Instacart shoppers besieged by bots that snatch lucrative orders | Hacker News Reader

Instacart shoppers besieged by bots that snatch lucrative orders

seattletimes.com

189 pointswallflower139 comments

When I was in college 10 years ago I did Amazon's Mechanical Turk to pay my way through. One of the things I did to make my life less miserable was to automate a lot of the searching, sifting, and grabbing of jobs (hits.) Bot would only accept jobs that fit my specified time/money ratio, or known hits that paid well and predictably.

Having the experience of using these "gig" side jobs I don't think there's anything wrong with people automating them to make the work less miserable and more lucrative. Frankly, anything but the best paying jobs cost more to do per hour than a federal minimum wage job. Thanks to automation I made around $16/hour (and I also had to pay taxes on that so it wasn't even that good!)

Don't hate the player, hate the game. And the game here is cutthroat and designed to exploit 1099 workers for below market labor cost.

If there was an alternative, higher wage they would pick that. This is the market labor cost.

"below market" because Instacart (and uber, etc) forgo normal labor law, have by-default but not by-contract full-time workers without any full-time benefits. Companies can achieve great benefit by externalizing cost on society, but should they? (And should they be able to?)

e.g. https://www.forbes.com/sites/ebauer/2019/12/16/is-uber-cheat... https://www.forbes.com/sites/clareoconnor/2014/04/15/report-...

Free market idealists think the market can determine the minimum wage without government or organized labor enforcing higher wages.

Free market idealists also tend to be very against the natural way for workers to consolidate and leverage negotiation power: trade unions.

Yeah, I never understood how "free everything" advocates tend to be so much against the freedom of the individual to refer decisions/actions to a collective when that happens to be the better course of action.

“Free market” isn’t “free everything”. In a free market the bid-ask spread is king. It’s just another kind of tyranny, which, like all such things, benefits those who are set up to take advantage of it, and leaves most others in the dust.

Maybe your lack of understanding is because you're arguing against an imaginary straw man?

A group forming a collective and negotiating as such fits perfectly fine with free market ideals. This exists in many forms that all manner of libertarian, free-market, whatever label people are OK with.

When that group uses violence, threats and coercion to their benefit is where people object. Examples - blocking an employer from hiring non-union employees, harming or threatening to harm workers during a strike, forcing people to join a union to get/keep a job, preventing new workers or new firms from entering the market through regulation, etc.

If you want to say "You can hire these 20 union workers at a rate of $$$, we all stand together" that's fine. When you say "You need to hire these 20 union workers, and if you don't, we'll surround your business and threaten, harass and intimidate the people you hire instead" is where you've jumped into violence and extortion rather than free market negotiation.

I think people in this thread are confusing "free market idealists" (usually libertarians) with conservatives/Republicans. A free market idealist has no problem with a free labor market that self-organizes into unions as long as the government doesn't mandate unionization. Conservatives and Republicans tend to advocate for big business and have therefore a natural antipathy towards organized labor.

I would also presume that in addition a `free market idealist` would also want to ensure that Unions aren't given any special status, and that employers could also come together to agree wages.

If unions don't aren't somewhat mandatory then workers are trapped in prisoner's dilemma.

Yes. That's an argument against libertarianism, not an argument that libertarianism must include mandatory unions.

The prisoner's dilemma is modeled with only two sides. A union might be better for average members, but it's strictly worse for non-members and employers and customers. Competition is a solution, not a problem to be avoided.

Comments like these are why I've moved to Blind for more quality discussions.

Not exactly. We are against the way unions are implemented and make work not a free market. I'm not against your union so longs as I can decide not to join and still have equal work with you.

Your thinking is a fallacy, because a union uses the same principle that the republican party uses to maintain control: solidarity.

I'm curious why you would want to opt out of a union in the first place, since unions generally double workers' wages. I live in the right to work state of Idaho, which has some of the lowest wages in the country, especially for things like farm work. I realize that this is a bit of a straw man argument though, which doesn't touch your main point.

I'm having trouble thinking of a case where a union charges more in fees than it provides in additional wages, benefits and other protections. So I think my main point is that you are fixated on a motivation that doesn't exist. It's like being angry that you must pay for a stamp to mail something through the post office, even though that costs a fraction of what UPS or FedEx cost. You're free to work somewhere else or use those other services.

