Meet Toby: The AI Analyst Inside Your Shopify Dashboard

The problem with dashboards

Here's a scenario most Shopify founders know well. You open your analytics dashboard on a Monday morning. Revenue is down 12% week over week. You can see it in the chart — the line goes down and to the right. You spend the next two hours clicking through tabs: orders by channel, cohort retention, ad spend by campaign, traffic sources. By the end, you have a lot of additional data and no clearer answer. Was it the new creative? The Meta CPM spike? A product going out of stock? A competitor promotion? The dashboard showed you everything. It told you nothing.

This is the fundamental limitation of the dashboard-as-product model. Dashboards are excellent at visualising data. They are, by design, neutral. They present. They do not interpret. The interpretation — the hardest and most valuable part — is left entirely to you, your gut, and however much time you have on a Monday morning.

For large brands with dedicated analysts, this is fine. An analyst can dig into the data, form hypotheses, run the regressions, and come back with a coherent narrative. But most Shopify brands don't have a dedicated analyst. They have a founder, maybe a marketing manager, both of whom are already doing twelve other things. The dashboard is sophisticated. The bandwidth to interpret it is not.

Toby is our answer to that gap.

What Toby is and what it can do

Toby is Nuso's AI analyst. It lives inside your Nuso dashboard and has full context on your store's data: orders, revenue, cohorts, ad spend, channel performance, inventory, and customer behaviour. You talk to it the way you'd talk to a brilliant analyst who happened to have memorised every row of your database.

You type a question in plain English. Toby reads your data, reasons about it, and responds with an answer — not a chart, not a redirect to a different dashboard tab, but a written response that explains what's happening and why, with supporting data points.

Toby can:

• Diagnose revenue trends — explain why revenue went up or down in a given period, with specific contributing factors from your data
• Analyse cohort performance — compare customer cohorts by acquisition channel, creative, or period and surface the ones behaving differently
• Evaluate channel efficiency — assess which paid channels are performing above or below their historical efficiency, and flag emerging divergence
• Interpret MER movements — explain why your blended MER has shifted and which variables are most likely responsible
• Surface inventory risks — flag products trending toward stockout based on recent velocity, or identify slow-moving SKUs tying up cash
• Answer ad hoc questions — "Which product had the highest repeat purchase rate last quarter?" "What percentage of November customers have bought again?" — the kinds of questions that would otherwise require a custom report

Toby in action: example conversations

What makes Toby different from generic AI

You might reasonably ask: can't I just paste my data into ChatGPT and ask the same questions? In theory, yes. In practice, the experience is completely different — for a few important reasons.

Generic AI assistants don't know your store. They don't know that your Midnight Serum went out of stock on Wednesday, or that your January cohort skewed toward discount creative, or what your typical weekly revenue baseline is. They can reason about data you give them, but you have to give it to them, formatted correctly, every single time. Toby already has all of it. Every conversation starts with the full context of your store, without you having to do any data wrangling first.

Toby also understands the specific domain of Shopify analytics and DTC ecommerce. It knows what MER is, what a healthy cohort retention curve looks like, what typical CPM seasonality patterns are. Generic AI has broad knowledge but not domain depth. Toby is calibrated specifically to the questions Shopify brands actually ask.

Finally, Toby's answers are grounded in your actual data. It doesn't hallucinate figures — when it cites a number, that number comes from your Shopify or ad platform data, pulled in real time. You can verify any figure Toby gives you by clicking through to the relevant dashboard view.

What Toby can't do (and won't pretend to)

We want to be honest about Toby's limits, because we think that honesty makes it more useful, not less.

Toby is excellent at pattern recognition within your historical data. It is not a crystal ball. When it says "your January cohort's repeat rate is slightly below pace," that is a data observation, not a guaranteed forecast. Toby will flag trends and surface likely explanations, but the interpretation and the decision remain with you.

Toby also won't replace your judgment on brand and creative strategy. It can tell you which creative type has historically generated higher-LTV cohorts from your data. It cannot tell you whether a new creative concept is right for your brand voice, or whether a price increase will affect how customers perceive your positioning. Those are human questions about brand strategy that require context Toby doesn't have.

And Toby can't tell you things your data doesn't contain. If you haven't connected your Google Ads account, Toby can't analyse Google performance. If your UTM parameters are inconsistent, Toby's channel attribution will reflect that inconsistency. Toby is only as good as the data it has access to — which is why Nuso prompts you through a complete data connection setup before you start using it.

Where Toby is headed

Toby is live inside Nuso today, and we're continuing to expand what it can do. In the near term:

• Proactive alerts: Toby will ping you when it detects an anomaly worth investigating — a MER dip, a stockout risk, an unusual cohort divergence — without you having to ask first
• Weekly store briefs: A Monday morning summary in Toby's voice: what happened last week, what to watch this week, and what looks like it needs a decision
• Budget scenario modelling: "What would happen to MER if I increased Meta spend by £5,000 per week?" — directional answers based on your historical response curves
• Creative feedback loop: Connecting Nuso's AI Studio (where you generate product and ad creative) to Toby's cohort analysis, so Toby can tell you how specific AI-generated creatives are performing against LTV benchmarks over time

The broader vision for Toby is an analyst that gets more useful over time, not less. As Nuso accumulates more of your historical data, Toby's pattern recognition becomes more accurate and its baseline context becomes richer. A Toby that has watched your store for 18 months understands your seasonality, your cohort norms, and your channel dynamics in a way that a fresh tool never could.

We built Nuso because we believe Shopify brands deserve analytics that actually helps them make decisions — not just data that looks impressive in a dashboard review. Toby is the most direct expression of that belief. It's the analyst in the room that most brands have never been able to afford.

Give it a question you've been wondering about your store. You might be surprised by the answer.