Good NumbersCustom ecommerce dashboards, with an AI analyst your team can ask.
We build dashboards around the decisions your team makes every day, on top of your own data warehouse, with logins per role and an AI assistant that answers questions from your verified numbers.
- Built on your own warehouse
- Read-only AI with the query shown
- Fixed quote before work starts
Off-the-shelf dashboards show everything except what you need.
Most dashboards show every metric a tool can produce. Your team still opens five tabs and a spreadsheet to answer the questions that matter each morning.
- Built for the tool, not the team
- Generic layouts that ignore how your team actually decides on budget, creatives and stock.
- Slow on real data
- Reports that time out or sample once you have a few years of orders and ad history.
- Numbers nobody trusts
- Dashboards on top of unreconciled data repeat the same disagreements in a nicer layout.
- Questions wait for an analyst
- Anything outside the dashboard becomes a request, and an answer days later.
What's included
Decision mapping
We start from the questions each person answers every day and design the views around them, not around the data.
Custom dashboard application
A fast web dashboard in your brand, built on your warehouse, that stays quick with years of data.
Logins per role
Each person sees what they need: finance, growth, creative and leadership views on the same numbers.
Creative and campaign views
Ad-level performance with previews, so creative decisions happen next to the numbers.
AI analyst
An AI assistant connected read-only to your warehouse. Ask in plain English and get an answer with the query it ran, so it can be checked.
Alerts
Slack or email alerts when a metric moves outside its normal range or a data source stops updating.
Looker Studio option
Where a custom application is more than you need, a well-built Looker Studio dashboard on the same warehouse.
Hosting and handover
Deployed in your cloud account with documentation, so you own the code and the data.
How it works
- 1
Scope
We map the decisions, users and data sources, and confirm the warehouse layer the dashboard needs.
- 2
Fixed quote
A written plan with screens, roles, milestones and a fixed price before anything is built.
- 3
Build in milestones
Data first, then the dashboard, then logins and the AI analyst, each reviewed by your team.
- 4
Launch and support
Handover with documentation, plus optional monthly support for new views and sources.
Common questions
Anything else, ask in the free check form or email hello@goodnumbers.io.
Should we use Looker Studio or a custom dashboard?
Looker Studio is the right choice for most brands that need standard reporting. A custom dashboard is worth it when you need logins per role, creative previews, an AI assistant, or speed on large datasets.
How accurate is the AI analyst?
It only answers from your warehouse, has read-only access and shows the query behind each answer, so anyone can check it. Answers are as good as the data underneath, which is why the warehouse and reconciliation come first.
Is our data used to train AI models?
The assistant runs through a commercial AI API with read-only access to your warehouse. We set it up so your data stays in your own cloud account and is only sent to answer the questions your team asks.
Do we need a data warehouse first?
Yes. A dashboard is only as reliable as the data behind it. If you do not have one, we build it as part of the same project.
How long does it take?
Typically three to five weeks including the warehouse, depending on sources, roles and the number of views.
Related guides
What is a good ROAS for a Shopify store? (with a break-even calculator)
A good ROAS depends on your margins, not an industry average. Work out your break-even and target ROAS in a minute, then check it against the right number.
AnalyticsHow to find your true ROAS across Meta, Google and Shopify
Platform ROAS, blended ROAS, MER and break-even ROAS explained with worked numbers, plus a weekly scorecard that shows what ad spend really returns.
AttributionWhy Meta Ads revenue doesn't match Shopify (and which number to trust)
Meta almost always reports more purchase value than Shopify. The 8 reasons why, how to measure the real gap, and which number to use for which decision.
Other services: Server-side tracking setupGA4 audit for ShopifyMeta CAPI setupBigQuery data warehouse
Plan your data build.
Tell us which sources you use and which decisions you want the data to drive. We'll reply within one business day with questions, a proposed architecture and a fixed quote. No call needed to start.
Plan a custom data build