Maybe someone else can answer this better than I can. I really do want to understand why unions are so controversial, because I've only experienced the downside of not having them. Like when I was moving furniture 20 years ago and the warehouse charged $34/hr and only paid us $10/hr, even though we were doing all of the work. My feeling is that had we been unionized, we would have made at least $17/hr.

>since unions generally double workers' wages.

>My feeling is that had we been unionized, we would have made at least $17/hr

You can dismiss this as anecdotal if you wish, but I went from a non-union job to a union one, requiring much the same skills, and the latter pays less.

On a per hour basis, my current job pays 95% of the non-union position, but the main difference is 7.5 hours vs. 8 hours a day. That means my yearly salary is more like 87%.

On the plus side, you are guaranteed raises over time, which my former employer explicitly disclaimed, saying all pay raises were based on "merit". It's also nice that you know everybody at the same level is making the same amount.

I'm not going to generalize about all union jobs from my experience, but from a theoretical perspective, if a union offers security and better benefits then it's plausible people would be willing to give up a certain amount of pay.

One aspect of having a union that I hadn't considered was that in March, they had to have a lot of intensive negotiations about working from home, because the existing contract did not allow it.

>>On a per hour basis, my current job pays 95% of the non-union position, but the main difference is 7.5 hours vs. 8 hours a day. That means my yearly salary is more like 87%.

Over your career, doesnt getting raises matter more than immediately more money that will never increase? and without a union could be ended at anytime.

Over a 25 year career, assuming 3% raises, you make 30% more. Plus you worked 3125 hours less.

What is a corporation but a union of capitalists? Why must only the owners be allowed to unite for better outcomes?

> Why must only the owners be allowed to unite for better outcomes?

Well, there's a reason the politico-economic system is called “capitalism” and not “laborism” or “socialism”, and it's not because it is structured for the benefit of labor or society as a whole. Such labor rights as have been established were a small but important step away from pure capitalism, and advocates of capitalism naturally want to peel them back.

A counterargument (that isn't necessarily convincing to me) would be that a union of capitalists is called a cartel, and that is often regulated or opposed, considered illegal or illegitimate; thus a union is equivalent.

I believe there are also libertarians (small-L) who do believe in the above, but think cartels and unions should be unregulated.

There is no difference. I shouldn't have to join either a union or a company.

The way I interpret what you say is that you'd like to have the benefits of being in a union, without being in a union. If that's the case (otherwise, sorry for misrepresenting your views) I don't see how that can work. How are a bunch of independent individuals going to secure their common rights without collaborating with each other?

A union is, in an economic sense, a cartel. Their job is to restrict supply of labour so that the price they can demand for it increases, making their members more money. This is why it takes over a year to join the plumbers or electricians union in many states and why it takes over 10 years to join the longshoremens union.

I am against all cartels as a matter of principle, be they the ILWU or OPEC.

OK. I'm against corporations as matter of principle. A corporation is also a cartel.

A corporation is a legal entity that provides some liability protection, and is not a cartel in any sense. Several corporation can collude to form a cartel, but a corporation in and of itself does not in any way meet the definition.

The Freedom Markets™ faithful proclaim

  - efficient markets

  - rational actors

  - no transaction costs 

  - power imbalances are moot

Any notions otherwise are apostate and must be purged.

> Companies can achieve great benefit by externalizing cost on society, but should they?

Yes, indeed they should. But the government also needs to claim taxes from companies to pay for these benefits. This concept is not even new, heck this is how most of Europe works.

Last time I checked, USA did have corporate tax. Just like we got in Europe.

Corporate income tax was just cut by 40% (from 35% to 21%). Almost 2.5% of the entire federal government's revenue.

So? It's not sky-high in Europe either. In my European country it's 15%, 5% for micro businesses.

The entire point of identifying an externalized cost is that it is not being paid for by the beneficiary. You can't both allow it and have them pay for it or it's no longer externalized.

> .. I did to make my life less miserable was to automate a lot of the searching, sifting, and grabbing ..

So it is not bots, it is your own customers, that know how to code. You should not fight your own customers, right? Coding skills in this century is like reading skills in previous one.

Sometimes you should.

Apart from it being zero-sum and this depriving other people of opportunity, with some of these services there's a promise of "starting fulfillment" upon acceptance. This is voided when a bot picks it up.

