Understanding Cohort Analysis

Cohort analysis is conceptually pretty simple yet it’s one of the most important and powerful analysis approach a startup can adopt. I had in my earlier post discussed the importance of Lean Methodology for startups to minimize wastage of resources and getting to product/ market fit first before scaling up. Cohorts play a crucial role in helping us understand user behavior on each iteration or improvement to the product. There are plenty of other business questions that can be understood better using Cohort Analysis. To give you some examples:

1) How are the optimizations made to the product in a defined period affecting conversions?
2) Which traffic source is generating maximum conversions?
3) Which source tends to bring in users with maximum engagement on the platform?
4) Are customers acquired via email marketing more likely to repeat purchase or are they more likely to upgrade, compared to those acquired e.g. via AdWords marketing?

And more. Products such as Mixpanel and Kissmetrics enable us to easily create and analyze cohorts. Cohorts have never been a core part of Google Analytics, however there are certain hacks you can do to make it work. Even then there are restrictions to creating different types pf cohorts using GA, for eg: a cohort based on the date of purchase of any product on the website. With the latest update GA does allow one to segment users based on the date of their first visit.

What is a Cohort?

A cohort is simply a group of people who share something in common and is time bound, ie, they had something in common when the grouping was first made. A Cohort is very similar to a segment and often there is a lot of confusion on the difference. To understand better, you can consider a segment as “Employees working in the Marketing Department” while a cohort would be more like “Employees who joined in November 2013”.

Cohort Analysis
Cohort Analysis is very popular in medicine where it is used to study the long term effects of drugs and vaccines:

A cohort is a group of people who share a common characteristic or experience within a defined period (e.g., are born, are exposed to a drug or a vaccine, etc.). Thus a group of people who were born on a day or in a particular period, say 1948, form a birth cohort. The comparison group may be the general population from which the cohort is drawn, or it may be another cohort of persons thought to have had little or no exposure to the substance under investigation, but otherwise similar. Alternatively, subgroups within the cohort may be compared with each other.
Source: Wikipedia

We can apply the same concepts for an online portal/ startup to understand better the different type of users and their behavior on the platform. How we define the cohorts to compare and what we compare about their behavior will depend on the business question we are seeking an answer for. In the case of a Lean Startup, the basic premise is that the product is constantly iterated to find the product/market fit and then iterated on to optimize conversions and scale. This is one of the prime applications of a cohort analysis. We can use Cohort Analysis to compare the users acquired during each iteration and compare their behavior on the platform in terms of retention, engagement, conversions etc. Joshua Porter’s excellent blog post on twitter’s use of Cohort Analysis to track engagement with product improvements is a great example of this.

image

If you look at the fig, it has rows for cohorts ( User acquired during each month is grouped as a separate cohort) and the columns give the engagement or retention figures for the cohort over a 12-Month period. As you can see this is the only manner in which one could clearly understand if the iterations and product improvements which twitter was rolling out on a regular basis was continually improving the engagement on the platform. Under a normal graph where in the cohorts are not present, many a times this picture won’t get reflected as the engagement from the early set of users will mask the engagement metrics of a particular group, be it in a negative or a positive manner.

The above example from twitter represents just one application of Cohort analysis. There are various business questions as discussed earlier that can be answered using cohorts. Let’s first understand the various ways to define cohorts:

1. Cohorts defined by when the user first Visits:
Many a times a user does not sign up or engage the first time they visit a platform. Grouping users based on their first visit will help one to understand the number of touches required before they sign up or engage on the platform and on what product iterations does one increase the conversion or the engagement metric based on the date of first visit. The earlier case study of Twitter is a good example of using cohorts to understand user engagement for a product.

2. Cohorts defined by when the user Converts:
By Converts, I mean any type of conversion or micro-conversion on the platform. It could be signing up, registering, making a first purchase, subscribing to the list etc.

3. Cohorts define by what channel the user was acquired on:
It’s really important to understand the best channels of user acquisition and the behavior of the users acquired through each channel so that one can focus more on the channels that yield best results. Cohorts based on the Channel of acquisition helps in this.

4. Cohorts based on User behavior:
Users can also be grouped based on the behavior they exhibit on the platform. For eg: In case of Zoomdeck, there are users who are frequent visitors and infrequent visitors. Users can be grouped in to various cohorts based on their re-visit rate and engagement on the platform. This is important as it helps us better understand them by having a look at other metrics exhibited by them. For an e-commerce companies one would need to strategize differently for frequent buyers vs infrequent buyers and this can be done better through cohorts.

