Business problem: Stagnant growth rate
Buying v/s renting was the major dilemma of the target consumer because the Indian market would only get a sense of ownership by Buying.
Aim
Draw insights using web analytics tools like Mixpanel and Google Analytics
To launch an email marketing campaign and identify potential leads
Search for 3rd party marketing vendors
Optimize consumer journey on it's website to maximize conversions
Company Profile
About RentoMojo: RentoMojo is the early adopter of subscription-based commerce in India, catering to furniture and appliances. Their target is to have 100 million users within ten years. Their target audience is millennials with a disposable income of INR 10,000-50,000 and between ages 18- 30 years, who are relocating and want to experience the products but not own them.
Core value proposition offered:
Hassle free experience
Free Delivery for both intercity and intracity
Commitment free ownership
Competitor Analysis
RentoMojo had direct competitors like Furlenco (Quadrant A), brands with similar audience.
Online brands considered by customers like Quikr and Olx (Quadrant B).
Offline markets considered by customers to buy and rent (Quadrant C & D).
Online brands to buy furniture like Pepperfry.com and Urban Ladder (Quadrant B).
Initial Growth Strategy
As an initial solution, RentoMojo offered long term renting for buyers which would cost less. Online and offline strategies were crafted to increase the user base.
Online- Taking the digital route
Utilized Facebook to reach out to those who shifted to new cities, changed cities, and/or changed jobs
Utilized search engine ads to reach people looking for home furnishings
Offline- Broker Network
A broker network was created as 90% of their inventory was either unfurnished or semi-furnished. These brokers would source leads of the people looking for home furnishing to the RentoMojo sales team.
Average Customer Journey and Lead Nurturing Process
Customer conversion took an average of 3-4 weeks. Conversion through SEM where the intent is already built took up to 7-14 days. Whereas display and Social Media channels up to 45-60 days on 1st click attribution.
To showcase the various value propositions of renting, a sequential email/ lead nurturing campaign was created. A detailed customer journey was created by adding UTM tags and incorporating events with clickable elements. For session wise data and funnels at source/ dimension levels, Google Analytics was used. For individual-level data, Mixpanel was used to create funnels. Metrics on Google Analytics for awareness campaign were:
Bounce Rate
Pages per session
Average session duration
Metrics measured for conversion campaigns were:
Conversion rate
Add to cart rate
KPIs to track:
Source/Medium
City
Type of Campaign
Browser level report
OS report
Analysis and Interpretation
Tackling the business problem
RentoMojo had a stagnant growth in the last quarter at 8-10% whereas companies grew more in the last quarter as per the general trend. The growth became stagnant because customers were not moving to new houses, customers were not moving to new cities and only travelling in this period, and spending on buying instead of renting.
Channel Analysis
Facebook and SEM were stagnant, Display ads were not generating results. Email was to be explored, thus added email partners to source leads. Email was a new channel for the company and wasn't tried in the past. Four email partners were hired on a CPC model.
Netcore
Intellect
iCubes
OMG Network
The providers worked with numerous networks, sourced leads through various channels, and categorized them, based on their interaction, into different segment groups. Based on RentoMojo's past data, few KPIs were undertaken for these partners to qualify with an average CPC at Rs. 7. The primary KPIs for email vendor qualification were
Bounce rate less than 40%
Average session duration more than 2mins
More than 3 Pages per visit
Additional quality KPIs were:
Cost per Product view
Cost per Add to Cart
Add to Cart Rate
Protection against Bot traffic
To tackle the bot traffic problem. in-depth KPIs were used, like:
Bounce rate
Average Session Duration
Pages per Visit
Product View Rate
Add to Cart Rate
Acceptance Payment Rate
Benchmarks were setup using past data to churn out email-partners based on:
Add to cart rate greater than 3%
Product view rate above 15%
Later, another qualification KPI was added which marked that 80% of traffic should be new users. The criterion was added because the email partners had limited leads and the same leads were visiting the website again.
Mixpanel Lead Scoring
Identified reasons to capture leads intent of purchase and relevance
The leads sent by the vendors had to be grade and scored to be segmented into different groups based on the user journey on the website so that users who showed positive behaviour on the website could be contacted first. The important touch points to segment users were add to cart, products viewed and toggle tenure button.
