The Calculus of the Batch

When Instacart announced in February that it had added nearly 6,000 Family Dollar stores to its swelling marketplace of retail outlets from which consumers could now receive fast online ordering and delivery, it seemed like a pretty good deal for all involved: The retailer got an option to expand reach into the growing on-demand shopping modality from which it could win incremental sales and share, and thanks in part to a to minimum-order threshold associated with delivery, tapped into a new kind of shopping trip generating what for Family Dollar were bigger and more profitable baskets.

For Instacart, having thousands of new outlets broadened its profile as a kind of an “everything store” in the on-demand omnichannel space and built still greater market density to better optimize its corps of gig-based shoppers and drivers. And Family Dollar customers got the option of a more convenient way to shop.

One group that wasn’t so excited? Gig drivers and shoppers who fret over what they perceive to be an expansion of jobs that potentially net them too little compensation to be worth their time.

“Family Dollar batches are likely to be high-risk/low-reward for the crowd-sourced workforce,” one experienced Instacart driver, who asked not to be identified, told WGB. “Experienced shoppers avoid low- and no-tip batches that result in single-digit hourly wages. Instacart will need to pull out all the stops to get them filled.”

A Service-Value Disconnect?

If you want to find an Instacart driver in the wild, pull into any high-volume Costco parking lot first thing in the morning, and you’ll find them staring into their phones, ready to pounce. Successful gig workers tend to be self-styled entrepreneurs and hustlers by nature. Many perceive their work as a kind of challenge, not only among fellow drivers in their area, but about gaining whatever edge they can on the algorithms that deal them job opportunities in the first place. They discuss customer interaction in manner not unlike hourly retail workers on the sales floor: They understand good service is especially important, but customers can still drive them crazy.

Instacart broadcasts its jobs—or what it calls “batches”—to participating workers on their apps, who then race to accept and fulfill them.

Opinions vary widely as to which jobs are more attractive than others, and no two driver strategies are necessarily alike. But among Instacart workers in general, competition seems highest for those jobs where drivers can earn more than $1 for every item picked between the “batch payment” (or what Instacart pays between an incentive, mileage and any extra pay for heavy loads) and a customer tip. Other factors also play a role in the jobs offered: For example, shoppers with better customer ratings tend to see better opportunities, as do those with higher “acceptance rates.”  

All of these points—the mileage, the pay, the tip, the retailer or retailers involved, along with any number of personal considerations, like limits on trunk size, familiarity with store layouts, relationships with staff at stores or customers, and a worker’s willingness to lift heavy items or climb stairs—play into a kind of quick-twitch calculus gig workers must process as batches arrive, sources describe. Good jobs tend to vanish in seconds; many today say they are also competing with “bots”—or software that sniffs out and claims lucrative jobs (this practice enters a murky area of the terms and conditions for workers that could get violators deactivated from the platform, but is another common complicating factor in the distribution of gigs and is not limited to Instacart, workers say).

Instacart for its part acknowledges there are “many considerations for every order that need to be taken into account” when it determines batch pay, including the volume of customer demand, geographic profiles, retailer mix, items, weight and projected delivery times. While declining to comment on specific retailers, the San Francisco-based tech company said shoppers over the past year have been earning up to 60% more on every batch, which includes higher payments from Instacart and nearly double the tips from customers, as opportunities exploded amid pandemic-related demand. And despite oft-stated uneasiness with Instacart’s role in retail, its roster of physical stores continues to grow: Walgreens and Michael crafts stores announcements followed the Family Dollar announcement this year, expanding the stores its contracted shoppers visit to nearly 600 retail brands.

Sources in the gig world say that disparities have emerged between the perceived attractiveness of jobs by retailers, making the stores whose clients tend to have big shopping baskets and see delivery as a service (a Costco run, for example) a seemingly better deal for gig workers than a trip to a Family Dollar store, whose shoppers might perceive delivery as an element of the value those retailers offer. The disparity, some workers say, is not because Family Dollar shoppers don’t tip, but because they say retailers and Instacart aren’t doing enough to inform customers that drivers rely on tips. The combination of this and other factors they say, can upset the balance between gigs, despite the algorithmic adjustments intended to democratize them.

Costco runs, which typically encompass large baskets, can be lucrative for Instacart shoppers when they include percentage tips, workers say. Photograph: Shutterstock 

“From my experience delivering from different retailers and being an organizer, the feedback I get is that the clientele of stores like Family Dollar typically does not tip, because they perceive us as an extension of the retailer, or a delivery person for the retailer that’s already being compensated with an hourly wage,” Willy Solis, a Texas-based independent delivery driver who typically takes Instacart and Shipt orders, told WGB in an interview. “These companies don’t make it clear that we’re a separate service that relies on tips in order to subsidize our pay. That’s where the disconnect is.”

Solis is quick to reject the notion—expressed by at least some gig workers—that the bargain-minded clientele of discount stores like Family Dollar or Aldi may be less able to tip.

“We find that to be actually be the opposite,” he says. “People in the lower-income bracket typically tip us better than affluent customers, which is surprising when you think about it, because they're in a situation where they’re least able to give, but they give us the most typically. So I would tend to believe that the issue is revolving around these stores not making it clear that we are independent contractors that rely on tips.”

