Marketing analytics inside your product.

Client results under the agency's own brand.

A marketing dashboard as one of your customers opens it, scoped to their own rows.
Try it

Experience Sumboard

Builder view and customer view.

Switch between the workspace where your team builds and the branded dashboard your customer sees.

View perspective
BuilderCustomer
  • 01Create data insights
  • 02Customize with your brand
  • 03Embed in your app
Loading interactive demo…
The problem

What goes wrong in Marketing dashboards.

The people reading these numbers are measured by them, so each metric is shown with its counting rule.

The people reading these numbers are measured by them. A metric label without its counting rule lets each team assume a different definition.

The people reading these numbers are measured by them, so each metric is shown with its counting rule.
Who sees what

Permission resolution at query time.

An agency sees its own clients, and each client sees only itself and usually does not know the agency's other clients exist. White-label is a requirement here: the wrong logo on an export is a commercial problem.

What gets measured

One metric definition, shared across teams.

Sessions

Visits in a period, with the dashboard stating whether internal and bot traffic are excluded. Tools count sessions differently.

Cost per acquisition

Spend divided by conversions for a stated window. Conversion windows differ per platform, so a blended figure states its window.

Return on ad spend

Attributed revenue against spend. The attribution model is stated with the figure; two models on the same data give different results.

Attributed conversions

Conversions assigned to a channel by a stated model. Last-click and data-driven models give different counts, so the model is named.

Failure modes

Three failure modes in production.

Metric names without their definitions

A card that says 'Conversions: 1,284' and nothing else leaves the definition to the reader.

A vendor's brand on a client-facing report

A report with a vendor logo on it shows the agency's client who the supplier is. Branded exports and sender domains are the requirement.

Blended totals with silent gaps

When one platform's data has not landed, a blended total reads low. Per-source freshness makes the gap visible.

Data freshness

How fresh the numbers need to be.

Campaign data arrives from several platforms on different schedules and some of it restates for days after the fact. The dashboard shows when each source last landed; a blended number is as fresh as its slowest input.

Reference

Further reading for Marketing engineering teams.

Frequently asked questions.

Can each client see only their own data?

Yes, and they do not see that other clients exist. Scope is resolved in the query, not in the interface.

Will our clients see your name anywhere?

No. The dashboard, the scheduled emails, the PDF and the spreadsheet export all carry your brand and your sending domain.

Can we show what a metric counts inside the dashboard?

Yes. Definitions are part of the model and are shown with the number.

Can clients schedule their own reports?

Yes. They set the frequency, the format and the filters themselves.

Other industries