Grade A: Add to cart
Grade B: Viewed more than 3 products and toggled through different tenure points for price adjustments
Grade C: Viewed less than 3 products but toggled through different tenure points
Grade D: Only visited the site and did not qualify for above mentioned events.
A retargeting list was made through Mixpanel by identifying the users who added to cart. The users were shortlisted based on targeting parameters defined previously i.e. demographic and location. These users had a high purchase intent and the buyer required a nudge which was achieved by urgency creation/incentives to complete the purchase.
Evaluating Traffic Quality
Identified the email partner generating most new leads
Evaluated the email partners based on new users sent by them to the website. The qualifying criteria was 80% and maximum number of new users. The focus here was to choose the email partner who generates maximum new leads by finding the ratio of new users to total users from that partner because acquiring new leads can help in increasing awareness and to build a lead base of potential customers for nurturing them in the future.
Netcore: 4,163/4,834 = 86.1%
Intellect: 2,736/3,457 =79.1%
iCubes: 1,672/2,982= 56.1%
OMG: 1,457/2,254= 65.1%
Identifying Email partner with low intent leads
Classifying the partners on the basis of purchase intent helped us prioritize the customer base, to retarget only those customers where the probability of sale is highest, which is the customers with high purchase intent.
The best measure of purchase intent is the conversion metric 'Add product to cart'. It clearly shows the users purchase intent and require a small push to complete the transaction. Continuing with a customer base that has low intent will thin out the budget and thus increase overall conversion cost. Other conversion metrics that are also applicable are 'Conversion Rate' and 'Revenue Generated'.
Another measure that can be used to analyze purchase intent is the behavior metric 'Average session duration'. The longer a user spends on your website the more engaged they are, this engagement can develop into purchase intent. However, there are limitations to this metric and needs to be combined with other metrics to make correct decisions eg. 'Pages per session', 'Number of sessions'.
Going by either Avg. Session Duration or or all of Add to cart rate, Goal Conversion rate, Revenue or Profit, Pages per session, iCubes and OMG performed the worst.
Another aspect is the percentage of returning users that indicates paid traffic - a malpractice that can be used by email providers to show high CTR, which is detrimental for the business.
Identifying email partner bringing best quality traffic, users with higher engagement
To analyze user engagement, Average session duration is a behavior metric. The longer a user spends on the website, the more engaged they are, which leads to purchase intent. However, there are limitations to this metric and needs to be combined with other metrics like Pages per session, Number of sessions, to make correct decisions. Intellect had the longest Average session duration of 6.73 mins.
Identifying email partner with the highest cart abandonment
Users who had added product to cart clearly showed the highest buying intent and were scored the highest. Users with abandoned cart need a small push in terms of urgency creation or incentivization to complete the transaction. Retargeting this pool has the highest the probability of sale. The cart abandonment rate has to be correlated with the conversion rates of the email providers. If they have a good conversion rate too, then the email partner is already doing well at converting users. To drive conversions these customers can be enticed with some discounts.
Cart abandonment rate = [(Add to cart rate – Conversion rate)/Add to cart rate]
Based on the formula above, Netcore had the highest cart abandonment rate.
Interpreting the Consumer Funnel
Using Mixpanel visualization report, it was observed that the sign-up page appeared at a later stage in the consumer journey. Redesigning the website to bring the sign-up process earlier in the consumer journey is quite feasible since a large number of people initially arrived at the home page and dropped off by the time they came to the sign-up step. Hence, shifting the sign-up page to an earlier location will help in increasing the sign-up rate significantly.
Conclusion
Segmented users by their purchase intent using lead grading and scoring based on user journey and qualifying parameters from demographic, location and behavior factors.
Placed sanity metrics to evaluate the quality of traffic by checking new visitor percentage, and identifying low intent leads, highest engagement and highest cart abandonment rate clubbed with conversion rate for the email partners.
After detailed analysis of the four email partners we concluded that the Netcore and Intellect were the best options going forward for RentoMojo.
Improved sign-up rate by using consumer funnel through the analytics tool.
Disclaimer: The data used on this RentoMojo project was shared with me for educational purpose by Upgrad. The analysis are based on the particular dataset and are my own interpretations of that data.