Overview, Acquisition & Media, Profit, Customers, Products and Email, built against your own data overnight and grouped by the conversations your team actually has. Ask for another in plain English and it joins them.
Figures throughout come from a sample store modelled on real brand patterns. The gaps they illustrate are real; the sizes will be different on yours.
Most reporting tools section themselves by where the data came from: a Meta page, a Google page, a Klaviyo page, a store page. That's the four-versions problem rebuilt inside one product: four tabs, four definitions, four answers to the same question. Truefold groups by the conversation the numbers are for instead, so the argument about whose number is right happens on one dashboard, on purpose, rather than everywhere at once.
Growth, profit, efficiency, and where the money went, at total level. No tabs: the moment a glance has tabs it becomes a fourth place you have to go and look.
Efficiency · By channel · Platforms & campaigns · Attribution reality
P&L · Cost structure · Discounting · Margin by channel · What's measured, what's assumed
The base today · Cohorts & retention · Acquisition quality · Repeat behaviour
It isn't a daily dashboard and doesn't pretend to be.
Margin by product · Mix & basket · Returns & risk
Campaigns and flows kept apart, because a blended email number is the average of two different things.
Six is the whole list, and it's meant to be. Every prebuilt surface is one more thing we have to keep honest, so the seventh is the one you ask for. It's built from the same metrics and sits beside the rest.
Most weekly marketing meetings run the same way. Five minutes on the top line. Then each channel owner takes their turn and answers the same three questions: what worked, what didn't, what they think is going on. Then somebody decides where next month's budget goes. The numbers are meant to be ready beforehand, and usually someone is still assembling them while the first person is talking.
Acquisition & Media is built in that order.
Is the engine working, blended? Efficiency after discounts, acquisition cost, spend and revenue, each against the previous period. Blended on purpose: this tab is the five minutes before anyone gets specific.
Who's carrying, who's dragging, and where the budget should move. Efficiency and contribution margin for the same channel, side by side, one attributed and one not.
Where a manager actually operates, and what they'll be asked about: campaign-level performance, spend by platform, where budget is capping delivery.
The one every meeting has and nobody wins. The platforms claim more conversions than Shopify credits them, and the channel split doesn't explain all the revenue. Both figures are on the tab. Truefold shows the disagreement rather than picking a side and asking you to trust it over the platforms.
It's ready before the meeting because it's ready every morning. Nobody assembles it, and nobody is still assembling it at ten past.
Describe what you want and Truefold builds it: tabs, panels, filters wired across all of them. It draws on the same library the six are built from, so a panel you asked for and a panel that shipped mean the same thing by revenue.
What it won't do is invent a number to satisfy the request. If the ask is ambiguous it asks. Some things can't be answered at all. Ask for a site conversion rate and nothing in the four sources counts sessions, so it tells you that, and what it would need to answer it.
“A weekly view for the paid team: efficiency after discounts, contribution margin by channel, and where spend is capped by budget.”
You get the tabs, the panels, and one channel filter driving all three.
Change it by asking.Ask for one edit and you get one edit. Truefold compares the result against what you had and puts back anything you didn't mention: a panel that came back different, a chart that quietly went missing, a tab label that was set and came back empty. We watched that last one happen during an unrelated edit. Every save is a version with a name on it, and you can read what changed and roll back to any of them.
Rename a dashboard, rewrite its description, move panels. None of it needs anyone's help.
There's no drill-down menu and no filter maze. Select a panel, open it in chat, and it arrives with its context already loaded: the metric, the date range, the filters that were applied. Nothing to retype, nothing to rebuild.
From there it's a workbench rather than a search box. Ask why the line moved. Cut it another way. Chart it against last month. Keep what's worth keeping. The same works from a Signal: anything that tells you something changed opens into the place you can find out why.
A good dashboard's job is to make you ask a question. Most tools stop exactly there.
Panels are built on tested data models. When something underneath breaks, Signals raises it before the chart carries on rendering as though nothing happened. Broken charts are the easy problem, though. The hard one is a chart that renders perfectly and quietly means something other than its title.
Ask for net profit by channel and you don't get one. Rent isn't caused by paid search, and splitting it by revenue makes a channel look worse in the month it sold more. Contribution margin by channel takes the slot instead. That's the honest answer to the question underneath.
Blended efficiency split by channel stops being blended. It has quietly become channel ROAS, which is a weaker number and a different one. So the panel is labelled channel ROAS, not efficiency.
None of this makes a dashboard prettier. It makes it quotable, which matters when the number goes in front of somebody who will ask where it came from.
Truefold pulls; it never writes. Nothing in your store, your ad accounts or your flows changes because you connected us.
One BigQuery dataset per organisation, used by nobody else. Customer names, email addresses, phone numbers and shipping addresses are dropped before anything is written.
Rename metrics to your team's language, correct anything that doesn't match how you run the business. Version-controlled, with a name against every change. We don't train AI on your business data.
Sign in to Shopify, Meta Ads and Google Ads, paste a Klaviyo key. All six are populated by the next morning, on your own numbers.