Real life example: I ordered an Uber, which was accepted by a "driver" just two streets down. Except the car still hadn't started moving 10 minutes later. So I walked over, and the GPS signal seemed to be coming from a parked car, no driver in sight. Messaged and called the driver, but no response, even 20 minutes after acceptance. Had to contact Uber to get the cancellation fee back.

> some shoppers are paying software developers who have created bots — in the form of third-party apps — that run alongside the legitimate Instacart app and claim the best orders for clients

Instacart is a real-time market. In real-time markets reaction time is paramount. "Cranking up pressure against app makers and banning violators when they find them" is swimming up a waterfall.

The solution is in the market structure. Fundamentally, Instacart needs to shift to batch processing.

This could take several forms. They could, on order submission, open a 60-second auction window. The best qualifying bid wins. Bids, mind you, could come in dollars or time to delivery or something else. Alternatively, shoppers could set orders parameters within which they guarantee acceptance. Orders could then be randomly assigned to online shoppers, perhaps with a bias towards ratings and other metrics.

Either way, a real-time market, where the first shopper to accept an order, seems under-optimised.

> cost of the third-party apps ranges from $250 to $600 in cryptocurrency or bank deposits

Instagram should acquire the differentiated apps and integrate them into its routing.

Showing every order to every shopper is inefficient. Some of these bots will just click on every order every shopper would find attractive. Those are zero sum. But some, I assume, help shoppers pick orders uniquely suited to them. Those are positive sum, and worth integrating into Instacart per se.

Fascinating stuff, nevertheless.

I'd be surprised if Instacart could pull this off, their app is one of the worst pieces of software I use on a regular basis.

Really confusing UX, simple things like looking up previous orders are several clicks and popups away. It's extremely, painfully slow (page loads take seconds and refresh often, I have a gigabit connection). Want to add things you always buy? It'll show you things you often buy in a weird side scroll thing. Pop ups of sales and extra things, confusing calls to attention everywhere.

Click that option to add what you've previously ordered? Hopefully you haven't already added some stuff because it's not smart about adding things, it'll just increment everything already in your cart that was also in your previous order.

Group cart where you have express? Make sure you're the one to click order, if a roommate does you won't get the free delivery. Lots of extra fees, order soda? heavy order fee, get alcohol? extra alcohol fee.

I've gotten used to it, but it's not a great experience.

Instacart rivals AirBnB in how bad their software is. Wipe from your resume bad.

As a mobile developer (and former web developer and even desktop developer going back to the 80's) I find terrible UI/UX embarrassing and unnecessary, but common. Sometimes it's the designers fault (pretty > usability) sometimes it's the product managers fault (same) but more commonly it's the executive's fault (I am more important than you and despite having zero clues about anything, my word goes).

Good design is freaking hard, but not rocket science; it requires the people in charge to understand how to say no and the people doing the work to be clear in why it's a bad idea when it is. While people sometimes revile Jobs for being an asshole, he understood this very deeply.

Instacart could care less, that's obvious to me.

What you're working up to is a reverse auction for delivery service. Who is willing to do a batch of deliveries for the least price? Constrained time TSP+Knapsack, a very challenging problem. (Because you can choose many different subsets of delivery jobs, run TSP for each, and calculate a bid price for each job that will win, or if you don't, minimises regret.)

The real problem with instacart is that some orders are better than others. If the delivery fee reflected the real costs then this wouldn't be such a problem.

EDIT s/your/you're. I swear my phone's predictive text is getting antagonistic.

Agreed, gaming the system like this is an indicator that pricing is incorrect

> The real problem with instacart is that some orders are better than others.

You don't need to eliminate differences to fix the problem - simply minimize them enough that it isn't worth building or buying a bot to give yourself an edge.

A simple solution would be to look at how quickly each order is snatched up, and reduce the price paid for orders that some AI decides will probably be snatched up quickly. I bet some simple linear model consisting of total travel distance, weight, number of unique items, and dollar value would suffice.

All of this is pretty simple. Instacart simply needs to block ability of those that make an order from decreasing the tip after the order is live and automatically charge the shopper a penalty in the amount of the tip and service fee for accepting an order and not completing it.

The first will remove incentives of customers to create orders that are tasty for shoppers under the false pretenses.

The second one will make it costly for shoppers to accept good orders that they cannot complete.