5. Cohorts based on Customer Lifecycle:
For a platform having a number of stages it’s important to track various metrics like retention, Customer Lifetime Value, Engagement etc. It could be a simple game having various levels and classifying users based on the levels they are in and understanding the various metrics exhibited by these cohorts would help one take better decision to incentivize the users and make them shift levels.

6. Cohorts based on User Characteristic:
There might be cases where one would also want to create cohorts based on certain user characteristics like Men Vs Women, The Country of Origin, Age Group etc to create targeted campaigns or provide customized incentives to improve the engagement, retention or revenue metrics exhibited by them.

We have covered in general the various cohorts that can be created, although I do agree there might be a few specific ones related to the niche you are operating in. Creating cohorts form just one part of the puzzle, the most important part is to use various metrics to understand the behavior exhibited by these cohorts which enables you to take business decisions. There are various metrics one would need to track depending on the niche, type of product and the product lifecycle stage the Product is in.

Metrics most often tracked between cohorts are:

1. Measures of User Engagement:
During the early stage of a product before validation, User Engagement (including activation) and Retention becomes two of the most important metric. Cohorts based on date of first visit/ conversion, enables us to understand how product iteration is improving user engagement or if any changes made to the product has negatively affected engagement. The earlier example of Twitter was about tracking engagement on the platform. Depending on the product you can define what user action is termed as engagement or activation on your platform.

2. Retention:
Just like engagement is important as a metric, any successful product should have good retention figures as well. I had covered the importance of retention and how it affects virality, cost of user acquisition and customer lifetime value in my earlier posts on Virality. Cohorts help us understand retention better by enabling us to accurately define what features and user flows are improving the retention numbers. Funnel tools don’t help us track retention which needs to record user activity over longer periods.

3. Customer Lifetime Value:
Customer Lifetime Value is probably the most difficult metric to track. One of the questions we might want to understand could be the channels of user acquisition that result in giving us the max. value for CLV, the particular activity that drives a user to upgrade plans, split-test different pricing plans to understand the optimum one, features or user flow changes that results in better CLV. All of these can only be understood better using a cohort group as it allows us to track a cohort over a period of time to better understand their behavior on the platform.

4. Measuring long life-cycle events:
A product undergoes many iterations and feature roll-out. It’s impossible to measure long lifecycle events using just funnels. A prime example could be measuring revenues or retention which is typically a long term thing.

Now depending on the niche and the stage of growth your startup is in, you would have to choose the various metric that you need to track and also for the various cohorts we had earlier described. At the end of the day for any product, things finally boil down to user growth, engagement, retention and revenue. Analytics enable us to improve on each of those metric and cohort analysis is a technique that gives us great insights in measuring metric that are typically long cycle.

Cohort Analysis Presentation (Example)

I love this presentation of Cohort analysis (quoted from this Blog post) :

image

What you can see immediately is that the area on the right (Period 5) stacks up the current status with users from Period 1 to Period 4. The really interesting piece of the puzzle comes into play when you are considering what exactly your users represent: active, subscribers, etc. So here is what we can infer from the chart:

  • The height of the chart at Period 5 (at 280) is the number of users currently using (or paying for) our system/app.
  • The individual stacks have a drop-off. As we can see, the drop-off is high in the beginning and then starts to level out but does not go down to zero. Since this is homogeneous across all periods, we can infer that there is something we are doing right: user behavior becomes predictable.
  • For each period 1 to 4, new users were signing up and the number of users from Period 1 makes up 17.8% (50 out of 280) of the users in Period 5.
  • The fall off of users from one Period to the next is higher in subsequent Periods, leveling out at about 25%  of the original sign-ups after 3 periods.

References:

Using Data & Analytic Tools to Better Understand Your Users – Measuring the Right Metrics

Startups be it a product or a services based one, is in an extremely competitive landscape vying for every impression it can get among the millions of potential customers available online. Getting your startup visible or discoverable is one thing, getting them to convert on your website and retain them is an even tougher task with the plethora of services and products that the consumer is forced upon. This is why it becomes so very important for startups to understand each and every activity of the user right from the first time a potential customer/ user discovers their service or product on the web to the point they convert and start coming back to their website.

There are plenty of data that’s available to a startup these days and a vast variety of analytic tools to analyze them as well. A few years back, one would have managed analytics and data tracking using just a Visitor analytics tool like Google Analytics, but that is no more the case now. With growing competition, you have far less room to fail. Based on your website and your requirements you can choose from the various Analytic Tools that’s available to you. More often than not, you would need to have a combination of these tools below to better understand user behavior. The below chart gives you the various classes of Analytic tools and their strength in measuring various parameters:

image

Source: www.moz.com

It is crucial for a marketer to appreciate the insights data can provide on user behavior and take necessary actions to correct and optimize wherever required. It is also crucial for a marketer to measure the right data and understand it’s essence for better improvement of the customer lifecycle on their website.

