Shipturtle x Optiblack: How Company‑Level Analytics Turned a Noisy Data Firehose Into a Revenue Engine

When the team at Shipturtle looked at their analytics, they saw… almost everything and almost nothing at the same time.

Events were pouring in from GA4, UTMs, Google Search Console, New Relic, and Mixpanel. Clicks, page views, keystrokes — millions of them, every month. Yet the questions that really mattered to the business remained stubbornly unanswered:

• Which Shopify merchants are actually successful and healthy?
• What does a winning multi‑vendor marketplace setup journey look like, per company?
• Which campaigns are attracting the merchants who go on to generate real revenue?

The data was there. The signal wasn’t.

This is how Shipturtle and Optiblack rebuilt analytics around the company — not the individual user — and turned a noisy firehose into a focused, revenue‑aligned analytics engine.

1. The Product Behind the Problem

Shipturtle is a Shopify app that lets merchants turn a regular Shopify store into a full‑fledged multi‑vendor marketplace.

For each merchant, Shipturtle:

• Onboards and manages vendors from a central dashboard
• Syncs vendor products into the merchant’s Shopify store
• Routes orders to the right vendor
• Handles commission calculations and payouts
• Integrates with 200+ shippers for label generation
• Gives vendors a white‑label portal that looks and feels like the merchant’s own system

Crucially, Shipturtle is merchant‑centric. The Shopify store owner — not the end vendor — is the paying customer. Pricing, value realization, and expansion all happen at the company (store) level.

Yet the analytics told a different story.

2. The Growing Blind Spot

As Shipturtle scaled, three intertwined gaps became impossible to ignore.

2.1 Company‑Level Questions, User‑Level Answers

The business is sold and billed at the Shopify store level, but the tooling was tuned to users:

• Fragmented tracking across GA4, UTMs, Google Search Console, and New Relic
• Mixpanel configured mostly around user‑level events

That made basic, business‑critical questions hard to answer:

• Which companies are truly active and healthy?
• What does a successful marketplace setup journey look like per company?
• Which acquisition channels bring in high‑value merchants, not just more signups?

The data model and the revenue model were out of sync.

2.2 A 14‑Day Trial With No Map

The 14‑day free trial is Shipturtle’s activation crucible. If merchants don’t hit value quickly — adding vendors, syncing products, processing orders — they churn.

But the team lacked a reliable, company‑level view of the trial journey:

• No robust funnel from app install → charge approval → vendors added → products synced → first order processed
• No clear picture of where merchants dropped off
• No evidence on which actions actually correlated with conversion and long‑term retention

Decisions about onboarding and activation were made with more intuition than insight.

2.3 High Analytics Cost, Low Signal‑to‑Noise

Mixpanel was already in place, but it was working against them:

• Auto‑captured front‑end events (clicks, keystrokes, generic page views)
• Roughly 5.7M events per month
• No strong grouping key at the company or domain level

The result was a dangerous combination:

• Rising analytics bills and the risk of getting locked into high‑volume paid plans
• Dashboards that were visually busy but hard to trust — teams questioned definitions more than they used insights

Shipturtle needed analytics that looked like their business: company‑centric, cost‑efficient, and trustworthy.

That’s when they partnered with Optiblack.

3. Reframing the Goal: Analytics That Match the Business

From the first strategy workshop, the two teams agreed: this was not just about “better dashboards.” It was about re‑architecting analytics around how Shipturtle makes money.

They aligned on four clear objectives:

1. Design a company‑centric analytics model aligned with Shipturtle’s pricing and value realization — Shopify store / domain, not just individual users.
2. Instrument a precise onboarding and activation funnel for the 14‑day free trial, from app install to first vendor and first order.
3. Optimize Mixpanel usage and costs by removing low‑value event noise and focusing on high‑signal, back‑end‑validated events.
4. Build trustworthy dashboards backed by historical data and clear definitions so business and product teams could act without second‑guessing the numbers.

With the destination clear, Optiblack started from first principles: what is the real unit of value in this business?

4. Designing a Company‑Level Analytics Engine

4.1 Choosing the Right “Unit of Truth”

Optiblack proposed a group‑based analytics model built around the Shopify domain.

• Primary group ID: shopify_domain — representing the merchant’s store / company
• Unique user identifier: user_id (database primary key) — to keep a direct line between Mixpanel profiles and back‑end records

This shift unlocked a different class of questions:

• Which domains are active each week and month?
• How many companies complete onboarding, and how quickly?
• What does retention look like at the domain level?

In other words: what did organizations do? instead of what did users click?

