How Instacart started is really a story about grocery price data. In 2012, the startup had no store deals and no product feeds. In fact, much of the data it needed was not online at all.

So its team shopped at regular store prices and set its own prices in the app. Later, Instacart moved from markups to retailer partnerships. Today, however, retailers set prices, and many now match in-store prices.

This guide covers each stage, from the first order to the 2025 price-test backlash. In addition, it shows what businesses can learn about tracking grocery prices.

How Instacart Started in 2012

Apoorva Mehta, Max Mullen and Brandon Leonardo founded Instacart in San Francisco in 2012. Before that, Mehta worked as a supply chain engineer at Amazon, according to Contrary Research.

Soon after, the company joined Y Combinator’s summer 2012 batch. In fact, Mehta won a meeting by using the app to deliver beer to a Y Combinator partner.

The first order was humble, too. Later, Mehta told investor Elad Gil that he placed an order himself, shopped for it and delivered it to himself. As he put it, he “gave myself a nice tip”. In short, this is how Instacart started: one founder, one app and one order.

Then, in August 2012, TechCrunch reported the launch. The app covered San Francisco, Mountain View and Palo Alto. It listed “tens of thousands of items”, while one-hour delivery cost $9.99 and three-hour delivery cost $3.99.

How Instacart Started: “Ninja Shopping” at Shelf Prices

Big grocers were not ready to work with a tiny startup. In fact, Mehta said signing a Fortune 500 retailer like Kroger or Costco “was just going to be impossible”.

So Instacart used what he called “ninja shopping”. Its staff simply walked into stores, bought items at shelf prices, and delivered them. At launch, those staff were paid hourly and did not accept tips, according to TechCrunch.

At first, revenue came mainly from delivery fees. Store partnerships were only a possible future source of revenue at the time. That shaped how Instacart started making money.

Soon, however, Instacart set its own item prices. In 2013, its FAQ said that “Instacart prices are our own and vary from the store’s price”. That same year, a customer compared receipts and reported a markup of about 30%. He also wrote that Instacart support said it did not have “any type of relationship” with the stores.

Later, Profitero compared Instacart prices with Costco shelf prices in 10 ZIP codes. Its February 2016 study found an average markup of 23%. By contrast, Instacart itself quoted a “15%+” markup.

How Instacart Started Its Catalog Without Store Data

Delivery was only half the job. First, Instacart needed a catalog with product names, photos, and prices.

Mehta said he was “totally okay scraping stuff, but you just couldn’t find it”. In other words, much grocery data was simply not online. Some SKUs “didn’t even have any online information about them at all”, he added.

So the team did it the hard way. They bought “one of every single thing from the store”, photographed each item in a studio and uploaded it. According to Mehta, buying the entire store cost about $50,000.

The effort paid off quickly. After Instacart added that store, “overnight our demand doubled”, Mehta said.

This part of how Instacart started answers a common question. Instacart was open to scraping, but in 2012 there was little grocery data online to scrape. Today, by contrast, Instacart shows products from nearly 100,000 stores across more than 2,200 retail banners.

From Markups to Retailer Partnerships

The turning point came in September 2014. That month, Instacart announced a partnership with Whole Foods Market, according to Fortune. Mehta later described it as Instacart’s first major deal. By then, it had more than 1,000 personal shoppers in 15 cities.

Even so, he called it “a terrible deal” where no math “would ever make sense”. Still, it “legitimized what Instacart was to other grocers”. As a result, the next deal was three times better, he said.

In April 2015, Instacart then changed its revenue model. It moved from food markups to deals with grocery stores, earning a cut of the sales it brought them. In addition, the app began showing how its prices compared with store prices.

Partner stores such as Whole Foods matched in-store prices. Meanwhile, non-partner stores still carried markups. For example, in 2017 a Reddit user found an order cost 42.8% more than the store receipt, before fees.

Fees and memberships grew, too. Instacart Express launched in August 2013 and became Instacart+ in June 2022. Also, a mandatory 5% service fee arrived in April 2018. Then, in May 2019, Whole Foods ended its US partnership, reportedly because of its ties to Amazon. Over time, partnerships replaced much of the markup model from how Instacart started.

Finally, Instacart went public on Nasdaq in September 2023 under the ticker CART.