In my previous post, we had discussed the importance of measuring the right macro metrics. For understanding and validating Product/ Market fit, one needs to measure Activation and Retention. However to completely understand the lifecycle of the Customer one needs to also measure the other three elements: Acquisition, Revenue and Referral.

image

Funnels are a great way to understand user behavior on your website. They are visual, simple and map well to most of the events related to measuring the macro metrics. But Funnels alone have their limitations as well. Imagine if you wanted to measure the impact of repeated product iterations you have pushed out to during a period on the revenue. It becomes extremely difficult to track the same using only funnel, one because the impact on revenue is a long term thing and also because you would need to segment users who signed up during the period when each iteration was rolled out to effectively understand the impact on revenue for the set of users who started off with a particular variation of the product. This is where cohorts play an important part. Think, I would cover cohorts in the next post and explain in detail the methodology to track metrics like retention, revenue, impact of feature iterations on both and more. In this post, we will focus on using Google Analytics in tracking the channels resulting in any of your user interacting with your brand, converting on your product/ service and also on coming back to your product/ service. The Digital Marketing Funnel as represented in the figure earlier can be broken down in to 3 components:

  • TOF – Top of Funnel
  • MOF – Middle of Funnel
  • BOF – Bottom of Funnel

Top of Funnel:

Top of the funnel represents the first interaction a user has with your brand/ product. There are plenty of channels on which the interaction would happen and one would need to optimize for each of the channels the interaction happens on. The best solution is to always focus on at max two of the channels where the interactions seem to be most effective. With the new Universal Google Analytics Tool, you can get the channel details at Acquisition » All Traffic.

image

The above table gives you a good understanding of all the various channels that drive traffic on to your platform. You can export the data to an excel sheet and then use a pivot table to understand what medium acts as the best option to drive first time traffic so that you can focus and optimize for that channel/ medium.

You can drill down further to understand the best referral sources through Acquisition » All Referrals

image

Determining which sites have referred the best traffic to your website is important as it enables you to focus on those channels. You can focus on important parameters like Bounce Rate and Time Spent on site to understand the engagement of the users coming from various channels. Not only that, you can also identify websites that are similar to the ones driving traffic on to your website by doing a search on Google [ Use the search query related:”site name”]or on Similar Web to try and leverage on to the similar audience on those sites to generate traffic. For eg: If weheartit.com is a major referrer to your site, then doing a search for related websites on google gives you these results:

image

The above search result gives you a healthy number of similar sites with similar target audience who would be interested in your site. Refining and cross-posting your contents across these websites can also help you in getting additional traffic. You can even automate a few of these by using a service like IFTTT where you create recipes for simultaneously posting on a number of these platforms.

Remember, it’s always a good practice to tag the various URLs you use to drive traffic from various campaigns on referring sites. You can use the standard URL builder which google provides to generate tags.

By generating campaign URLs, you can identify the source of referrals to your website, whether visitors found the link from within a newsletter, social media post or other marketing campaigns. By naming the three main campaign tagging elements:  source, medium and campaign, Google Analytics will display information about where the referral originated. Simply complete the tool’s three-step form.

Here are just a few examples of valuable KPI data points you might consider tracking as part of acquisition:

  • Organic Search (SEO)
  • Paid Search Marketing (SEM)
  • Social Campaigns
  • Banner Campaigns
  • Links from External Sites
  • Links from Online Videos
  • Email Recipients
  • RSS Subscribers

Another important parameter which you would want to track is the landing page and how you can optimize them for better conversions. Google analytics helps you identify the most important landing pages on your site and the user flow thereafter. This would give you a better understanding on which pages are performing badly and helps you understand what you can do to further improve user interaction on those pages. [Behavior » Site Content » Landing Pages or Content Drill Down ]

image

On Improving weak landing pages:

  • Optimize the content to make it relevant if it’s outdated.
  • If it’s your main landing page, change the message or positioning if required. Use the heatmap tool to better understand the user interaction on the pages and optimize your page accordingly.
  • Make the content more comprehensive so that more people will find it interesting and informative.
  • Build more relevant internal links to the weaker pages to give them more link juice.
  • You can prompt the user to sign-up for email newsletters or at least try and convert them on any of your micro-conversions before the user leaves.

