-structured default dashboard would reduce or eliminate the need for MyView
Most users across segments (finance, ops, sales) share enough overlap to be served by a single dashboard view
Product & data
Heavy MyView usage is a significant contributor to slow load times
Fixing the default experience will restore enough trust to bring users back from external tools
Global filters (timeframe, marketplace, brand) will cover the majority of filtering needs without per-widget customization
Progressive disclosure — summary first, detail on demand — maps to how managers actually work through their day
Business context
Item-level profitability insights are the highest-value addition for the core persona
Channel-level filtering is a prerequisite for the dashboard to be useful to multi-marketplace sellers
Integrations (Google Ads, Shopify, etc.) are important but secondary to getting the core data model right first
Doubts — we don't yet know…
Users & behavior
Whether trust in the dashboard can be rebuilt quickly, or whether a transition period alongside the old version is needed
How often managers actually return to the dashboard intraday vs. once in the morning
Whether user needs across segments diverge enough to require separate views, or whether one dashboard with filters is sufficient
Which KPIs are truly universal vs. specific to a user segment or merchant type
Product & data
Whether data discrepancies are a front-end display issue or a deeper pipeline problem — and who owns the fix
What the actual performance impact of removing or replacing MyView would be
How frequently data needs to refresh to feel "live enough" for daily operational use
Whether the new data model can support item-level cost consolidation without significant backend work
Business context
Which integrations are table stakes for launch vs. post-MVP
How widespread the pain is — whether it's concentrated in power users or felt broadly across the customer base
Whether a better default experience is enough, or whether some level of personalization will always be expected