Two-way reviews in Taxi Apps: The Good and the Bad

Nostalgia is a strange thing. It makes you appreciate the good things long after they have floated away into oblivion.

As a harried, hassled, hapless Mom she always had trouble keeping those levitating bodies called kids tethered to one place. Only rules, and tough ones, could have helped her survive the mayhem. No homework done – punishment. TV at wrong hours or more than allotted time – grounded in the room for two days. Playtime exceeding study time – folding laundry for five days.

At that point it had seemed quite a bucketful. But now when she looks at her daughter reliving the cycle of life, she scratches her chin even if it’s not itching. Her modern-day-supermom-version daughter has two kids and she has devised a weird way of managing them. If the brother is playing a video game more than the allowed quota, the sister would report him. And if the sister is browsing anything that is not related to her studies, the brother would jump in as a tattletale with glee.

She had tried asking her daughter – why would she give away the reins of discipline to the objects of discipline at all? She did not even turn her gaze up from her dishwasher as she answered – Mum, I have no idea where the lines get blurred between school work and idle-Internet; how am I supposed to keep track of their sneaking and tip-toeing? Let them play cops to each other.

Ok, so this is how the world is going to be when her grand-kids grow up. Ok, so this is why the tables turned when one day her daughter told her to behave and say Good morning to the driver.

Not for manners, but for reviews.

Alas, this is the state of things in the new sharing economy that we inhabit willy-nilly. Even if Uber or another aggregator app that you use for anything (a holiday accommodation, a cycle on rent, a second-hand good, a holiday dog care kennel) is not displaying it in your face, you as a customer are being rated on a report card the same way you strike a tick for the driver, merchant, host or the career you picked.

In fact, so impactful and hushed can such ratings be that, reportedly, there have even been some hacks into getting one’s Uber score and the company has to change its code time and again to fix such attempts at report-card thefts.

Now isn’t that a good thing for frazzled and clueless moms of the regulatory system?

Specially as the distinction between erstwhile players and aggregator-companies are stubbornly fuzzy whether it is the taxi world, the hospitality industry or the retail markets. Every time a new incident raises its elfish head up, the law-makers/operators find themselves arranging and re-arranging the furniture of their familiar-worlds before they can even deliver a slap on someone’s/anyone’s wrist.

The peer-to-peer networks have doodled a new world and it’s often too slippery and hazy when it comes to dealing with the laws and the systems we are inured to.

Their answer is simple and hard to contend with – why listen to laws when the very technology and platforms we use allow strong and adequate self-policing?

Double-lane Reviews: Cops without a Uniform

If transport regulators are struggling with the challenges that taxi aggregators are pouring every now and then, along with protests from regular taxi players; then other regulators are not spared either. Airbnb has unleashed the Jumanji board of a new level altogether as authorities accuse hosts and landlords of circumventing legal hygiene under the vague ambit of a networked economy.

But then, how do they answer the logic that Uber, Airbnb and their ilk throw forth – The sites themselves are armed with rigorous and two-way rating systems, leaving not much room for negligence of safety, privacy, decorum, and other basics that laws insist on.

The nature of these reviews is distinct too: they are collective, individually-marked yet shared, and dilute the hierarchies of order with their unique digital steam.

Algorithms and automated ratings iron out intermediaries, gatekeepers or any form of prejudice that could render the process pointless – at least that’s what the sites and apps promise.

When there is freedom from interference by intermediaries, collaborative consumption and access economy provide ample room for an experience-based review as the guide for future consumers/users. And that’s not all, the ratings are two ways – so if a passenger or guest is adding to white-listing or black-listing a driver or a host (on Uber or Airbnb respectively), then the passenger and guest is also being rated by the reviewee (the driver, the host) and is equally vulnerable for a blacklist if the line is crossed.

This self-regulatory mechanism of apps-driven, decentralized, sharing economy does weed out external controls that were often merely a fly-on-the-wall view of an experience. They insert real, hands-on experiential flavor into the regulation system.