Middle of Funnel:

Middle of Funnel in the Digital Marketing Funnel is the point where in the user is moving from an initial product or brand interaction to a first sale/ to any major interaction on the platform. You might not be able to get a user to convert during this stage but it’s crucially important for companies to target micro-conversions during this stage.

It’s important to track the sources or channels through which the users come back to your site during this stage and it’s also important to measure the paths taken by the users in completing the micro-conversions or goals set on your page. For understanding user paths, GA has an option called Visitor Flow under Audience that visually represents the user path on the website and the drop-offs at each stage. The Visit Flow Report is a nice and a better representation of the traditional click path report. One can view the visitors moving between nodes. One also has the option to view particular segments of users based on region, campaign, traffic source, country etc and their flow/ browsing pattern on the website.

image

You can also create your own funnel for any of the goals you have set using GA to better understand where the users are dropping off. For setting up goals or micro-conversions in your site, you would need to clearly define the business objectives for creating goals (micro-conversions). Few examples of good engagement goals to track:

  • Account signup
  • Email signup
  • RSS subscription
  • Watching video
  • Content interactions (e.g. photo zoom, faceted search attributes, etc.)
  • Product Purchase

The goals would vary based on the type of website you are measuring for. To set up these goals, you can login in to the admin panel of your Google Analytics dashboard and then click on the Goal tab.

image

You have different goal types to chose from: Destination, Duration, Pages/ Screens per visit or Event. In case of an E-commerece website for eg, if the marketer needs to track how many users complete the check-out process, then he/ she would have to chose the type of the goal as “Destination” in the first step. In the second step he/ she would have to define the destination page which would complete the goal (Conversions).

image

For creating the funnel, you would need to specify each step (page) the user traverses before completing the final goal. The funnel visually represents each stage in the micro-conversion process also specifying the drop-offs at each stage. You can create, based on your requirements, multiple mini-conversions and funnels to better understand user flow during this middle stage of user lifecycle.

image

[Fig: A funnel representation of a goal set to White paper Downloads from the start page clearly indicating the conversions and drop-offs at each stage.]

In the middle of the funnel (MOF) for the Digital Marketing Funnel, it’s also important to analyze the most effective and popular channels that bring the user back. For this, GA provides Multi-Channel attribution tools under the “Conversions” section. There are various attribution models one could use. For a full guide refer this. The Linear Attribution Mode, which gives equal weightage to any channel in the funnel irrespective of where it appears,  gives us great insight in to which channel accounts for the most revenue overall. You can use the Model Comparison Tool in GA to find this out:

image

For figuring out the most popular channels in the MOF, we would have to do some manipulation using excel to weed out the first and the last interaction channels.

Bottom of the Funnel:

The bottom of the funnel is the last touch before someone buys. These channels are very important as it let’s you identify which channels to focus on to complete conversions. You can find this data in Conversion > Attribution > Model Comparison Tool and select your model as the Last interaction.

image

You can use these data on the best channels for driving traffic on to your website to further improve and optimize.

Segmenting:

In addition to standard segments that are available in GA to chose from ( You would have noticed this when we discussed the User Flow path), there are also a wide variety of custom user segmenting options that lets you better understand each set of users. You can create your own segments from the dashboard by clicking on the drop-down next to the All Visits tab that’s present as default. GA with the latest update now has the ability to segment visitors and not just visits, which is something GA lacked compared to tools like Kissmetrics and Mixpanel.

Now click on the Create Segments Icon to define your segments. There are a wide variety of parameters you can use to create segments or else you can use any of your own created events as well to define a segment.

image

image

Refer this post for a great list of custom advanced segments which you can use.

Using segments, you can slice and dice your audience in ways never imagined before. You can create segments based on first purchase value, browser being used, platform being used, device on which the visitor opened the site, purchase value during a period etc. I can very well use this data to do a cohort analysis which is very important at an early stage especially if you are on a lean methodology and constantly iterating, measuring the behavior of the set of users who come in during each of these iterations. Even otherwise, there is tremendous amount of insights analyzing segments will give you. I will cover Cohort analysis in detail in the next post.

Making your products Viral : Understanding Virality

One of the major challenges for a marketer or an entrepreneur is to get users and grow for an eternity. Paul Graham would tell you that you ain’t doing it right if you are not growing by a minimum of 5-7% Week-on-Week. And there are plenty of channels one could use to grow, be it the Press, Text Ads or Visual Ads, Partnerships. All of these techniques require money to be spent proportionally to the amount of visits/ click throughs or conversions you are going to get. Wouldn’t it be so much better if we could get hundreds of users for an eternity for virtually no marketing spend. This is where the inherent Virality of products help.

