General
Do You Actually Need a Data Stack? Honest Answer for Stores Under $500K
Somewhere around the $50K/month mark, most ecommerce owners start hearing the term "data stack."
It shows up in newsletters, on Twitter, in agency pitch decks. Snowflake. dbt. Fivetran. Segment. A whole vocabulary of tools that sound essential and expensive in equal measure.
Here's our honest answer: if you're doing under $500K/month, you almost certainly don't need one.
That's not a sales-friendly thing for us to say. We build data infrastructure for a living. But we've seen enough stores waste six months and $20K building something they never used to know that recommending it broadly would be dishonest.
What a "data stack" actually means
Strip away the jargon and a data stack is four things:
Connectors that pull data out of your tools automatically (Fivetran, Airbyte)
A warehouse where all that data lands in one place (Snowflake, BigQuery)
A transformation layer that cleans and models it (dbt)
A visualization layer on top (Looker, Tableau, Power BI)
It's genuinely powerful. It's also genuinely expensive — realistically $1,500 to $5,000/month once you include tooling and someone's time to maintain it.
The question isn't whether it works. It's whether the problem it solves is a problem you have yet.
The problem a data stack actually solves
A data stack solves scale and complexity, not clarity.
You need one when:
You have so much data that spreadsheets and native connectors physically can't handle the volume
You have so many sources that manual reconciliation takes days, not hours
You have multiple people who need different views of the same underlying data
You need historical data modeled in ways your source tools can't produce
Your queries are slow enough that people stop running them
Notice what's not on that list: "I don't know which ads are working." That's a clarity problem, and a data stack is a wildly expensive way to solve it.
What stores under $500K actually need
In almost every audit we run on a store in the $10K–$500K range, the real problem is the same: the data exists, it's just scattered and nobody has connected it.
Shopify knows your revenue. Meta knows your ad spend. GA4 knows your traffic sources. Klaviyo knows your email performance. Each one is right. None of them talk to each other.
You don't need a warehouse to fix that. You need:
One dashboard that pulls from all four via native connectors
Five to seven metrics that actually drive decisions, not forty that look impressive
A weekly rhythm where someone actually looks at it
Looker Studio does this for free. It connects to GA4, Google Ads, and Sheets natively, and to Shopify and Meta through inexpensive connectors. For most stores in this range, a well-built Looker Studio dashboard is 80% of the value of a full data stack at 2% of the cost.
The signals you're actually ready
We'd start seriously discussing a data stack when three or more of these are true:
Your Looker Studio dashboard is timing out or hitting row limits. You've outgrown what free tooling can handle. This is the clearest signal.
You have more than six data sources that genuinely matter. Store, ads, email, inventory, support, subscriptions, wholesale. At that point manual connectors start breaking constantly.
Multiple people need different views. Finance wants margin. Marketing wants CAC. Ops wants inventory velocity. When one dashboard can't serve all of them, modeling matters.
You need history your tools don't keep. Most SaaS platforms only retain 13–25 months. If you need three-year cohort analysis, you need to be storing that data yourself.
Someone owns data as part of their job. A stack without a maintainer becomes an expensive broken pipe within six months. This is the one people skip, and it's the one that kills projects.
What overbuilding actually costs
We audited a store doing around $180K/month that had spent eight months and roughly $30K building a proper warehouse setup. Fivetran connectors, Snowflake, dbt models, the works.
Nobody used it.
The dashboards on top were never finished, because finishing them was always the next quarter's project. Meanwhile the founder was still checking Shopify and Meta in separate tabs every morning, exactly like before, because that was the thing that actually worked.
The infrastructure was competent. The sequencing was wrong. They built the plumbing before establishing that anyone wanted the water.
The cost isn't just the money. It's eight months where the actual problem — no single view of performance — stayed unsolved while everyone was busy.
The honest sequence
If you're under $500K/month, do this in order:
Get one dashboard live with your four or five core sources. Looker Studio, free, a few days of work.
Cut your metrics down to the ones that trigger a decision. If a number changes and you don't know what to do about it, remove it.
Establish a weekly review. Thirty minutes, same time every week. This is the step everyone skips and it's the one that creates value.
Run that for six months. Find where it breaks.
Then, and only then, build for what actually broke.
Step 5 might be a data stack. It might just be a better connector. You won't know until steps 1–4 have told you.
Where this leaves you
If you're under $500K/month and someone is pitching you a data warehouse, ask them a simple question: what specifically breaks in a Looker Studio dashboard for a store my size?
If they can answer concretely, listen. If they answer with words like "scalability" and "future-proofing," you're being sold infrastructure for a problem you don't have yet.
Build for the stage you're at. Upgrade when you outgrow it.
Not sure which stage you're actually at?
Book a free 30-minute data audit with lebombo. We'll look at your current setup and tell you honestly whether you need to build anything at all.
Want to work it out yourself first?
The Data Stack Blueprint inside our checklist maps the right tools to your revenue stage — what to use at $10K/month vs $500K/month, and what to deliberately skip.
[Get the Ecommerce Data Stack Checklist — $37 →]