And almost every app, Uber, Airbnb, TaskRabbit, even the NHS website that is asking patients and their friends and family to comment on and rate the service, the doctors etc. is doing it.

At Airbnb a consistently bad review for either host or guest can effectively translate into no return business. At Uber, being sloppy with time or misbehaving can have equally dire ratings, whether it is the driver or the passenger.

It’s not just riders but drivers get to rate riders on a 1-5 star scale as well, thus, ensuring that arrogance, time-wastage, rude behavior and vomit can mean being blacklisted as a passenger for future rides.

Sounds almost alien and about time- How good is this mould-breaking system though?

T for Trust, T for Transparency

Peer reviews underpin the very nature of collaborative consumption if we think it from a model perspective. A rating system that works immediately, anonymously and struck by the affected party directly is potent enough to create that missing piece called ‘trust’ in our marketplaces.

Both the service-user and the service-provider have something to lose and something to gain depending on how they observe their boundaries and expectations. They are intangible but their outcomes are quite tangible if we look at the way they incentivize correct behaviour and the manner in which they enrich feedback inventory for future users.

Nothing could be better for a two-sided market like Lyft, Uber, Airbnb than two-way rating system.

Let’s look at what a judge in California put a spotlight on when denying Uber’s motion for summary judgement on the California drivers’ class action suit. Edward Chen reasoned that customer ratings, give Uber an “arguably tremendous amount of control over the ‘manner and means’ of its drivers’ performance.”

In a government-run system, one has to get one certification and go along with it, with the next one coming after a specific time window. Here, in an aggregator system, every next review is a certification that the next user will watch and decide upon.

They turn the very idea of policing on its head – why not all of us police all of us? Responsibility shifts easily from government courtrooms to online courtyards. In fact, Professor Sara Dolnicar from the University of Queensland has often reasoned in favour of Airbnb that the inherent review function of online share economy accommodation could be more rigorous than legislated regulations

In a study titled ‘Regulating the sharing economy’ Kristofer Erickson and Inge Sørensen,  University of Glasgow, argue that while the rise in sharing economy will impact the composition and constitution of new and established industries, it will affect policy on all these levels. They suggest that the approaches to studying the sharing economy will need to take account of social as well as economic aspects.

If online and platform-driven ratings can influence correct functioning then we don’t need laws, right?

Don’t say it. Wait, it’s not that easy.

History teaches Future

Many researchers pick eBay as the reference point that started the ball rolling here.

It was presumably eBay’s review system that snowballed into a market that made traditional retailers sit in shock with a deer-in-the-headlights epilepsy.

The online feedback system soon catapulted this retailer into the leagues of formidable retail giants. Because now buyers could simply look at what others buyers had to say about a merchant or a product and that was it.

Yet, eBay pioneered not just the model but also the pitfalls and complications it can run into.

  1. The Power-See-Saw:

When in 2008 eBay reportedly saw itself grappling with sellers coercing positive transaction reviews from buyers, it learnt something for the industry as a whole.  University of Maryland’s Siva Viswanathan and Goudong Gao explain in their report how merchant outrage erupted upon eBay’s subsequent ban on negative and neutral reviews by sellers.  Seems like the site came to a U-turn of sorts when sellers protested they were being stripped of protection from deadbeat bidders or unreasonable customer demands. Then, of course, there was the problem of retaliatory seller feedback and even a weeklong boycott from sellers.

The study shows not only how an online reputation system’s design affects seller behavior but also how some merchants can exploit what looks like an even, two-way rating system between buyers and sellers on the surface. This not only compromises the marketplace but also the T word.

The balance of power can shift in the wrong way from users to service providers.  What about the absence of calculating ‘revoked feedback’ data in a reputation rating system? Can users post honest ratings when they themselves are worried about the boomerang of a rating from the other side?