 What is Viral growth? Viral growth is nothing but an existing user bringing you new users either through a generic invite sent on any of the platforms the potential user is on or by directly using the product ( sharing a file link on dropbox) or by any means possible. Google with gmail was phenomenally successful in creating a viral growth. Google initially started with a base of 1000 people who were given a limited number of invitations to share with friends/ family. Gmail finally went public in the year 2007 but by April, 2006 Gmail had through viral referrals grown phenomenally to a base of 7.1 million users. Quite incredible. Products like Instagram, Dropbox, Youtube etc grew rapidly to a million users through virality.

As with any product the key to being successful in growing virally is to have a world-class product, a product people would love to use and would love to share with their friends. Word of Mouth is a great, free channel for products to grow. But that’s not the only way to build virality in to your products. Look at products that grew phenomenally and you would understand that they built in and utilized at least one or two incredibly viral features in their products. Let’s examine the various viral features a product could have:

1) Inherent Virality : It’s incredibly difficult to achieve this type of virality in all products. There are certain products and niches where the products are inherently viral like gmail or Whatsapp or facebook. These products thrive on users inviting others users because the user gets no value out of them without his families or friends or someone else. But do understand that the easier you make it for a user to invite his friends or family, the more invitations they send out whereby increasing your virality.

image

2) Signature Virality : Remember the messages “sent from my Blackberry” or “Sent from my ipad”? This type of virality encourages people to include the messages as signature because they think it makes them cool. Again, you would need a world class product that people would aspire to use to truly achieve this. Could you imagine someone using the signature “sent from my Nokia?” Kidding. But yeah, the point is to spread the message like Hotmail did with a simple “ Get your free email at Hotmail” signature and grew rapidly from a nominal base to 1 million in 6 months and in the next 5 weeks to 2 million. Remember this was a time when there were only 70 million Internet users and in 18 months they had about 12 million users.

image

Paypal with their autolinks on ebay is another great example. It automatically inserted Paypal logo to the bottom of each of the listings of the sellers who used Paypal. This was incredibly successful in making Paypal grow virally.

3) Incentivized Virality: Companies like Fab.com or Dropbox are great examples of this. They incentivized their users to send invitations to their network for either monetary benefits or extra storage space in the case of Dropbox. It worked and people brought in an incredible number of referral traffic. Think of Affiliates as well. They thrive on this. The company grows and sells products by incentivizing the affiliate marketer to sell more or bring him more buyers. Amazon has achieved an incredible amount of success through their affiliate networks.

My facebook feed is filled with shares from this new to be launched service :Trevolta

image

Of course one is going to share this with their friends, there is no better thing in this world than travelling around the world on someone else’s money! 🙂

4) Embeddable Virality: The biggest example of this is Youtube. Youtube was not the only video sharing website available during its initial stages but what made Youtube a leader was when they made the videos embeddable. People started embedding Youtube videos on their website and with it Youtube amassed massive views and made itself visible to an incredible number of people. This shifted the balance in youtube’s favor and there was no looking back.

5) Social Virality: In this case, Products depend on Social Network like facebook, twitter, pinterest etc to rapidly spread their base. There is a psychology behind Social virality. The key here is always to give people a set of tools to create something awesome which they would want to flaunt with their social graph. Instagram exploded because they could make photos beautiful and people loved flaunting their good looking self to the world. Services like twitter or Scoop.it grew virally because they allowed people to project a certain persona. Even the content shares that are done on any of these networks is in effect a way for a user to project a certain type of persona. If one could get this aspect right, then the product is a sure shot bet to grow virally. What I like about Twitter or Tumblr is the re-tweet or re-blog option which enables a user to create content effortlessly while actually he or she is curating content. It increases engagement on the platform and also gives a sense of satisfaction to the user that he or she is actually creating content.

I guess it’s easy to understand virality but its difficult building virality in to a product and even more difficult trying to measure it accurately.