> open a 60-second auction window

Then people would complain that they are only able to get batches which are letting them earn way less than they already are. This can also be automated to keep going lower up to N dollars that the bot user is okay with.

> shoppers could set orders parameters within which they guarantee acceptance

Same problem as above

> Orders could then be randomly assigned to online shoppers

This would mean people who continuously exceed the customer's experience potentially lose out to those who are below them. This is a lose-lose situation

> bias towards ratings and other metrics.

This means that people will feel they don't have opportunities to improve their rating.

This is a really hard problem to solve in this market, imho.

Addendum: I agree with the sentiments about fees reflecting the order quality.

> This could take several forms. They could, on order submission, open a 60-second auction window.

It will just end up being sniped like it happens on eBay. Lookup gixen.

A Dutch auction is a good approach here... The earnings increase every second, and whoever is willing to do it for the lowest price clicks the button first.

I always thought the same thing about ebay.

I thought... if someone bids at the last minute, it should extend the auction a short amount of time. Let people get into it personally, not through pre-bid and/or software.

Except that eBay uses a particular auction model where by extending the time doesn't matter much.

If you bid the most you would be prepared to pay for something - you don't pay that price - you just pay enough to beat everyone else.

If you bid in a genuine way there's no pipped at the post situation - you just weren't prepared to pay more.

Now there are some secondary psychological effects, like a high price bid so far might increase bidders willingness to pay more, so there may be some advantage to bidding later, but overall there's less of a problem than it initially appears.

Perhaps bidders just need to be more honest with themselves about how much they were really prepared to pay.

I think you're grossly underestimating the effect of other visible bids on behaviour. When I was buying a lot of IT equipment, years ago, sniping was the only reliable way to win auctions. It was almost like people would only bid on things others had already bid on.

EBay has a fundamental flaw with bid retractions that discourages early bidding.

If you bid $100 on an item that is currently at $5, it becomes $6. If someone else (possibly in cahoots with the seller) bids $150 and then retracts their bid, the fact that you’re willing to pay $100 has been leaked.

oh, that's an interesting idea.

I sort of assumed that cancelling bids was a no-no and you would probably get banned, but I don't know the reality of it. (sort of like speeding in your car - most normal people assume you'll get in trouble and are reasonable, but maybe you can do it and nobody will care)

you are giving out information by bidding early. if everyone snipes at the last possible moment, you are right, winner pays the lowest they can. However, by bidding early, other competitors now have info on you and can adjust accordingly.

There's a much simpler solution -- just pay all the shoppers flat, fair, per-hour rates and eliminate tips.

If shoppers get paid for their work, not a percentage of orders, there won't be any of this mess.

UPS drivers don't get a percentage cut of what's inside the boxes they deliver; grocery drivers shouldn't be treated any differently.

In a time where so many people are looking for jobs, this is gonna be a race to the bottom.

So groceries are high-frequency trading now?

I suppose the effect of technology is to accelerate every market, so we'll gradually see this and its accompanying sniper bots in more places. Food. Housing.

I like everything you said, but one nit:

> Instagram should acquire the differentiated apps and integrate them into its routing.

Are they actually differentiated? My understand was they're just auto-grabbing jobs, maybe based on some parameters the shopper enters.

My problem with this, which is brought up in someone else's top-level comment here, is that there is an implicit contract when I use Instacart as a customer - that someone is starting to shop immediately (or as close to is practical). I have no interest in paying Instacart's pretty high fees and having my job sit in someone's queue for any non-trivial amount of time. If I wanted to do that I would just do grocery pickup instead of delivery. I'm honestly not sure how to reconcile these competing desires while still being fair to the shoppers, which I do want despite how that previous sentence may have come across.

Reminds me of working at amazon doing prime deliveries and using an Android app where you can create flow charts to automate grabbing hours. The difference is amazon at the time didn't care. We even reported what we were doing as a bug, and recommended a queue system instead, yet we only got automated responses not mentioning the issue at all.

This automation included an annoying screen refresh, where you hit back and then reenter to see if the button is lit up to get hours. I remember feeling terrible sitting in the warehouse waiting for an order and seeing others clicking back and forth as fast as they can the entire time, all while my phone was silently, with the screen disabled, 'scanning' for hours.

I think they consider the warehouse drivers like what I used to do temporary, till they get their drone fleets up.