Plus, there is ample scope of manipulation in a reputation system that is plaited around algorithms and software. After all, eBay faced the loopholes in an unanticipated way too when sly merchants gamed the two-way rating systems by shipping fragile items and stamping negative retaliatory feedback for buyers.

  1. Robust or Rogue:

We can also take a leaf out of how the eBay’s system has evolved to incorporate specificity and context, when it introduced Detailed Seller Ratings (DSRs) that were introduced to bring in accuracy, quality of communication, shipping speed and shipping charges instead of one quick score.

Now while online marketplaces may seem self-sufficient to take up regulatory responsibility delegated to them, a certain level of government or neutral oversight cannot be ruled out.

Ratings can potentially be manipulated by dishonest or malicious participants. One cannot deny the lurking fear that commercial bias can crawl inside such algorithm-powered platforms too. Also, ratings may miss the context, thereby doing more harm than good to a player being rated.

Where can such possibilities and their regulation get displaced if there is no other cop to watch over these cops?

Even if a player is not a hotel or a taxi in the technical/traditional sense, they can’t turn a blind eye when it comes to traffic jams, crisis and emergencies or fire safeguards, can they? In another study, Merethe Dotterud Leiren and Jørgen Aarhaug revert to the semantic trick between taxi and non-taxi again to bring attention to how classification and resultant regulation is being evaded.

  1. Half Page Blank and Grey and Smudged:

Then, there is the half-lit, half-dark alley phenomenon. Riders can’t see the ratings that drivers give them. Users apparently want to, and will hack for it too if they can to, get a peek into how the provider rated them. Incidentally, even eBay ran a test for that possibility and when it did Jeff Terrell, its head of community, stated that it wanted to gauge if providing buyers with information about seller ratings is a distraction. Sellers, understandably, did not dovetail with the idea and pointed at it as a suicidal move that could drive more buyers away.

So, making rating systems, even if they are automated, self-driven and in-theory transparent; robust and credible substitutes for laws would not be easy for digital platforms and aggregators.

Like, Paul Belleflamme, professor at Université catholique de Louvain points out, for intermediaries there is a big need for designing strategies to induce agents on both sides to participate. How do you do that though; because to attract one group, you need to attract the other one?

Catch 22 again.

It’s not easy to ‘just’ rate

Also, how can we ignore that ratings are just averages whipped up from experiences that actually deserve qualitative evaluation and context. Scores can be highly misleading and polarizing if one relies on how statistics and averages work.

To achieve a thorough and objective feedback when one stares at the lack of established protocols, guidelines etc that aggregator platforms should provide; is quite an uphill task.

  1. Loopy Larks:

Good conduct differs from person to person. Being too talkative as a driver can be bad manners for one passenger. For another passenger, that may count as sitting with an uptight, snobby driver. The same goes for passengers too, even if one does not like the volume or the kind of music a driver is playing. How can one side be objective when there is so much subjectivity in how we like to be treated in a utopian world, and at the same time there is the repercussion angle from the other side waiting in the wings?

Being a rater is like getting a corner-room, even if for those adrenaline-turbo-charged two seconds. Becoming a boss can be akin to being drugged. It’s a powerful feeling but in the wrong place, and in the wrong hands, there is no guessing who can harm who and to what extent. Think of how much a driver’s choice to pick someone or not can be influenced by the perception s/he forms of a rider who might look like the one who gives unfavourable ratings. Think of the anxiety it causes for a driver lest s/he commits a minor mistake and pays through a bad rating received. Think of the artificial smiles and ceramic manners it elicits from drivers. Reminds of you flaky hotels all over again in an ironic twist.

  1. Unfair and Ugly:

Then let’s not fool ourselves on the labor angle here. All said and done, sharing economy is quintessentially a gig economy. There are untrained, voluntary employees running all around without a supervisory fabric, and assuming that self-regulation is the gossamer that will hold this storm together is such a fallacy.