For measuring Virality, one needs to understand two components:

  • Viral Coefficient
  • Viral Loop time

Let’s assume the scenario where:

image

This implies that each user brings you an additional user within a time frame of 10 days ( the Viral Loop time), which is absolutely incredible if you are able to achieve it! J As we had discussed earlier there are different types of virality and in this case we are assuming a simple scenario where each user is sending out invitations to get their friends in (it could be incentivized or simply because your user loves your product)

Now if we were to look at the growth the product would have by the 20th day:

image

Understanding Viral Loop time is important because Virality is inversely related to it. The shorter the Viral loop time, the better virality one would be able to achieve. Imagine if the Viral loop time in the earlier case was 1 day, ie, each user invites a set of users and the new user signs up all in a day’s time. That would make the user acquisition 5 times faster than the earlier scenario and your table would look like this:

image

Let us plot a graph to understand our growth curve in the first scenario:

image

Assuming a product has a viral coefficient that is equal to or greater than 1, it results in a steep upward growth curve. In reality a product having 1 or a number greater than 1 as its viral coefficient throughout its lifetime is impossible although there might be intervals during which the product shows such a viral coefficient. In reality a viral coefficient of 0.4-0.6 for a product is extremely good. Now let us consider such a scenario where the Viral coefficient is 0.5 assuming the rest of the numbers remain the same from our earlier example.

image

And if we were to plot this on a graph, the growth curve would look something like this:

image

The growth curve flattens out after a particular interval. It’s important for growth hackers and marketers to understand that in reality for most of the viral product this is how the graph would look like if they only depend on user acquisition through virality. So it’s important to plan out the metrics in such a manner that you constantly boost up user acquisition from other channels as well to have a steep growth curve which a product requires to be successful.  Remember Paul Graham and his number for the ideal growth rate for a startup? Utilize not just the virality of the product but also other channels like Press, Market Places , Creation of Viral Content, Paid Advertising or anything that boosts traffic and discoverability of your product/ service which drives conversions in order to maintain an upward trending growth curve.

Now if I were to simply consider the scenario earlier described with users sending ‘n’ invites and x% converts from them giving us a Viral Coefficient of K=n*x%, then the User Base at any particular point of time would be (considering only viral growth):

User Base (t) = User Base(0) * (K ^ (t/vlt +1) – 1)  /  (K-1)

(where vlt is the Viral loop time)

[Reference: David Skok’s article]

The above is not a comprehensive model as there are various things we have left out which includes:

  • The sending invitations process is always staggered. We have just assumed it to happen in one go. If I were to give an example – Imagine dropbox, you will always end up inviting people in a staggered way as you interact with them and share docs with them. It does not happen in one go. And If I were a user of dropbox and If I were to stop using it all together one fine day, then dropbox loses out on any referral signups from me.
  • The churn your product will have as it affects the above mentioned parameter.
  • We have not considered virality across the many channels and the different forms of virality.
  • We have also not included the saturation of a particular channel. If I were using a platform which has a total base of 10Million as the target base for sending out invitations, once I cover the entire user base I can’t rely on the formula.

The Viral Loop

image

The viral loop highlighted in the diagram is what can be called as the single viral loop. It’s important where it’s possible to have a double viral loop to fasten your user acquisition. This is possible especially in case of Social networks. Like we discussed earlier, retention is a key component that defines the Viral Coefficient. An increased retention will result in increased Viral Coefficient and hence a faster user growth. There are simple techniques one could do to improve retention and engagement on the platform. This forms part of the double viral loop. Re-connection always increases engagement and retention and hence it’s important to re-connect people by prompting them as well as by making it easy for them.

For ex on LinkedIn, after we sign up, it prompts us to export contacts from our address books and re-connects us. This removes the friction normally people will have in searching for people and then connecting with them. Also, it helps in retaining dormant users. This is a technique employed by many of the Social networks to bring back dormant users on to the platform. Notifications on follow improves your chances of brining back dormant users.

image

This simple step resulted in an increase of 16% in the number of invitations sent. Check the stats below:

image

Source: http://www.slideshare.net/joshelman/josh-elman-threegrowthhacksgrowconf81413

Or in case of twitter they take you step by step through the various things one can do on twitter and by it helps you in getting content on your feed and making you follow a few popular people on login itself. It alleviates any friction the user will have initially to engage on the platform and also interacting with the popular users sets the context for them to get active. In doing so Twitter achieves more invitations and requests sent to users and prospective users and also re-connections and engagement between existing users. That’s a double viral loop.

image

 image

Similarly, it’s important for any marketer to understand the viral loop of their product, one would have to iterate and measure to understand in detail the parameters and the best possible viral loops.