4.2 Making Roles Explicit: Merchants vs Vendors

Under the hood, Shipturtle has two very different types of users:

• Merchants – the paying customers (Shopify store owners)
• Vendors – non‑paying users linked to a merchant via parent_id

Optiblack wired all back‑end actions to user_id and used a fetch_bootstrap API to surface role information to the front‑end. Events could now be tagged with who was acting and on whose behalf.

This kept the merchant‑centric lens intact while still capturing the white‑label vendor experience.

4.3 Taming Mixpanel: From 5.7M Events to Under 1M

The next step was reducing noise — and cost.

Optiblack:

• Disabled auto_capture to stop collecting generic clicks, keystrokes and page views
• Introduced a structured event taxonomy, initially focusing on a small, high‑value set:
• Account Creation
• Vendor Invite Sent
• Vendor Invite Accepted
• Implemented server‑side events to validate client‑side coverage (which can drop to ~70%) and ensure that critical business events are never missed

At the same time, they consolidated Mixpanel into two clear projects:

• Test shipturtle development
• shipturtle prod

The effect was dramatic: event volume dropped from ~5.7M to under 1M events per month, unlocking Mixpanel’s free (or much lower‑tier) pricing — while increasing the signal in the data that remained.

4.4 Backfilling the Past to Trust the Future

New tracking alone wasn’t enough. If the dashboards contradicted years of internal expectations, teams would distrust them.

So Optiblack designed a backfill and data integrity process:

• Shipturtle exported historical data for:
• Users
• Shopify domains
• Parent / organization mappings
• Optiblack mapped and cleaned this data into
• Mixpanel group profiles (by shopify_domain)
• User profiles (by user_id)
• Shipturtle then imported the mapped data into Mixpanel

This meant trend lines didn’t start “from zero” on the day of the new implementation. Dashboards could show coherent histories, and leadership didn’t have to debate whether the tool or their intuition was wrong.

5. Implementation: From Local Setup to Live Dashboards

5.1 Laying Technical Foundations

To move quickly without fighting the stack, Optiblack first aligned on environment and workflows:

• Standardized development on Linux (Ubuntu) to match Shipturtle’s Laravel/PHP tooling
• Established a branching protocol for new analytics work
• Unblocked local back‑end setup so developers could work against realistic data

With a stable environment in place, they connected Laravel back‑end events to Mixpanel, attaching:

• user_id for per‑profile tracking and debugging
• shopify_domain and parent_id as group properties for organization‑level analytics

5.2 Mapping the Onboarding Journey

Next, we designed the onboarding and activation funnel that truly reflects the Shipturtle trial experience:

1. App install and first session
2. Charge approval and start of the 14‑day free trial
3. Vendor onboarding (manual or self‑serve)
4. Product sync and approval flows
5. First order processed

On top of this, they built dashboards to:

• Track completion and drop‑off at each step
• Compare 1‑day, 3‑day and 7‑day completion rates
• Analyze time spent per step and how often users re‑attempted onboarding

An early and powerful finding emerged: merchants who did not complete key steps within three days almost never returned.

This “three‑day rule” became a north star for product and success teams — a concrete window for interventions.

5.3 Cutting Event Volume and Governing Tracking

To keep Mixpanel lean and meaningful, Optiblack:

• Authored and deployed scripts to turn off auto_capture and rely on explicit, code‑defined page view and business events
• Reduced session replay coverage from 100% to 10%, focusing on paid traffic sessions (identified via UTM parameters) where diagnostic value was highest
• Put in place a lightweight governance process to:
• Review final event schemas
• Ensure shopify_domain and user_id are consistently attached
• Coordinate deployments with Shipturtle’s engineering team

5.4 Bringing Stakeholders Into the Dashboards

Optiblack’s goal wasn’t just to ship charts — it was to make them usable.

They introduced four real‑time dashboards, each with a clear audience and purpose.

Quick Executor (High‑Level Overview)

Built for leadership, this dashboard surfaces:

• Daily Active Users (DAU)
• Monthly Active Users (MAU)
• Session duration
• Retention curves over days, weeks, and months

During a live demo, a small but telling moment happened: a filter for MAU was corrected from “contains shipturtle.com” to “does not contain” to exclude internal staff.

The result: the dashboard surfaced a true MAU of 380 merchants.

That single change visibly increased confidence in the data — a signal that the system now reflected reality, not noise.

