Integrate to Differentiate

Many ecommerce business sell parts of a solution (individual SKUs/services) conveniently, quickly and cheaply. Services like Uber/Ola, Airbnb/Booking, 1mg/netmeds/Practo/Portea, Bigbasket, Amazon and similar have made life immeasurably better than it used to be earlier. However, they still do not always provide a full integration solution for the user's problem and may be missing large value, as well as a way to differentiate themselves and create user loyalty.

Recommendations done well can guide users towards full solutions. We at Euler Systems have been able to use them to help clients increase revenue dramatically. The attempt is to mimic a good and responsible sales agents who is skilled enough to save the user who came in to buy a squash racket multiple trips back to the shop as he realizes on his first visit to the squash club that he cannot step on the court without non-marking shoes, on the second that the balls the squash club supplies are not bouncy enough for the beginner, and a few games later that a racket works much better with a separate grip. The goal, of course, is to find everything the user needs but no more.

An entire look or a set of looks for summer created by the provider is much more convenient and coherent than having to buy individual shirts, trousers, shoes, suits, ties, belts etc. StichFix just launched an algorithm developed by a 100 data scientists that can help users put together a look by helping figure out what items of clothing go together with #recommendations. They plan to charge a styling fee for this service.

An entire meal for all guests in a party chosen in a restaurant based on past history + weather + time of day + accompanying guests?(#Cred or maybe #Zomato could do this) is much better than ordering the same old, boring individual items by hand after wasting 20 minutes hemming and hawing through the menu. The reason you end up with the same boring stuff is because the host or the decider doesn't want to offend religion, health or taste. Rather than forcing time with friends to be wasted on these procedural questions, all this should be left to algos - democracy might survive the danger from Swiggy knowing I like dosas thrice a month but probably not after that. Even at home, 5 dinners worth of recipe boxes for a household of two would be much better than shopping individual items from?#Bigbasket

Holiday pacakges a la?Disney cruises or bespoke ones like #Pickyourtrail?are better than combining Flights from?#Kiwi/#Kayak, cabs (at least with Uber, that doesn't require prep in mainstream destinations) + tours & activities from?#GetYourGuide?+ stays with?#Airbnb/#Booking by hand.

An entire treatment for TB/Diabetes + diagnostic tests + doctor visits/telemedicine + attendant help + Monitors + Physiotherapy, if a single provider were to provide it, would be better than cobbling together a solution by hand by using disparate parts from #1mg?#netmeds?#Practo?#Portea, a hospital and the insurer

In content consumption, Netflix has done a good job of solving the "problem" of having ample spare time with enough content and good recommendations. Youtube's recommendations still aren't quite as effective. One might blame the lack of serialized content as the culprit, but music services' playlists do a very good job of stitching together an engaging listening session from disaparate pieces of content.

The whole is greater than sum of parts for the user in many contexts. Recommendations can help users solve the full problem. A full solution that understands user context can drive value for user and allow the provider to charge for it. Multi-SKU aggregators/D2C must realize they can get beyond perpetual discounting only with bigger purchase baskets solving entire problems. Customers will pay or tolerate some margin for a full solution while also building loyalty.

For Twilio Verification.