AKAshutosh Kumar: Our organization and offering is data centric, so quite naturally analytics is at the centre of our organization. Capillary focuses on productizing the analytics to build a more scalable and deployable model. The goal is to use technology to such an extent so as to provide the retailers and end customers of our product the power to see automated results and insights of their own data. We want to convert the analytics and CRM business of a retailer from a cost centre to a revenue centre, and thus we are very much ROI focused. Also, the focus is on providing an end to end solution to the retailer where we take care of every aspect of CRM during our engagement, from data capture to analytics, technology, campaign execution, delivery and consultancy.
AIM: Please brief us about some business solutions you work on and how you derive value out of it.
AK: We are a product-focused company and we execute most of our analytics using the products and technologies developed in house. One of our solutions is the Instant Vouchering System which analyzes the past customer data as soon as a bill is entered in the system and the data is sent to our servers on the cloud. The analytics engine then executes a targeted campaign immediately based on customer’s historical transaction data when the customer is still in the store and the voucher code is sent to the customer instantaneously, which can be redeemed through our product installed at the retailers POS. This is just an example of how we develop and deploy analytical solutions using cutting edge and patented technology.
There are few more solutions like lifecycle, micro-segmentation and loyalty program management. The key focus is to derive value out of it so that we can show the retailers a 3x-6x ROI during the pilot phase itself. The entire value proposition depends on the quality of insights which we use to drive the promotions and campaigns
AIM: How does a typical requirement gathering to delivery cycle looks like for you?
AK: We have separate sales, technology, analytics and delivery teams apart from a dedicated account manager for any client. All teams work closely with each other to understand the requirement, which could be a combination of product and/or analytics plus execution. The execution part is important because as mentioned earlier we make sure analytics/CRM does not remain only a cost centre but also becomes a revenue centre.
The mode of engagement is generally a 3-4 months pilot where we show ROI to the client which also acts as a proof of concept. If the pilot is successful, we roll out a long term plan of engagement in consultation with the client and our internal teams. Both in pilot and rollout phase, the focus is on solving the problems/issues faced by the client and business objectives like increasing sales / improving customer retention / increasing frequency of purchase/ increasing avg ticket size etc. Once the business problem is understood and we obtain the data, we figure out the part which we have already built in our automated system, and the part for which manual effort will be required. Typically this ratio is around 75:25 in the pilot phase which gets reduced to as much as 95:5 in the rollout phase.
AIM: Please brief us about the size of your analytics division and what is hierarchal alignment, both depth and breadth.
AK: The analytics division at Capillary is approximately 60 members strong. The hierarchy is pretty flat, with the teams divided into different geographies based on client location, and a manager heading each geography (India or international market). Being a product company, Capillary also has a separate technology team which creates solutions aimed at automating the reporting and execution part and developing the in-house analytics algorithms into a deployable and marketable product.
AIM: Would you like to share any example of an Insight that generated a huge positive impact for your clients?
AK: I would state an example of a pizza delivery chain which is one of our biggest clients across India and South East Asia. We had observed a distinct trend in the customer’s data, where the customers stick to their behavior in a lot of cases, in terms of time of order, size of order and the frequency of order. We created specific segments of customers based on their spending pattern, and also on the type of pizza & sides they order. Marrying these two segments gave us a clear profile of the customer w.r.t taste segment (what type of pizza he likes, how much side orders, loyal to one type of pizza, etc) , and behavior segments (when does he/she orders it, with the range of spend, when was the last time he/she ordered). We further drilled deep into the data to figure out based on the past transactional history when is the customer expected to make another transaction, and what type of offer he/she would prefer. Few more filters and constraints were applied, and we ended up creating 17,000 unique micro-segments for pizza customers.
Now the more challenging part – how to manage so many segments? This is where we get help from technology team which has created an automated Lifecycle Engine that is used to report, manage & execute distinct campaigns for all the 17k micro-segments. This product is scalable and is deployed across clients in Capillary.
By implementing this solution for the client, we have been able to achieve a 5-6% SSTG (Same Store Transaction Growth) consistently per month for the last one year and more, which translates to a huge jump in sales and transactions without bombarding the end customer with too many promotions, and keeping marketing costs to a minimum.
AIM: Do you think it’s possible to become too married to the data that comes out of analytics? Where do you draw the line?