At Zoomdeck, we are creating a platform for making photos interactive. A User can spot anything interesting inside photos to ask a question or add notes or add spots to highlight an interesting story or experience about an element inside photos, then make it much more engaging by linking the spots to audio, video, products, people, places or any link relevant. Users would be able to discover and share the stories and elements in photos using interactive spots and have contextual conversations around each spot. We have a web version, an iOS app and an embedding option which along with viral content shared across various Social Media sites would be a key driver of traffic and user acquisition for us. When I look at the various channels of virality for Zoomdeck, I have:

  1. Embedding: Bloggers and Publishers embedding Interactive Photos on their website. Similar to how Youtube utilized embedding as an important element of their Viral growth.
  2. User joins Zoomdeck, takes photos and makes them interactive by adding spots. Shares it with their friends and family (Invite). ( This will have a longer Viral Loop time) Similar to how Instagram or Pinterest built their viral loop.
  3. Directly recommends the product to their contacts through the invite option in the app or in person.
  4. Sharing of Interactive Photos they find interesting on Zoomdeck( Content) on Social platforms ( Facebook, twitter or Pinterest). Their network discovers, finds it interesting and shares with their friends. ( This will have a much shorter Viral Loop time) The advantage of having content that is viral in nature is the Viral Loop time significantly reduces as you are providing ready made things for people to share and not asking them to create which is always time consuming and requires an effort and hence would always have friction. A Youtube or Twitter is a great example of this.

The four basic viral loops in the case of Zoomdeck as mentioned above would each have different conversion ratios. While the first option and the fourth option would enable Zoomdeck to reach a much larger base of audience and that too multiple number of times, the conversion percentage is going to be a lot lesser than the second and third option where in our chances of conversions are much higher. Similarly, the Viral loop time for the first and fourth option would be much lesser than the VLT number for the other two. So measure the various parameters continuously and optimize for the ones that give best results.

Importance of Seeding :

Imagine for calculation purpose the current user base of a product as 5000 and consider only the 2nd and the 3rd channels of virality listed above as the growth channels for easiness in quantifying. (Assume a Viral Coefficient of 0.6 and a Vlt of 10 days) We would have a table that looks like this

image

And our user graph would look like this:

image

Now, assume acquisition of a constant number of users from other channels, we have (all values are hypothetical):

image

image

image

Seeding initially is critical as that’s what enables the viral growth to kick in. In the first example we flatten out our user base after 2 months. This is why seeding should always be an ongoing process to leverage maximum value from virality or else we should have a viral coefficient greater than 1 to have an eternal upward curve for our user graph which is very difficult to achieve throughout the lifetime of the product. There would be short bursts when the viral coefficient is greater than 1 and would result in phenomenal growth especially if you are on a larger base as well but not through the lifetime of a product.

Key-points:

  • Virality is something that should be inherent in the product. It’s important to design and incorporate virality during product conceptualization itself.
  • Always measure and track various metrics to understand what works best and dig deep into those channels.
  • Iterate as fast as possible to understand the best viral channels. The longer the iteration cycle, the longer it will take for you to spot your best viral loop.
  • Reduce the number of steps required to do any action that results in virality. Make it as easy as possible for the users to send invitations. Understand that the easier you make, the better your metrics would look.
  • Two factors that influence Virality are: Viral Coefficient (K) and Viral Loop time (Vlt). Increase ‘K’ and decrease ‘Vlt’ for rapid growth.
  • Retention and re-connection are important factors that help in Viral growth.
  • Important to seed users initially.
  • Exponential growth from Virality kicks in after a threshold limit. Make use of various channels for seeding the initial audience.
  • It’s very difficult to achieve sustaining growth through virality where you require a viral coefficient greater than 1. Hence, compensate for this and balance it out by seeding users through other channels as well – if required paid channels also to maintain momentum.

References:

http://www.forentrepreneurs.com/lessons-learnt-viral-marketing/

http://www.linkedin.com/today/post/article/20130402154324-18876785-how-to-model-viral-growth-retention-virality-curves

http://andrewchen.co/2007/07/11/whats-your-viral-loop-understanding-the-engine-of-adoption/

Customer Engagement

Customer Engagement – What are the key metrics to track and why?

B2B SaaS is extremely competitive especially for horizontal SaaS products. If you are in the SMB space then that makes it even more challenging for you to survive and then grow. There are a few important metrics the product needs to track assiduously –

  • CAC ( Customer Acquisition Cost)
  • LTV ( Customer Lifetime Value)
  • Payback Period
  • Churn
  • NPS ( Net promoter Score)
  • Sales Velocity

I’m sure most SaaS companies do track these numbers. The key to success is to reduce Churn, CAC and to increase LTV, NPS. One of the key factors that enable a SaaS product to achieve this is customer engagement. But how do you define and measure customer engagement?

 

What is Customer Engagement?

Customer engagement is the interaction/ activity of your customer on the platform. The customer engagement could be a positive or a negative one and it’s equally important to understand the nature of this engagement.

  • A negative engagement increases the risk of Churn, so there are immediate actions that need to be taken to ensure the customer stays.
  • Similarly, a happy and engaged customer provides you with an opportunity to up-sell or cross-sell.

