In November 2021, Bodybuilding.com shut down BodySpace. Years of workout logs, progress photos, and personal records went offline. Some users had downloaded their data. Most hadn’t.
In 2020, Under Armour sold MyFitnessPal to Francisco Partners for $345 million after acquiring it for $475 million in 2015. The acquisition went through despite a 2018 data breach exposing 150 million accounts. Users of premium features watched the app’s roadmap shift. Some walked away. Most stayed because moving 5 years of nutrition logs was harder than tolerating the new direction.
These aren’t unusual stories. They’re the default story for any closed platform that holds your data.
The pattern in fitness coaching tools
Online coaching software follows the same pattern. The coach starts with 3 clients on Trainerize. By client 22, all of those clients’ programs, logs, photos, and notes live inside Trainerize’s database. The coach doesn’t have a local copy. The client doesn’t have a local copy.
Trainerize stays in business. The data stays accessible. Until it doesn’t.
The risks aren’t hypothetical:
- Pricing changes (Trainerize’s tier restructure in 2023 raised costs 20-40% for many tiers)
- Acquisitions (ABC Fitness Solutions acquired Trainerize in 2021. Direction changed)
- Deprecations (TrueCoach removed several reporting features in 2024 in favor of “premium” tier)
- Platform shutdowns (BodySpace, MyPlate, several smaller coaching SaaS over the past 5 years)
Every coach with 3+ years of client history on a closed platform has a continuity risk. Most underestimate it because the platform is currently working fine.
What “your data” usually means
When a SaaS says “you own your data” it usually means:
- You can export your data (sometimes to a useful format, sometimes to PDF only)
- The data still lives on their servers (you have a copy, they have the original)
- Pulling the data out doesn’t replicate the system that uses it (your exports are not functional; they’re just data)
This is fine if everything goes well. It breaks when the platform changes direction in a way you don’t agree with. By the time you decide to leave, migrating 22 clients’ historical data to a new system is a 40-hour project. Most coaches just stay.
What real data ownership looks like
Real data ownership has three properties:
- The data lives in a system you control. Not “exports available” but “the canonical copy is on hardware or cloud storage you own.”
- The data is in a portable format. Open formats (Excel, CSV, JSON) that any tool can read. Not proprietary database dumps that require the original software to interpret.
- The data continues to function without the platform’s middleware. You can open a sheet without our app and it still makes sense. The formulas are visible. The structure is human- readable.
This is harder to engineer than “we’ll let you export.” It requires the system architecture to treat your data as the source of truth rather than a derivative.
Why we built Coach Sheet on Google Sheets
When we mapped the design space, the question wasn’t “should the data be portable” but “what’s the simplest way to make portability the default.”
Google Sheets is the answer because:
- The coach’s Google account owns every file
- The format is open (xlsx export with one click)
- The system continues to function without us (Apps Script can be disabled and the sheets still work as plain spreadsheets)
- The structure is visible (formulas, sheets, columns are all human-readable)
If Coach Sheet shuts down tomorrow, every coach using it has every client’s full history sitting in their Google Drive. They can open any sheet, see the data, see the formulas, and continue using the spreadsheet manually.
This isn’t a hypothetical fallback. It’s the design contract.
What this costs us
Data-ownership-first design has tradeoffs we accepted:
- We can’t build a multi-tenant analytics dashboard across all coaches’ data (we don’t have it)
- We can’t run aggregate ML models on client outcomes (we don’t have access)
- We can’t sell anonymized data to researchers (we don’t have it)
- We can’t easily add features that require central database joins (each coach is a silo)
These tradeoffs are intentional. The product is a coaching tool, not a data platform. We make money from coaches paying $15/month for the integration and support, not from monetizing the data underneath.
The Helpful Content System angle
Google’s Helpful Content System (rolled out 2022, integrated into core ranking 2024) explicitly penalizes content that exists primarily to extract data, attention, or money from users without giving back proportionate value.
The same principle applies to product design. A coaching tool that exists primarily to lock the coach’s data into the platform is the product equivalent of zero-value content.
A coaching tool that exists primarily to make coaches better at coaching, and treats the coach’s business as an independent entity, is the product equivalent of helpful content.
Coaches notice the difference even if they can’t articulate it. The platforms with high churn are usually the ones with high lock-in. The platforms with low churn are usually the ones that made it easy to leave.
What you can do today
If you’re using a closed coaching platform:
- Export everything to your hard drive or your own cloud (Dropbox, Google Drive, OneDrive) monthly. Set a calendar reminder.
- Maintain a parallel “master” sheet outside the platform with your client roster, current programs, and key metrics. This is your insurance.
- When evaluating new tools, ask: where does the canonical copy of the data live? If the answer is “in our platform,” weight that risk against the benefits.
- If you’re starting fresh, choose tools that don’t lock you in. (We hope that’s us; if not, pick something with explicit open-format guarantees.)
The cost of data lock-in is invisible until the platform changes direction. By then, you’ve already paid it.