Solis serves an organizer for Gig Workers Collective, a nonprofit group seeking better compensation and benefits for freelance drivers. From his perspective, retailers have used services like Instacart primarily to quickly and cost-effectively present themselves as omnichannel merchants amid the influence Amazon has brought on the industry, and so have found it either inconvenient or not in their purview to consider the concerns of freelance workers shopping their stores and carting their deliveries to customers.

“At the end of the day, what they’re doing is presenting us to the customer as a last-mile solution. They’re saying, ‘Don’t come to the store, we’ll bring it to you.’ That’s the value that they're trying to add to their particular brand,” he says. As internet shopping exploded during the pandemic, this took additional gravity, Solis says, as crowdsourcing in effect “transfer[ed] the liability of becoming ill from coronavirus over to someone other than their customers themselves.”

Aaron Hageman, founder and CEO of Delivery Drivers Inc. (DDI), an Irvine, Calif.-based third-party administrator specializing in last-mile delivery, acknowledges that the retailer associated with the order can matter as workers process gigs. He boils grocery driver decisions to three considerations: Who is the customer? What’s in their basket? And does the associated retailer take drivers into consideration as part of their offering (for example, auto-populating the tip field in the delivery app)? “If I’m a driver,” Hageman says, “I have a good relationship with a store if all three things check off.”

There’s nuance associated with tips that’s often out of the control of the driver, and is evolving as the country adjusts to an economy that’s more heavily influenced by gig work and omnichannel shopping, Hageman says: “Do you tip the pizza guy? Of course you do. But do you tip the flower delivery driver? And the answer there, nine times out of 10, is no. And that’s because I don’t want my girlfriend to have to tip the driver if it’s a surprise for her. The idea is, it’s all taken care of.”

According to Hageman, about 70% of delivery customers tip drivers currently.

Peak Boost

If it’s indeed true that some gigs are perceived to be better than others, how does Instacart assure its retail partners it can deliver on the promise of on-demand service? That gets to the remark from the shopper above about “pulling out all the stops.”

One common solution involves combining several batches into a single “gig.” Instacart also runs various promotions offering bonuses to workers for the number of batches completed over a period of time. This can help clear small orders that might not otherwise draw sufficient interest on their own. 

Another tactic involves “peak boost” bonuses indicated by an accompanying “fire” icon on the app. While these gigs can be lucrative, experienced Instacarters tend to view these offers with some suspicion: Adverse weather conditions, shopped-down stores or troublesome customers could be awaiting them.

The Driver-to-Gig Ratio

Experience in the gig economy has made Solis decidedly skeptical that apps like Instacart and the retailers that work with them will do more for the freelancers building their business without some collective action by workers: The recent #DeclineNow movement among some DoorDash drivers is one such example, as activist giggers encourage a boycott of low-reward trips. Part of the issue, Solis says, is the nature of crowdsourcing itself: a gig-worker-to-gig ratio that, in the driver’s estimation, is considerably larger than ideal and creates scenarios where workers need to make quick and often hard choices for themselves that frequently include a wild card of some kind.

“There is no real cost associated with onboarding new shoppers,” Solis says of the delivery apps. “It’s very minimal for them. They don’t have any of the exposures that are associated with traditional employment. So they go out of their way to make sure they have supply—which is what they’re selling to customers: We are the product and the service that they’re using in order make their operation work.

“The overstocking of shoppers really creates a scenario where when these orders are presented, we have literally seconds to decide, without having the ability to be able to properly dissect the order, to understand the distances that are involved and the retailer that’s involved and all the other data that we need in order to make sure that we’re completely understanding of what it is we’re taking on,” he continues. “Without that, it’s not fair. We don’t have enough data or enough knowledge about the order.”

Hageman of DDI—a company used by employers like Walmart to proactively recruit drivers for its Spark Delivery program and maintain an optimal balance of service—sees things a little differently in this regard, describing a goal of reaching stability between the promise of on-demand delivery with fair compensation for the labor to achieve that.

DDI this year anticipated that it will recruit 140,000 gig workers on behalf of its clients this year, on top of a fourfold increase in drivers it engaged in 2020 as businesses and consumers turned to delivery to adapt to the pandemic restrictions. Gig work, or what Hageman calls the “1099 workforce” in reference to the freelancer federal tax forms, has also grown as an option for millions whose jobs that were curtailed or lost in the COVID crisis.

“If we accept the 1099 workforce for what it is, they go into a pool, and as a worker we have to accept we're in that side of the pool,” he says. “The workers can balance that out being in a pool that jobs are thrown into by the ability to join multiple platforms. Hopefully, that keeps the balance.”

In assessing the need to recruit, Hageman describes the challenge of meeting the key performance indicators of drivers and customers at once while also assessing expected growth of the industry.

For drivers, Hageman says DDI looks at average revenue per job and the economics of time: He says about two-thirds of gig workers are full time, and 60% of them work across multiple platforms. The company also considers wages by markets, he said.

For customers, key variables revolve around the ability to provide on-time performance and meet consumer expectations, noting that falling short on these promises could have a considerable impact at scale.

“Do I get the window I want? Did I get an Uber in five minutes or in 25? If I order Chinese food, do I want to wait 35 or 40 minutes? Or an hour and a half?” he describes. “This is what the world of last-mile has to consider when we look at the economics of resources we give driver partners.

“The divers love it if you’re understaffed,” Hageman concludes. “The customers love it if you’ve overstaffed.”