DTC operators spend 9.3 hours per week searching for and gathering information across their tools. That stat comes from McKinsey. In practice, it means half your working week is spent being the connector between Stripe, Shopify, Klaviyo, Gorgias, and whatever spreadsheet currently holds the closest thing to a complete picture. You are the API. The data passes through you, not a system.
This isn’t a productivity hack problem. It’s a system design failure. Your tools are excellent at their individual jobs and completely unaware of each other’s existence. So you fill the gap. You open tabs, cross-reference email addresses, export CSVs, paste numbers into Google Sheets, and try to assemble a view of your business that no single tool was designed to provide.
How much does tab-switching actually cost?
McKinsey’s 9.3 hours per week number is the headline, but the IDC research underneath it is more specific: knowledge workers spend 2.5 hours per day just searching for information. Not analyzing it. Not acting on it. Searching for it.
For ecommerce operators, the shape of this is painfully familiar. You export a CSV from Stripe, open Shopify in another tab, pull up Klaviyo’s analytics, check Gorgias for recent tickets, and then try to hold all of that in your head long enough to answer a single question about a single customer. Ecommerce automation surveys put this cross-tool work at 5-10 hours per week for operators doing regular reporting.
The cost isn’t just the time. Every time you switch from one tool to another, you lose the thread of whatever you were analyzing. You were looking at a churn pattern in Stripe, then you opened Gorgias to check for related tickets, and by the time you found the tickets you forgot which cohort you were investigating. Context switching has a cognitive cost that doesn’t show up on any dashboard but affects every decision you make.
The tab bar becomes a to-do list you can’t close. Twelve tabs open, each one representing a question you started answering but haven’t finished. That is the real tax.
Five ways being the human API costs more than you think
1. You’re always looking at last week’s data.
Manual assembly means the picture is always stale. By the time you’ve stitched together churn data from Stripe, engagement metrics from Klaviyo, and ticket history from Gorgias, the patterns you’re seeing are already a week old. Maybe older. Your intervention window for at-risk customers is measured in days, according to Express Analytics research showing that a 40% engagement drop over two weeks makes churn 3x more likely. But your data lag is a week at minimum. You’re making decisions about what’s happening now based on a snapshot from last Tuesday. The gap between when a pattern becomes visible and when you see it is where customers disappear.
2. Your analysis has a sample size of one.
When you pull data manually, you pull it for the specific question you’re investigating right now. You don’t pull it for the question you haven’t thought of yet. If you want to know why a specific customer churned, you look up their Stripe history, their Gorgias tickets, their Klaviyo engagement. You might find the answer. But you won’t find the pattern, because patterns require looking at hundreds or thousands of customers at the same time. A system sees correlations across your entire customer base simultaneously. You see correlations one customer at a time, and only for customers you’ve already decided to investigate. The customers whose churn would have surprised you are the ones you never looked up.
3. You can’t hand off the process.
If you get sick, go on vacation, or hire someone to take over your role, the “system” breaks. Because the system is you. Nobody else knows which Stripe export to pull, which Gorgias filter shows the tickets that matter, which spreadsheet tab has the formula that connects payment failures to support interactions. The knowledge lives in your muscle memory, not in any documented process. I’ve seen operators spend two full days trying to bring a new hire up to speed on their weekly reporting workflow, only to have the new hire build a completely different version because the original was too idiosyncratic to replicate. You haven’t built a system. You’ve built a habit that only you can perform.
4. You miss the cross-domain signals.
Commerce Signals found that 47% of retail marketing spend is wasted. That waste happens in the gaps between tools, in the correlations that exist in the data but don’t exist in any single dashboard. A support ticket spike from customers acquired through a specific Instagram campaign. A churn pattern tied to a specific product SKU. A correlation between email disengagement and payment failure timing. A product return rate that spikes for customers acquired during a promotion. These patterns are invisible when you’re looking at one tool at a time, because each tool only sees its own domain. The most valuable signals in your business sit at the intersections. But nobody is looking at the intersections, because no tool was designed to show them, and you can only hold so many variables in your head while toggling between browser tabs.