Instacart Prices Today: Who Sets Them and What Changed

Today, Instacart pricing looks very different from how Instacart started. The company now calls itself a grocery technology company. Its retail partners control item prices on their Instacart storefronts. Fees, however, are separate and can vary by distance, basket size, and delivery speed.

The 2025 Price-Test Backlash

In December 2025, a study by Groundwork Collaborative and Consumer Reports raised new questions. It found that 74% of tested items showed different prices to different shoppers. For those items, the average gap between the lowest and highest price was 13%. However, the authors said their sample did not represent the whole US population.

The tests ran through Eversight, a pricing tool Instacart bought in 2022. Then, on December 22, 2025, Instacart ended all item price tests. It said the tests “missed the mark for some customers”.

Separately, Instacart agreed to pay $60 million in refunds to settle FTC claims about its fees and free delivery offers. Even so, the company denied any wrongdoing.

The Move to No-Markup Pricing

Meanwhile, more grocers now match in-store prices on Instacart. Hy-Vee and Raley’s dropped markups early in 2026, and Fareway and others followed. By September 2026, Instacart said more than 90% of Americans could reach a “grocery parity retailer” on its app.

In July 2026, Instacart also published new pricing principles. For example, it promised the same item prices for the same store at the same time. It also ruled out surveillance pricing and dynamic item pricing.

On top of that, the business keeps growing. In Q2 2026, Instacart handled 90.3 million orders and $10.35 billion in GTV. Advertising and other revenue reached $297 million, or 2.9% of GTV.

Lessons From How Instacart Started for Grocery Price Data

The story of how Instacart started holds useful lessons for any business that tracks grocery prices:

• Channel prices differ. App prices, in-store prices, and delivery fees can all differ for the same item. So track each one separately.
• Store and time matter. Prices can vary by store, ZIP code, and day. Therefore, collect data at store level with a timestamp.
• Fees are part of the price. Service and delivery fees change the real cost of an order. Also, memberships can change it again.
• Markups change over time. Retailers move between markups and price parity. As a result, a one-time snapshot soon goes out of date.
• Data is easier to get now. Instacart once spent about $50,000 to build one store’s catalog by hand. Today, by contrast, public grocery data can be collected at scale.

In short, grocery prices are a moving target. That is the biggest lesson from how Instacart started, because the winners measure prices often.

How FoodSpark Helps With Instacart Price Data

Unlike how Instacart started, you do not need to build a catalog by hand. Instead, FoodSpark offers Instacart store pricing data across cities and ZIP codes.

• Product fields: name, brand, category, pack size, and SKU where available.
• Price fields: list price, discounted price, promo label, and unit price. Since Instacart does not always show unit price, FoodSpark computes it.
• Availability: in-stock status and suggested substitutes.
• Location: retailer, store ID, city, state, ZIP code, and timestamp.
• Refresh: prices every 2 to 6 hours or daily, while promos update hourly during campaigns.
• Delivery: a REST/JSON API, flat files, or BI-ready feeds for Tableau, Power BI, or Snowflake.

FoodSpark collects publicly available data only. In addition, it does not scrape personal or identity-related information.

You can also explore the Instacart grocery extraction service, the Instacart API, and grocery data scraping. Then check food data API pricing for a free 100-row sample.

How Instacart Started: FAQs

Who founded Instacart, and when?

Apoorva Mehta, Max Mullen, and Brandon Leonardo founded Instacart in San Francisco in 2012. Then it joined Y Combinator’s summer 2012 batch.

Did Instacart scrape grocery prices in its early days?

Not in the usual sense. Mehta said he was open to scraping, but much grocery data was not online. So the team bought, photographed, and listed products by hand.

Why were early Instacart prices higher than store prices?

Instacart had no store deals at first, so it set its own prices. As a result, many items carried markups, such as the 30% one a customer reported in 2013.

When did Instacart partner with Whole Foods?

Instacart announced the partnership in September 2014. However, Whole Foods ended its US partnership in May 2019.

Does Instacart still mark up grocery prices?

It depends on the retailer, because retailers now set prices. Some add markups, while others, such as Hy-Vee and Raley’s, offer no-markup pricing.

How can I track Instacart prices by ZIP code?

Use a data provider such as FoodSpark. It collects store-level Instacart prices by city and ZIP code, and it offers a free 100-row sample.