It’s Time to Dump Data-in-a-Box Companies Like Baremetrics and ChartMogul
Data-in-a-box companies are going out of style like bell bottom jeans and mood rings. There are easier ways to manage your data.
Have you ever looked at a report in Baremetrics or ChartMogul or SaaSOptics and realized the numbers are way off? Not just slightly off, or maybe a day out of date, or only “correct” if you make that One Special Assumption about netting out certain customers?
Back in 2019 when you were just a $50k MRR business, you may have been quite excited to hear about Data-in-a-Box solutions. The Data-in-a-Box pitch works well:
- Integrations to Stripe, Shopify, or other payments systems
- A set of pre-made charts and tables showing you your data
- No need to hire data engineers or analysts
- Relatively cheap to get started
Then, you bought the solution, got a set of reports teed up for you. Your Data-in-a-Box vendor solution worked okay for basic stuff, even if the dashboards were sometimes a big laggy or a day or two out of date.
The problem is, a year later your dashboards got worse. You started selling in multiple channels, which means your questions got more complex, your ad spend on multiple platforms drove business to multiple sales channels, and you tripled your MRR, which means you started paying even more (Baremetrics and similar charge you on MRR, they do have all your transactions data, after all) for dashboards and insights that are became further divorced from the realities of your business.
Nothing on the dashboards seems right, plus if you want an issue fixed your Data-in-a-Box customer service rep has offered to help, but only if you pay services fees for one of their analysts to address your problems.
Congratulations. You’ve outgrown your Data-in-a-Box solution.
You’re now actually paying 2–3x for high level insights that aren’t even correct anymore, and you can’t double-click on them to drill in or get deeper insights.
Worse, if you have operations people running their own reports off this data or spending ad budget on incorrect data from a Data-in-a-Box solution, you’re likely just lighting money on fire.
Even worse yet, most of these Data-in-a-Box solutions upsell customers on new report sets that are also based on MRR. Baremetrics offers a core ‘Metrics’ package, the ‘Cancellation Insights’ and ‘Recover’ and ‘Forecast’ as upsells.
As a data engineer and analyst for years I’m going to let you in on a little secret here: this is bullshit.
Once the data is loaded from the payments platform, these Data-in-a-Box companies are simply soaking customers by packaging these as upsells and then charging you on your MRR. These additional dashboards and insights packages are about 1 day of work to make, if that.
I’ve watched enough businesses get soaked by Data-in-a-Box platforms that don’t scale up that I’m mad. There is a better way, and no it doesn’t need to involve hiring large teams of data engineers and data scientists and analysts.
The problem with Data-in-a-Box solutions can be boiled down to this:
Your data is unique, but your data problems are not unique. Data-in-a-box platforms don’t address the uniqueness of your data, the specific nuances and exceptions and logic that is specific to your organization. Instead they only focus on high level data problems like MRR, churn, and things that any business would want to know. Quickly, value decays.
Introducing the Activity Schema: Answer the ‘Why’ Without a Large Data Team
Every single business can be broken down into:
- Activities, which are real-world events like a click on an ad or a purchase
- Entities, like customers or visitors or suppliers
- Time, which is the record of when an activity occurred
That’s the whole ballgame.
500 customers (entity) came back to the site (activity) in the last two weeks (time) after clicking (activity) on the Reengagement Email (entity) we sent last month (time).
Thinking about your business in terms of activities, entities, and time informs how you ask good questions about your business.
- Do customers who bought on Shopify make second purchases through a different channel?
- We know MRR in Baremetrics and similar Data-in-a-Box solutions, yes, but how much of that comes from customers making their third or fourth order?
- Are specific customers churning after they bought a specific product?
As your business becomes more sophisticated and you need to ask more sophisticated questions and tinkering with margins becomes more critical, Data-in-a-Box solutions quickly lose value, even as Data-in-a-Box companies start charging you more, as they charge you on MRR even as they contribute less to that MRR bump.
As you scale, your questions become about the impact of data held in one system on data held in another system, like the impact of emails in an ESP like Klaviyo on incremental revenue lift, which is held in your OMS.
This is basic funnel analysis, which is where most of these Data-in-a-Box companies fall flat since they’re only collecting your payments data. They fundamentally cannot answer questions of impact: how did x marketing campaign affect y revenue upsells.
Narrator, the company behind the Activity Schema, offers what is probably the easiest solution for Breaking Out of the Data-in-a-Box.
By simply combining everything on the backend into an activity, the timestamps of the activities, and an entity, anything can be easily modeled and presented, and any question can be answered across marketing systems, payments systems, and any other applications where data is housed.
Any new question is simply a matter of minutes to hours to answer — no going back and forth with a Data-in-a-Box Customer Success rep who has never heard of LTV, let alone how to make reports.
Plus, Narrator charges on platform and seats — if you double MRR in a year because you’re crushing it with customers, you won’t all of a sudden get charged extra simply because your MRR increased.
Bringing It All Together
In the last decade of cheap money, we’ve seen all sorts of data promises and snake oil.
The rise of the data scientist, the rise of data engineering, the rise of the analytics engineer, the rise of point solutions that say they do one thing and then charge you on your revenue that they had nothing to do with — these days are over and it’s time to get serious. Bad tools and expensive, bloated headcount teams are out.
All that has really been accomplished here is some people got new job titles and demanded higher salaries and made overly complex solutions.
Companies that will win in the next few years will not be those that care about the size or volume of data collected or the size of their data science teams, but organizations that get stuff done and answer basic questions quickly and accurately.
Consider dumping your Data-in-a-Box tools or your 10-person data science team for something that actually helps you move fast and answer your organization’s most pressing questions without all the jumping through hoops.
— — — — — — — — — — — — — — —
Lauren Balik is the Owner of Upright Analytics, a boutique data and strategy consulting firm focused on cleaning data messes and getting actual strategic work done without all the data hoopla and circus acts.
Lauren has dumped complicated bad tools and complicated data stacks in favor of Narrator and the activity schema.
Say hello: lauren@uprightanalytics.com
