“Basedash is the data analyst we never had to hire. It replaces an entire full-time employee for us.”

Parallel is the TikTok distribution network for brands. Instead of relying on a single brand account, Parallel manages a network of more than 10,000 organic accounts that publish content for clients hundreds of thousands of times every month — accounting for roughly 0.2% of all daily TikTok posts. The result is organic-style content delivered at the scale and consistency of paid advertising, at a fraction of the CPM. Parallel powers programs for large CPG brands and consumer applications.

Operating that kind of network produces an enormous amount of data. Every account, post, template, sound, and brand campaign has its own metrics, and any of them can be the lever that makes a campaign work or quietly stop working. Co-founder and CEO Peter Solimine needed a way to interrogate all of it without pulling engineers off product or building yet another internal dashboard.

“Our entire business is content moving through accounts. If you can’t see what’s happening at the post and account level, you’re flying blind.”

Before Basedash

Parallel’s production data lives in Postgres on Supabase. Behind every viral post is a chain of structured records — the account that published it, the template it was built from, the sound it used, the client it ran for, the views and conversions it generated. Useful data, but not useful at all if the only way to query it is to hand-write SQL or file a ticket with engineering.

In the early days, that’s exactly what happened. Most non-trivial questions about network performance went through engineering, which meant most questions never got asked. The cost of finding out was higher than the cost of guessing — a familiar trade-off at any operationally heavy startup, and a costly one when your business is built on continuous experimentation.

Why Basedash

Peter pointed Basedash at Parallel’s Postgres database and immediately had a conversational interface over the same data powering the distribution platform. There was no schema documentation to write, no dashboards to scope, and no engineering project to greenlight — just questions and answers.

“I can ask Basedash anything about the network and get an answer in seconds. The kinds of questions that used to feel too expensive to ask — how are accounts in this group performing this week, which templates are driving views for this client, where are sound URLs failing — those just happen now.”

Today, Peter runs regular conversations across the business, and a growing portion of the team uses it for their own questions across product, ops, and account management.

What Parallel actually asks Basedash

The questions Parallel runs through Basedash are specific to the operational reality of running a large distributed content network:

• Content performance. Which templates, sounds, and posts are driving the most views and conversions, sliced by client, account group, time of day, and platform.
• Account health. Identifying dead accounts, zero-view active accounts, and silently underperforming groups — the kind of network hygiene that gets invisible without someone actively looking.
• Per-client reporting. Pulling dedicated views of how each brand program is performing, from posting cadence and template mix to top-performing posts.
• Operational integrity. Spotting failed sound URLs, broken assets, and data quality issues in the publishing pipeline before they become a campaign problem.

“It’s the first thing I open in the morning. I’ll run a handful of conversations before my first meeting just to see what’s happening across the network.”

A founder-led data culture

Most analytics tools target data teams. Basedash works for operators. For Peter, that distinction is the entire point — as the founder of a business built on running tens of thousands of accounts in parallel, he needs the analytical horsepower of a data team without the latency of one.

The conversational interface lets questions evolve naturally. A simple lookup (“how many posts did this client get yesterday”) becomes an investigation (“which accounts are underperforming for this client this week, and is there a template or sound in common”) without ever switching tools or writing a query.

“Every minute I save on analysis is a minute I get back for the actual business. Basedash compounds that across the entire team.”

Impact

Since adopting Basedash, Parallel has built a self-serve analytics layer on top of its content distribution platform:

• The CEO answering his own questions about the network every day, without writing SQL or filing a ticket
• Real-time visibility into account health, content performance, and per-client outcomes across thousands of managed accounts
• Engineering time reclaimed for product work instead of ad-hoc data requests
• Operational issues like broken sounds and failed posts caught early because they’re easy to look for
• A growing portion of the team self-serving their own analysis instead of routing through one person

“Basedash turned the data layer of our company into something self-serve. For an operations-heavy business like ours, that’s the difference between guessing and knowing.”