AK: Yes, people who work in this industry are typically those who eat, drink, and live by the data. A lot of times people have the notion that data based results are the ‘holiest of all’ whose authenticity cannot be questioned because data does not lie. However, there is another angle to it. Good analytics means good insights and good recommendations to the client, and good recommendations will only happen if the recommendations are based on both the data insights as well as business knowledge that might or might not be present in the data. A lot of data scientists believe that the results should be factual and 100% based on data only. I believe it has to be a 50-50 combination of data insights and gut feeling / business sense to make a really strong recommendation.
AIM: What are a few things that organizations should be doing with their analytics efforts that most don’t do today?
AK: Focus should be on quality rather than on quantity. Whatever reporting can be automated using widely available tools should be automated which can save a lot of time. Organizations should also look into how they want to scale up, especially for startups in this field. If they can build a product around their analytics offering, or standardize their analytics offering it will become much easier for them to sell, execute and scale up.
AIM: What are some of the data measurement points that are becoming more important for organizations?
AK: In the customer centric analytics, one has to go with both the attitudinal as well as behavioral data. This means that a retailer, while analyzing its customers has to take into account data from various streams, transactional as well as social/mobile data to paint a better picture of its customer and his/her behavior. This is becoming increasingly important due to high influence and penetration of social media and due to its importance as a media of communication.
AIM: What are the most significant challenges you face being in the forefront of analytics space?
AK: There are a lot of challenges. It starts with the scarcity of trained analytics talent in the Indian market. There are a lot of companies that do analytics and hire people for the same, but I can say that at least 50% of the analytics industry workforce work on reporting, not analysis. This leads to a lot of people who have worked in analytics industry, but a shortage of people who have actually done good analysis themselves and are trained in it.
Another challenge that we face specially in India is the lack of understanding among the clients w.r.t analytics methodologies, due to which analytics gets undervalued. The Indian clients are most interested in ROI coming out of analytics, and that’s the reason we have developed our business model in the same way. International clients are much more mature in understanding analytics due to the presence of already existing analytics industry in the international market.
AIM: How did you start your career in analytics?
AK: I started my career in Analytics at marketRx (now Enterprise Analytics Practice Cognizant) and worked in US pharmaceutical markets analysis. After that I joined Dunnhumby which works for some of the biggest retailers like Tesco and Kroger. I worked for 3 years in marketRx & dunnhumby combined and then joined Capillary Technologies.
AIM: What kind of knowledge worker do you recruit and what is the selection methodology? What skill sets do you look at while recruiting in analytics?
AK: We recruit at two levels – freshers and experienced professionals. In both the profiles, two things are common – the candidate has to be good with numbers and has a very good aptitude for learning new things at a fast rate. We believe that even if someone does not have prior analytics training but has these two traits, he/she can be trained to be a good analyst. That’s the way we have built our 75% of the team. For experienced hires we also look at the depth and breadth of their analytics knowledge, and how fast they can learn the business model of Capillary.
AIM: How do you see Analytics evolving today in the industry as a whole? What are the most important contemporary trends that you see emerging in the Analytics space across the globe?
AK: We have already seen, and will seeing more of the fact that Analytics is going to be at the centre of decision making in every industry where data is used. The application of analytics is already present in retail, finance, travel, media, banking, sports, leisure & luxury, and in numerous other fields. This is driving growth with a large number of new analytics companies coming up very fast. One thing I can see happening in the next few years is the consolidation of analytics space with a lot of mergers & acquisitions taking place.
Across the globe the focus is rapidly changing into two aspects: from resource driven analytics to a more scalable product based model, and emergence of big data technologies. A lot of analytics companies still work on AaaS (Analytics as a Service) – a resource heavy model where your employee strength grows linearly with your revenue. The way to go ahead and scale it up is to automate and productize the analytics as much as possible which will make it easier to deploy solutions across the clients. However, analytics being a lot of subjective and ‘intelligent’ exercise everything cannot be automated, but at least 60-80% can be. Imagine how much more can your employees do if you free up their 60-80% of time
Another thing is big data which needs no introduction. A lot of research is being done in this area but the applications are still limited, and it is not yet widely applied. However, customers across verticals have realized its importance and potential and are preparing themselves to implement it and getting benefit out of it. This is also important since a lot of data in the next few years is going to come from social media, mobiles, apps and such data heavy sources which will require big data technologies.