Regarding instacart, and these systems in general, a queue system is just the most fair option. Your take home pay shouldn't be reduced to how well you can click a touch screen or whether you happen to be aware of app automation. It doesn't even seem to benefit the platform itself, just degrade the experience of the driver's.

Fyi, the word is "queue".

Haha, your right.

you're

>I think they consider the warehouse drivers like what I used to do temporary, till they get their drone fleets up.

Why even bother with drones when you can pay a person $8/HR with zero benefits thanks to collusion with lawmakers to bust unions and stall the minimum wage?

Drones likely will be cheaper than that because of improved quality- fewer errors, faster, better tracking, less fuel, etc.

Same deal with fast food workers who make minimum wage. Automation will improve quality even though people are so cheap. Being able to perfectly make tacos or big macs or whatnot and remove scheduling challenges will be a plus.

Because drones don't blow whistles, newspapers don't write articles about how drones are mistreated.

Not to mention lowering the overhead of HR, payroll, etc functions.

> a queue system is just the most fair option

There are lots of systems better than first-come first-serve for logistical systems. Hiring someone with a background in queueing theory and financial market microstructure would be a good starting point for Instagram (EDIT: Instacart). It would likely improve customer outcomes, too, to boot.

Better in what sense? I want my order filled as quickly as possible.

It is better to give the on site drivers a number in a queue then to make them f5 spam some app.

No, as they may accept or reject an order. So then you can have an order being kicked from driver to driver and not being fulfilled.

> take home pay shouldn't be reduced to how well you can click a touch screen or whether you happen to be aware of app automation

Could not agree more. I attempted to do Instacart at one point, and noped out as soon as I saw this asinine, medieval, 'race to grab an order before someone else' bullcrap. I am not playing that stupid game.

If a large number of drivers are using automation software, Instacart should announce it to everyone and provide basic instructions on how to do this so the playing field is leveled. Make it part of the job. Or, better yet, get rid of this awful mess and institute a more fair system.

I’m a software developer who has been enthusiastic about technology pretty much my entire life, but lately I’ve started doubting that our current technological advances are even a net positive for humanity. Reading about things like this (and also about online ads, robocalls, tracking, spyware, ransomware, extortionware, massive environment-ravaging crypto-mining ops, state surveillance, “predictive policing”, fake news, state-coordinated FUD campaigns, etc.) makes me want to give it all up and live out what remains of my life in subsistence on a primitive farm.

Then, you will learn about the plight of the farmers. Really, checkout "Super Size Me 2". I don't think there is any industry where its all gravy. Each industry has its own problems, you can only do your best to contribute in its positive growth. Don't be too hard on yourself. Also, you aren't Ghandi (or insert XYZ person here), don't put too much weight on your shoulders. FWIW none of us really matter, this is a symptom of how North American companies operate. Survival of the fittest. I went through deep depression for similar reasons, so hoping some tough love helps you.

I suspect you're conflating/confusing subsistence farming with industrial "farming". As someone who has lived a "subsistence" lifestyle for a long time, now (decades) one can live really quite well on a "sub-poverty-level income" simply because you're quite decoupled from conventional-money economies. I've heard it called "abundance poverty" (not "abundant poverty", to be sure! :) )

It's worth a disclaimer, though: I practise this subsistence (aka "self-sufficiency") living outside the USA. USmmv.

I'm going to go out on a limb and guess that this is only possible on Android, right? The video linked by ultimoo shows an Android, and my perception is that iOS sandboxing is much tighter.

How can Instacart combat this? They can't require iOS among their shoppers. Is it going to be an arms race of heuristics and anti-cheat code like it is for games that want to prevent aim-bots?

Or they could start running their own algorithm on the back-end, release juicy batches first to shoppers that haven't gotten a good one recently.

They can add any sort of captcha challenge or publish fake juicy shopping orders as honeypot, since the bots will instantly grab them.

Of course, the next step will be to pay a sweatshop in a poor country to contantly monitor and grab the offers. There's no end to human greed, as evidenced by the Uber Eats couriers outsourcing to undocumented migrants at pennies on the dollar (or the euro) in France.

Smaller teams than Instacart have put up effective fights against cheating. It's an arms race, but so what? The goal is to make cheating uneconomical, not impossible.

There's always some cost to having to implement countermeasures.

In games, I've seen seen poorly-implemented anti-cheat software kick or ban legitimate players from their games.