5. You optimize for urgency, not importance.
When you’re the human API, the loudest signal wins. A Gorgias ticket queue with 47 open tickets gets attention because it’s visible and it’s red. A slow churn pattern across email-disengaged customers acquired through paid social doesn’t get attention, because nobody is assembling that view. It exists in the data. Three different tools each hold a piece of it. But it doesn’t have a notification, a badge count, or a queue. So it doesn’t get prioritized. The urgent work (responding to tickets, processing refunds, sending campaigns) crowds out the important work (understanding why customers leave, which channels produce retained customers, where your revenue actually comes from). You spend your week reacting instead of analyzing because reacting doesn’t require you to build the picture first.
The math on what this actually costs
At 9.3 hours per week, that’s roughly 480 hours per year spent on information gathering and manual integration. For an operator or head of ops making $100,000 to $150,000 annually, that’s $45,000 to $70,000 in time spent being the glue between tools that should already be connected.
That number doesn’t account for the decisions you didn’t make because you didn’t have time to assemble the data. It also misses the churn you didn’t prevent because the pattern was invisible across tools. And it completely ignores the marketing spend you wasted on channels that produce customers who leave within 90 days, which you would have known if acquisition and retention data were connected.
Salesforce’s research shows that only 31% of marketers are satisfied with their ability to unify customer data across platforms. DATAVERSITY’s 2024 survey puts it even more starkly: 68% of organizations cite data silos as their top data management concern, up 7 percentage points year over year. The problem is getting worse, not better, because every new tool you add creates another silo.
You shouldn’t be the API between your tools. That’s a system design problem, not a work ethic problem. Every hour you spend stitching data together is an hour you’re not spending on the decisions that data was supposed to inform.
What connected data changes
Riven automates the connection layer. Instead of exporting CSVs, maintaining Zapier chains, and cross-referencing spreadsheets, entities are linked across tools automatically. A customer in Stripe is the same customer in Gorgias is the same customer in Klaviyo. Their full history, from the ad that acquired them to the ticket they filed last week to the payment that failed yesterday, is visible in one place without you having to assemble it. The 9.3 hours per week you spend being the human API goes to zero. You stop searching for information and start acting on it.
Frequently asked questions
How much time do DTC operators actually spend on manual data integration?
McKinsey’s research puts knowledge workers at 9.3 hours per week on information gathering. IDC narrows it to 2.5 hours per day spent searching for information specifically. For ecommerce operators doing regular cross-tool reporting with CSV exports and manual lookups, the range is 5-10 hours per week, and it scales with your customer count and tool count.
What are the alternatives to being the human integration layer?
There are four options at different price points. Manual CSV exports work but cost 5-10 hours per week. Zapier or Make automations handle simple triggers but break on anything analytical. Reverse ETL with a data warehouse (Hightouch, Census, BigQuery) is the most flexible but requires technical staff to maintain. Purpose-built platforms like Riven connect tools automatically without requiring a data team.
When does it make sense to invest in connected tooling versus doing it manually?
When the time cost exceeds the tool cost. At $100K salary, 9.3 hours per week of manual integration costs roughly $45K per year. If you’re spending more than 5 hours a week on cross-tool data work, or if you’ve missed a churn pattern because the signals were spread across tools, the ROI math is already in favor of automation. Most operators hit that threshold somewhere between 500 and 2,000 active customers.
Does adding another tool just create another silo?
It depends on how the tool works. If it creates its own data store and its own dashboard that you have to check separately, yes. If it connects to your existing tools and links their data together without requiring you to change your workflow, no. The difference is whether the tool sits alongside your stack or sits between it. DATAVERSITY’s data shows that 68% of organizations cite data silos as their top concern, and the percentage is growing. More tools doesn’t have to mean more silos, but it usually does unless the tool was specifically designed to solve the silo problem.
Last updated: March 26, 2026