Founded in 2015 by IIT-KGP alumni Sachin Jaiswal, Nitin Babel, Shishir Modi, and Keshav Prawasi, Niki.ai has managed to post stellar growth.
As work lives get busier, and time always seems short to check off all the items on the to-do list, most of us could do with a personal assistant. However, economics and logistics don’t always permit that, but there is now a solution.
Niki.ai may not collect your laundry for you, or make that grocery run, but it sure can book movie tickets and cabs, pay utility bills, and initiate local searches and shopping deals. The app was officially launched at YourStory’s TechSparks 2015, and is a Tech30 company.
Based on Artificial Intelligence and Natural Language Programming (NLP), Niki.ai is a chatbot that users can 'talk' to. The app understands human language in the context of the products and services that a user wishes to avail, and initiates actions accordingly.
Now in its third year of operations, the startup has registered exponential overall growth. It has a user base of over 2 million, and in the last financial year, posted 800 percent year-on-year growth in Gross Merchandise Value (GMV), and 56 percent month-on-month growth in transactions, resulting in a revenue growth of 430 percent.
The Ratan Tata-backed startup has carved itself a place in assisting people to 'shop on chat'. Apart from Tata, Niki.ai is also backed by Ronnie Screwvala's Unilazer Ventures and SAP.iO, among other investors.
Niki’s revenue is generated through a channel partnership model, where it gets a commission for every order that it generates for its vendors, and partners include companies such as Redbus, Cleartrip, OYO, Ola, Uber, and Bookmyshow, to name a few.
Talking about the team’s motivation to work on the product, Sachin had earlier told YourStory:
"Many users are not able to keep up with the increase in the number of apps, due to low-memory phones and slow internet connections. There was a need to cut the noise around excess content to offer more personalised suggestions and facilitate decision-making. The answer was a chatbot which could simplify online ordering."
Contributing to Niki’s growth numbers is its conversational interface which remembers customer context, and offers an immersive, natural, personalised and unique experience to every customer.
Owing to this, Niki claims to have been able to achieve high retention and repeat customer rates. The company witnessed a 1.7-fold rise in the average order size during financial year 2017-18, even as the average number of orders placed by users shot up to seven per month.
Currently active on more than 7.5 lakh devices (including partners), Sachin noted that 60 percent of Niki’s users are from Tier-1 cities, and the company is seeing rapid adoption from users in Tier-II cities. The founders attribute this growth to its core value proposition and user interface.
Talking about Niki, Sachin said the company’s growth was significant in terms of technology. "When we started off, we had one architecture patent that we had filed for, and were looking to solve a few niche use cases like phone recharges and cab bookings."
Since its inception, the team has made multiple improvements to launch Niki 2.0, which is more conversational, and can also retain the context of conversations. The team achieved this by building its own NLP system that is able to understand 'conversational English', or English that may not necessarily be full sentences.
Sachin said Niki has filed two more patents related to its dialogue manager, which can retain and switch context during a conversation.
According to Niki.ai, the credit for its current traction goes significantly to distribution partnerships through SDK (Software Development Kit). While many companies in the internet space have struggled with high customer acquisition costs, Niki has collaborated with organisations such as HDFC Bank, HDFC Securities, ICICI Bank, Federal Bank, Karbonn mobiles, and Intex smartphones, among others, which have integrated Niki’s SDK.
From 300,000 last year to over 2 million customers now, the company has been able to grow its base rapidly and at extremely low costs.
"To supplement this with our customer retention strategy, we grew our service base three times in the last FY, giving customers more avenues to make transactions. The average orders per customer grew from around 1.8 last year to seven orders per month now. This has led us to recover the acquisition cost for a user within one month of her starting to use the bot."
It plans to replicate and build further on these strategies to continue on the road to hit a revenue run rate of $7 million to $8 million by the end of FY19.
"One of the advantages of Niki being a bot is it can reside in many environments, independent of Operating Systems or platforms (Android/iOS/Web). This paved a path for us to be able to partner with a brand and make Niki available to all its customers within half an hour."
He added, "Not only is Niki a unique value add to offer, but also works as a great engagement and retention booster."
It thus creates a win-win situation for brands and customers. What stands out is a customer being able to avail Niki's services on multiple platforms and apps, and still enjoy the convenience of making transactions over a consistent, single window conversational interface. Thus, the distribution strategy not only helps Niki save on acquisition costs, but also on overall costs incurred in re-engagement, re-marketing and from other channels.
"We've been able to keep the acquisition cost to a few cents per user," Sachin said.
With Niki being present on their own platforms, these partnerships offer a closet of services like bus, hotel, cab, movie booking, bill payments, recharges and much more on chat.
CEO Sachin Jaiswal says Niki.ai has always been about enabling people to go digital and take advantage of the convenience of buying products and services online over other traditional mediums.
"For this dream to come to life, we knew we would have to give control to the user rather than dictating HOWs and WHATs of the entire product experience. Therefore, with Niki, it all happens on conversations. The traction has been great so far and with millions of interactions, we’ve been able to strengthen our NLP and Machine Learning capabilities, making the text recognition accuracy reach above 92 percent."
Sachin says up next is voice recognition and multilingual support. "One will soon be able to simply talk to Niki in their own regional language to get tasks done. No learning curve, no fields, no buttons and no endless navigations. It will be as simple as talking to a friend," he said.