Onboarding Journey (Deep Dive)

This dashboard tells the detailed story of activation:

• Funnel from “app session started” through “onboarding completion”
• Step‑by‑step drop‑off rates
• Completion rates at 1, 3, and 7 days
• Time spent on each step
• Retry behavior for merchants who re‑attempt onboarding

Each metric is annotated with definitions, intended use, and guidance on interpretation, so teams spend less time arguing about what numbers mean and more time acting on them.

Retention & Engagement Dashboard

Here, teams can see:

• Weekly active domains
• Segmentation by paid_customer status
• Feature usage patterns by organization

This makes it possible to answer questions like:

• Which features are most strongly associated with long‑term retention?
• Which newly activated domains are starting to look like our best customers?

Marketing ROI Dashboard

Finally, the marketing team gets a direct view from spend to revenue:

• Google Ads performance (impressions, clicks, cost)
• Company‑level revenue tied back to campaigns and audiences

For the first time, the team can see which campaigns actually produce high‑value merchants, not just installs.

6. Outcomes: From Noise to a Revenue‑Focused Analytics Engine

With the new stack in place, Shipturtle’s analytics story changed in several measurable ways.

6.1 Quantitative Wins

• ~90% cost reduction in event volume — from ~5.7M to under 1M events per month, keeping Mixpanel in the free or low‑tier range.
• 380 Monthly Active Users surfaced as the true MAU after excluding internal staff.
• 99% data accuracy, validated by pairing client‑side tracking with back‑end events.
• Four real‑time dashboards covering executive overview, onboarding, retention/engagement, and marketing ROI.

6.2 The Three‑Day Onboarding Rule

One insight quickly stood out: if merchants didn’t complete onboarding within three days, they almost never came back.

This shaped multiple strategies:

• Product: prioritize in‑app tours, nudges and UX improvements front‑loaded into the first 72 hours.
• Success and Sales: time outreach sequences around that three‑day window.
• Experimentation: focus A/B tests on the steps with the highest early drop‑off.

6.3 Company‑Level Engagement, Not Just Clickstreams

By grouping behavior at the Shopify domain level, Shipturtle can now answer:

• How quickly organizations complete critical actions like vendor onboarding, product sync and processing the first order.
• Which domains are at risk of churn and which are expanding.

Analytics now mirror how revenue is generated and retained.

6.4 Marketing That Speaks the Language of Revenue

With campaign data tied directly to company‑level revenue, the marketing team can:

• See which keywords and audiences bring in merchants who actually activate and stay
• Scale spend on high‑LTV cohorts, not just high‑CTR creatives

The conversation moved from “which ads get the most clicks?” to “which ads create the best customers?”.

6.5 Faster Debugging and Stronger Support

Using user_id as the unique identifier across the stack means:

• Support teams can jump from a ticket or database row straight into the corresponding Mixpanel profile
• Engineers can see the exact event sequence that led to an issue

This shortens resolution times and reduces guesswork, without adding new tools to the mix.

7. What Shipturtle Learned Along the Way

Across the engagement, several principles emerged that extend beyond this one project.

1. For B2B SaaS, company‑level analytics beat user‑level analytics. When you sell and bill at the company level, your analytics must reflect that. Otherwise, insights are hard to turn into revenue‑aligned decisions.
   
   
2. Intentional tracking beats auto‑capture — for both cost and clarity. Turning off noisy, generic events and investing in a curated taxonomy of back‑end‑validated events cut costs by ~90% and made dashboards easier to interpret.
   
   
3. Backfilling is not optional if you want trust. Historical imports ensured that new dashboards aligned with past reality. This prevented endless debates about “which number is right” and encouraged teams to use the data.
   
   
4. Group analytics unlock more meaningful questions. Grouping by Shopify domain shifted the narrative from “what did users do?” to “what did organizations achieve?” — a far more actionable lens for a marketplace SaaS.
   
   

8. A Durable Foundation for Growth

What began as a problem — noisy, user‑centric analytics misaligned with a company‑based business model — became an opportunity to rebuild Shipturtle’s data foundation.

Through a focused partnership, Shipturtle and Optiblack:

• Re‑anchored analytics on the Shopify domain, matching how revenue is created
• Clarified the 14‑day trial journey and the critical three‑day activation window
• Reduced analytics costs while increasing signal
• Equipped product, marketing, leadership and support with a shared, trustworthy view of the business

Today, Shipturtle has not just “better dashboards,” but a revenue‑focused analytics engine that helps them:

• Understand how merchants build and grow their marketplaces
• Improve trial‑to‑paid conversion
• Allocate product and marketing resources where they create the most long‑term value

And underneath it all is a simple shift in question — from “What are users doing?” to “What are our companies achieving?” — backed by an analytics stack designed to answer it.