A sibling comment mentioned captchas. That would suck for legitimate Instacart shoppers. Imagine frantically looking for each photo that has a stop sign in it, knowing that the orders you're trying to claim could be snatched up by someone who's a faster captcha solver.

Somehow there's always a cost when bots start to wage war against other bots. Sure, they can figure out how to minimize it for the human, while making it uneconomical for the bots to exist.

I guess I'm just curious how they're going to do that. Because I can already see the blog post now: "Instacart Banned Me for No Reason"

Every service I've been involved with that was useful enough to succeed ended up implementing mechanisms to deter undesirable behavior. (I blame the humans.)

If you look at it the right way and are not currently under attack, they can be a lot of fun to work out and implement.

Physical robots that poke touchscreens exist, of course: https://techcrunch.com/2018/03/31/tapsters-robots-are-built-...

The analog hole stikes again.

They actually could require iOS among their shoppers. I did Postmates deliveries in ~2015 and they handed each delivery person an iPhone 5c with the courier app preinstalled.

I think this could be fixed by making each job be posted for a fixed block of time, say five minutes. Then it is randomly assigned to one of the shoppers that selected that job. This would remove the incentive around reaction speed and I don't think the customer would care about the slightly increased time for an order.

I was here to say the same thing. But you don't even need to extend the time frame to five minutes to defeat the bots. Lisa Marsh, in the article, complained "no human can click that fast." I bet she only needs 10-15 seconds to assess an order.

Not for nothing, this is the same mechanism by which the "unfair" advantages of HFT practitioners can be defeated. Run mini-auctions every whole second and randomly allocate fills if there are ties for best bid and/or best offer.

Of course, as long as the exchanges deeply care about vanity metrics such as trading volume (trading volume !== liquidity) such measures will never be broadly enacted.

No clue what Instacart shoppers do. What makes the orders lucrative/better? Is it high dollar orders, or multiple orders at the same grocery store, or is the tip pre-published, or what's the criteria when looking for orders to fulfill?

Lucrative orders are mispriced orders, no ifs and buts. That's how efficient markets work.

Instacart customers should be paying more for long trips and bulky items and short deadlines. Clarke once observed that “In orbit, caviar is cheaper than bread.”

Paying them more for valuable items only makes sense if the service includes preventing heists.

I don't know the exact formula, but they get tipped on the price of the order (like a server at a US restaurant would). Instacart customers are charged a flat delivery fee, so multiple orders also seems like a factor.

I think it means orders which will result in higher tips.

Is there a reason why instacart even allows you to pick orders? It seems like this exploit can be stopped just by only letting shoppers look at the first few items of the queue, and limiting the amount of skips.

That could be construed as Instacart assigning orders to their shoppers which would likely make them employees eligible for benefits, something they want to avoid at all costs. This may already be the case in CA regardless of what they do, but they definitely want to avoid paying any benefits for their workers in any state where this is not yet currently the case.

Definitely this. They don't care who gets the orders as long as they're gotten, and assigning the work to people rather than letting people pick from a pool makes them look more like an employer/dispatcher rather than a broker.

> construed as Instacart assigning orders to their shoppers

Uber & Lyft assign rides to drivers. [You can still cancel them after they ping you, but it's discouraged.] It seems to me like it works extremely well overall compared to what Instacart is doing.

Grocery shopping is a bit different from just picking up a food order. You may only have the time (or desire) to do smaller orders and not want to deal with larger more time consuming ones. Or maybe you don't want to drive farther to certain stores etc....

Sure, but all of those could be addressed by allowing setting available-for criteria rather than browse-and-select.

Or they could just not display tips in advance; since tips can be altered after delivery, it's misleading to treat them as part of the bargain of accepting a delivery anyway, advance entry should be viewed as a convenience for the orderer to enter a default amount based one expected service, not an inducement.

If the tips aren't visible in advance, bots can't snatch “lucrative” orders, because they can't find them.

I agree, I think showing tips prior to an order is absolutely stupid. I did Skip the Dishes delivery on the side and you could see the tip before you delivered to the customer. It always seemed silly to me to be able to know what was tipped. What if I had a big order and no tip? Honestly even as a customer I don't like this because you feel pressured to tip otherwise be afraid they will do something to your food. It would be like going to a restaurant and having to choose your tip before they did anything and have them know what you chose to gave them. It just is stupid.

To me it seems the problem is the very concept that some orders pay better than others.