Sometimes, ratings can become so overwhelming as a force that drivers and other service providers end up working on low-paying errands purely for ratings to pull their ranks a notch up. The risk of being barred from picking the next job or being blacklisted for a certain bracket or being elbowed out in search listings, is just too high for someone whose livelihood may depend on this gig job.

More so, because sometimes a rating is nothing more than a bad mood or a whim being stamped as someone’s bad performance. Worse, it can be a racist or prejudiced attitude venting out in form of a poor rating. So much so, that as a Harvard professor Benjamin Sachs hinted – if we look at Uber as an employer,  it can find itself perched in the violation rung of Title VII of the Civil Rights Act, for the use of customer ratings as a discriminatory tool.

  1. Easy-to-Carry?

Then, ratings, unlike a government certificate or nod, can’t be ported from one place or platform to another. A driver can quit a bad boss if he has a taxi medallion once. But here the medallions are only in the form of binary numbers that are closeted inside one platform. There’s no way to use them fluidly across the industry. Unless universal API data-medals of sorts are allotted to well-rated drivers or kept in third-party libraries, as some experts have been suggesting – platforms will wield their power through this control over rating data.

Ratings in this space conquer the problems of immediacy and ease that yesteryear customer-feedback forms faced. But removing arbitrariness from ratings is just one of the many challenges that aggregator platforms face. It’s a new industry, where unions and laws are still amorphous. Not to forget, the commercial bias monkey that still sits heavily on the shoulders of these app-driven companies. Apart from ensuring transparency and objectivity on the raters’ part, they also have their work cut out for ensuring they are not rigging ratings or leveraging monopolistic tactics that are easily amenable to them.

That means…

If we are wondering whether these new-wrinkle platforms can take care of fundamental issues like fire-safety, health inspection, identity verification, criminal background checks, and car maintenance etc.; then we are not being cynical. We are just cleaning the wipers again and again till the fog of confusion relents.

The sharing economy is, as of now, looking ginormous in impact and implications. If a third of Europeans were found using the services of a sharing platform in EU, 2016; 72 per cent of North Americans were seen there as per Pew Research Center, 2016.

If PwC estimates have sussed it out well, then by as early as 2025, we would be witnessing the sharing economy ringing in a whopping $ 335 billion in global revenue and this would be across verticals – finance, transport, accommodation, media, and secretarial services, creative production and software – everything.

We can’t then be casual about how a marketplace of overwhelming proportion operates and regulates. The platforms, their technologies and their postures of intent can gain and help immensely, if they evolve in-house and other regulatory systems.

Like Pierre Omidyar, eBay’s Founder urged in a 1996 letter to the eBay community

Make your complaints in the open. Better yet, give your praise in the open. Let everyone know what a joy it was to deal with someone. Deal with others the way you would have them deal with you. Remember that you are usually dealing with individuals, just like yourself. Subject to making mistakes. Well-meaning, but wrong on occasion. That’s just human. We can live with that. We can deal with that.

Until ratings can be unbiased, reliable, fair, qualitative, humane and helpful as good cops; we can’t throw other policing measures out of the window. At least, not yet.

Because the super-mom of this age still finds herself perplexed when she hears the smashing of glass and happy shrieks in the kitchen. Sometimes, even arch-rival siblings can be in cahoots.

That’s how the cookie jar crumbles.

Author: Shashwat DC



  1. binu poduthas

    March 16, 2017 at 9:09 am

    Ola & uber taxi alldrivers cheating

  2. LastJamaal

    January 26, 2018 at 9:36 pm

    I have noticed you don’t monetize your website, don’t waste your traffic, you can earn additional
    bucks every month because you’ve got hi quality content.
    If you want to know how to make extra bucks, search for: Mertiso’s tips
    best adsense alternative

Leave a Reply

Your email address will not be published.

Most Popular

To Top