Should I build my own ecommerce dashboard in Airtable, Power BI or custom code?

What the dashboard has to do first

Before you pick a tool, write down the five numbers you check every morning. For a multichannel seller they are usually sales by channel, profit after fees, stock that will run out soon, orders waiting to ship, and cash coming in. A tool is good only if it can show those five with data you trust. Most dashboards fail on the data, not the charts. For the metrics themselves, see my guide to an ecommerce profit dashboard.

The four routes at a glance

Route	What it is	Cost, as of November 2026	Main limit
Airtable	A database with forms and dashboard pages	Team $20 per seat a month billed annually ($24 monthly); Business $45 ($54 monthly)	50,000 records per base on Team, 125,000 on Business
AppSheet	Google's no-code builder that turns a spreadsheet or database into an app	Starter $5, Core $10 per user a month	2,500 rows per AppSheet database on Starter and Core
Power BI	Microsoft's reporting and chart tool	Pro $14 per user a month paid yearly; the desktop authoring app is a free download	8 scheduled refreshes a day on Pro
Custom code	A web app built for your shop	Build cost plus hosting	You pay for the build and its upkeep

Airtable

Airtable stores data in tables that look like spreadsheets but link to each other. It has dashboard-style pages called interfaces. It suits a catalogue, a supplier list or a small orders table you edit by hand. The free plan stops at 1,000 records per base and 5 editors. The Team plan raises that to 50,000 records, which sounds like a lot until you store one row per order line. A shop with 40 order lines a day fills it in about three and a half years, and that is before you add stock movements.

It does not pull Shopify or Amazon data by itself on a schedule at high volume. You connect it with sync features or a tool such as Zapier, which move data between apps. My guide on connecting Shopify to Airtable shows what that looks like. If you are weighing it against other databases, read the Airtable alternatives guide.

AppSheet

AppSheet builds an app from a Google Sheet or database, with forms, barcode scanning and simple charts. It is cheap per user, and you can test with up to 10 users at no cost. Its dashboards are basic, though. It is better for a warehouse screen where staff scan and update stock than for a profit view. The AppSheet database itself is capped at 2,500 rows on the Starter and Core plans, so for real order history you keep the data in Sheets or another source.

Power BI

Power BI is the strongest of the three for charts and slicing numbers. You build reports in the free desktop app, and you pay $14 a user a month on Pro to share them. Two limits matter for a shop. A Pro report refreshes up to 8 times a day, and the next level, Premium Per User at $24 a user a month, raises that to 48. So your dashboard is never live, it is a picture taken a few times a day. For a daily profit review that is fine. For a flash sale it is not.

The bigger cost is the work behind it. Power BI shows data, it does not collect it. Someone must load Shopify, Amazon fees, ad spend and costs into one clean table first. That job is the real project.

Custom code

A custom dashboard is a small web app that talks directly to your sales channels and your own database. It can refresh as often as the channels allow, hold as much history as you want, and show exactly what you decide on, nothing else. You can add logins, so a packer sees stock and an owner sees profit. The costs are the build and the upkeep. APIs change, channels add fees, and someone has to fix it when a feed breaks. My guide on building your own Amazon profit dashboard shows how much work one channel alone takes.

Refresh rates and upkeep: the part people skip

• Refresh. Ask how often the numbers update, and from where. Power BI Pro caps scheduled refreshes. Airtable depends on your sync. Custom code depends on how you build it.
• Row limits. Count your order lines per year, then compare with the caps above.
• Per-seat fees. All three charge per user. On Airtable, people who edit or comment are paid seats, while read-only viewers are not. On Power BI, people who view shared reports need their own Pro licence, so check who really needs one.
• Who fixes it. A DIY dashboard is only as good as the person who built it. If they leave, it decays.

A worked example with made-up numbers

A seller has three owners and a part-time packer, 120 orders a day, and sells on Shopify and Amazon. The three owners need profit views, and the packer needs a stock list.

With Power BI Pro for three owners that is $42 a month, with the packer on a free Google Sheet. Add a data tool or a few hours a week of someone's time to feed it. With Airtable Team for four people it is about $80 a month billed annually, and the order table passes 50,000 records in about 14 months at 120 order lines a day. Those figures are illustrations, so check each pricing page for today's rates.

When custom is the better choice

• You use three or more data sources and none of the builders connect them cleanly.
• You need a number that no tool calculates the way you do, such as profit after your own landing-cost rules.
• Your spreadsheet has become the business, and everyone is afraid to touch it. See turning an Excel sheet into a web app.
• You want different views for different staff, with real logins.
• You are paying many seats just so people can look at charts.

Not sure which of these fits? My overview of internal tools compares building paths in more detail.

How I'd approach this

Start in a spreadsheet with your five morning numbers and a month of real data. If you still check it every day after a month, move it into Power BI or Airtable. If you hit a limit, such as refresh speed, record caps or a cost you cannot model, that limit is your brief for a custom build.

That is where I come in. I build owner dashboards that pull from your channels, apply your own cost rules, and give each person the view they need, and you own the result. Describe the five numbers you check and I can tell you quickly whether a builder tool is enough.