Seems to me they should adjust the pay scale so that all orders pay similar when graded by time+difficulty.

I assume its to make sure orders get assigned to users who are actually available, but I am sure there are better ways of doing this.

Maybe the best strategy is to tip low when placing the order, then increase the tip after delivery to reward the guy who needed the money enough to take your order with a low tip.

why not just tip them in cash?

Many or most deliveries are contactless: https://news.instacart.com/introducing-leave-at-my-door-deli...

I doubt an Instacart contractor would turn down a cash tip. Pre-COVID, I exclusively tipped Lyft drivers cash. Yes I had to keep small bills around. Ended up not being a big deal.

Taping an envelope to your door clearly visible for the delivery person would still remain contactless. If the envelope is still there when you pick up your items, then no harm in trying

And if it's not, then you'll wonder if it went to the Instacart driver or the UPS guy.

(well, not my UPS guy, he's actually a she, and she gives my dogs treats when she sees him)

You kind of know if you have deliveries coming to your house, right? You get to schedule your Insta deliveries, so just don't make your Insta deliver 10:30am, 3pm, or 9pm. Should avoid the UPS delivery

I live in a condo with 4 doors in close proximity, so I don't really know when UPS, Fededx, Ontrac, Amazon Delivery or USPS are coming -- UPS could come anytime from 8am-6pm, and sometimes more than once a day.

Fair enough. I wasn't considering non-single family locations. At that point, you wouldn't know if it was a rando walking down the corridors let alone an actual delivery person. No good deed goes unpunished

This is what I do. I write INSTACART on it. Everyone loves cash.

How do you secure it?

People steal Amazon packages regularly, I think they would not hesitate to steal something that says "Instacart" on it.

Is there an app to deliver cash to my house? I haven't been to an ATM in at least a year.

/s but only kind of.

you do have a point. i have used cashapp more in the past 4 months than i have ever used it.

I don't even see the delivery driver, he drops the food off, knocks, and leaves.

Besides, I don't want to have to keep a supply of small bills around just to pay a tip that I could have just put on my credit card. I rarely pay in cash these days (even pre-covid)

Then you'd also incentivize workers to take orders with low tips and be disappointed when they really get a low tip.

Seems more fair to me to require the company pay their employees a fair wage and benefits rather than redefine those that work for them as contractors to dodge those obligations.

Most people who accept a low tip order would not be expecting to get any more than what was there

For a while, that's what you had to do to keep Instacart from taking a big chunk of the tip. They finally stopped doing that, after public pressure.

https://www.eater.com/2019/2/6/18214354/instacart-delivery-w...

The bots aren’t snatching the orders.

Other shoppers, competing for the same business, are using software programs to compete more effectively and efficiently.

This is not an exploit. Access to deal flow is first-come, first-served.

I was curious what these actually looked like in the wild and what Instacart shoppers thought. Here's some shoppers discussing bots:

https://www.reddit.com/r/InstacartShoppers/comments/gshom2/h...

https://www.reddit.com/r/InstacartShoppers/comments/fagva2/d...

https://www.reddit.com/r/software/comments/7mfxmh/a_good_fre...

The most commonly-mentioned example is "MurGee Auto Clicker" (a simple form clicker: https://www.murgee.com/auto-clicker/tutorial/) and miscellaneous Android form clickers.

A company offering technological innovation to gain advantages in a market is going to ban shoppers who use technological innovation to gain advantages in their internal market?

There's obviously competition among shoppers. The argument presented in this article is that those with technological advantages aren't acting fairly, but considering the dynamics of the shoppers' market, I would argue that they are not only acting fairly but are introducing what can be the next phase of innovation for the entire business. Instacart may have explored instant match in the past but found issues with it. Rather than drop it entirely, maybe they ought to revive the idea but refine it in consideration of what sniping is accomplishing.

I had come across a video couple of months ago showing this in action. Captcha was mentioned a few times on the thread. https://www.youtube.com/watch?v=nwCh4rtAOWc

I love how the kneejerk reaction to any bot problem is "captchas". I'm not sure how that would even help. If there are 20 jobs available, and 3 of them are high paying, having a bot sort through the list for you still gives you an advantage. Sure, it's better than having someone claim all 3 of them at the same time, but you can achive the same effect by having a cooldown of 5 seconds (enforced server side). Also, with a cooldown, you can't get an advantage (eg. entering a captcha faster with keyboard)

Instacart could wipe away this feature in one afternoon if they wanted.

Seems like a solution for eliminating automation could be to make such services only accessible through an [iOS] app, perhaps by requiring Face ID verification on accepting an order.

> When Marsh opens her Instacart shopping app, she sees promising orders disappear before she can act. “No human can click that fast,” she said. “Instacart needs to fix this. These bots are literally taking the food off my kids’ table.”

While I understand the wish to get the best/any batch, this is significant exaggeration. If your kids food depends on getting the batches from the website, then you have much larger problem than bots.

True, although I understand the complaint. Many platforms succumb to bots and power users and they often loose their appeal from that. Most prominent example for me is ebay. Didn't really used it to save money, but it was interesting to browse.

Now there is little opportunity for the average user to even visit it, or at least much less so.

In related news, lions have sued that the invention of spears by early tribes makes it significantly more difficult to hunt humans and it's taking food off their cubs' table.

This sounds very much like ordinary competition that any self-employed person would have to deal with. If I insist on using assembler to make websites for clients while my competitors use Rails, it's not a secret who is going to do better in the long run.

Uber handles this by not letting drivers know the destination before they arrive for pickup. Can't instacart offer the same functionality? Like you see the order, shop location, but do not see the amount and other "lucrativity" info. And there is a penalty for cancelling after you accepted.

This has been happening for some time now, unfortunately, IC doesn't give af b/c the orders are still fulfilled. In the 4 sided marketplace they balance, usually the shopper experience ranks last in priorities, like in all gig economy apps

This strikes me as unethical.

What is unethical to you? Is it the competition that is bad or is using technology to compete the bad thing?

Any system - or market - sophisticated enough will see the emergence of bots as the ultimate decision makers/arbitration players. It’s quite fascinating.

the problem stems from the fact that instacart is not doing a proper 'auction' of the orders to all participants.

if the instacart platform allows all deliverers to bid on a task (where each deliverer bids their % take of the order), and the lowest % take wins the bid, then this will give everyone a fair chance at taking all jobs.

My groceries are delivered by employees (either of the chain or of the postal service, and on rare occasions, of DPD). No need to hate the player when one simply avoids the precariat game.

Wow saw that video, maybe someone got into their API?

What if someone from a non-USA location started selling access to the best batches in prime USA locations e.g. SF/LA/SEA/NYC/DC then we could start trading and packaging up those batches and trading on them in crypto exchanges, maybe even make a new coin?

Also having someone put in a SSN is not any kind of verification if you’re trying to subcontract out to people not in the country legally/documented. The ssn algorithm is known and it is easy to generate, it just don’t get fixed because politics.

Maybe we have finally found a decent use for eth?

>e.g. SF/LA/SEA/NYC/DC then we could start trading and packaging up those batches and trading on them in crypto exchanges, maybe even make a new coin?

In that case you might as well make claiming orders be based on a dutch auction system.

And without ETH. Just with, you know, a server. With a dutch auction.

I stopped using Instacart years ago. My reasons were:

1) Shoppers would ring your phone for less than a second, ghost you and then show up with random replacements you never approved.

2) They would fail to deliver on time, then unilaterally reschedule a delivery for another time when I was not even going to be home.

3) Amazon Fresh and Prime Now are far superior services. Amazon Fresh has gotten slightly worse in the last couple of years, but it is still far better than Instacart.

Has the service improved lately? Should I use Instacart in 2020?

The last paragraph of this is heartbreaking.

https://archive.is/2TgtR

One solution is to make the contracts more even? Or allow some automated auctioning process.

.app is great for giving official-sounding domain names to scammers. Way to go, Google.

Eventually more .app domains will be squatted, the price will go up, and companies will buy domains proactively (like instashopper.app - instashopper is obvious shorthand for Instacart Shopper, the official term) but not before a lot of unethical enterprises take advantage of it.

lf these shoppers are really just "independent contractors" participating in Instacart's open market then Instacart is completely failing as a market maker.

i couldn't view this site, anyone have a mirror link?

Article also at https://www.seattletimes.com/business/instacart-shoppers-bes...

Ok, we'll change to that from https://www.bloomberg.com/news/articles/2020-07-31/instacart... on the assumption that more people can read it. Thanks!

Reply on news.ycombinator.com