 

So, how do you measure Customer Engagement?

Measuring customer engagement inside the product is the same process as lead scoring at the top of funnel. I had covered lead scoring earlier. Lead scoring is a top of the funnel score that we use to qualify leads based on their activity or interaction with various assets/ touchpoints of the product. You could measure customer engagement with either of the two options:

(1) Use 3rd part software tools that let you define and analyse various events inside the product. Here are a few tools you could consider

(2) Setup your own system where you log various datapoints in your DB and run queries to analyse the same.

In either case, you would have define the important events of engagement and also assign points for these events which would help you calculate the all important engagement score. The events that need to be tracked would be based on the application. For eg:

Helpdesk Software: Add support email, setup forwarding rules, setup DNS, Added Agent

A/B Test SaaS App: Create Test, Start Test, End Test, Share Results

Online Billing APP: Create Invoice, Send Invoice, Receive Payment

Once you have defined the events you can log them and also assign weights to each of these events to calculate your Customer engagement score.


Customer Engagement Score = (wt1*e1) + (wt2 * e2) + … + (wt# + e#)

where wt is the weight assigned and e represents the event being tracked.


Along with the consolidated user engagement score, you could also monitor certain specific or low level metrics that again define user engagement. A few examples are:

  • Daily Active Users ( DAU)
  • Weekly Active Users ( WAU)
  • Monthly Active users ( MAU)
  • DAU/ MAU Ratio
  • User Retention – Day1, Day7, Day30

The core metric that you need to track varies from product to product/ app to app. It’s for you to decide what numbers matter for your product.

 

What next?

Capturing and understanding these metrics defined above is the first step. Setting up steps to improve on these metrics is the next step. This entire process can be automated using a comprehensive automation tool like Marketo, Autopilot, Hubspot Enterprise etc. The right set of messages at the right time goes a long way in optimizing each of the above metrics.

An example:

Pipefy is a great tool for workflow/ process management. It lets you organize all your processes in one place. On signup up with Pipefy, they send you a set of emails to increase engagement.

One of the first emails that they send is a library of pre-existing templates ( most used ones) which would enable the users to get started immediately.

customer-engagement-1

They track weekly retention and send out a mailer to engage the inactive users. This is the second email they send out to inactive users –

customer-engagement-2

Then they follow it up with this email within a few days:

customer-engagement-3

Another example is how Groove improved customer activation using customer engagement data. Grove is a helpdesk software and one of the first things that a user should do after signing up is to setup a support email. They also measure the avg. time it takes for the user to setup the initial support email and if that doesn’t happen then they send an automated email. Here’s the template they use :

customer-engagement-4

They also track user retention and sends out mailers to inactive users to re-engage them. Here’s the template they use for that.

customer-engagement-5

These are proactive measures you can take to increase engagement and user engagement. You can personalize these messages/ automated communications that go out further by segmenting the data. An eg: For Horizontal SaaS products you get registrations from a bunch of industry verticals. You can further segment the user data based on industry vertical and send relevant use case for the industry/ use terminologies that the prospect could relate to. At FieldEZ, we segment prospects based on Industry and the use cases differ across Industry. FieldEZ is used as a Lead/ Sales management tool in industries such as BFSI, Pharma while it’s primarily used for Ticket Management in the Consumer Durables or Manufacturing industry segments.

Other than this customer segmentation also helps in:

  • Identifying what features matter most to a particular segment
  • Measure LTV, CAC, Payback Period, Churn, NPS etc for each segment and work on optimizing the same
  • Measure profitability of each segment
  • Test separate user onboarding techniques for each segment – Messaging and Core interactions based on what matters to the segment
Fundraising for Startups

Fundraising for Startups

Fundraising is difficult and that’s me being very polite. How many times have you heard a VC tell you – “Your product is really nice and very useful as well. I love it. I just need to see a bit of traction now to convince myself to invest”? If you are a early stage company looking for someone to invest, then I will bet my house on you having heard this from one of the VCs.

A couple of years back there was no shortage of VC money, they were making bets on anything and everything they could find and in doing so many of them have burnt their hands. When I say, burnt their hand, I mean they had literally drowned all of that money down the drain for crazily stupid ideas that would have never made it big, let alone India, anywhere else in the world. That’s changed the investment scenario for the better. The heavily discounted models which a few of these startups were running were extremely difficult to sustain. These models weren’t inherently changing consumer behavior or making consumers loyal to any brand/ product. What they were doing was just delaying the inevitable, that’s running out of money and not finding a backer who will invest at a higher valuation.

With funds drying up the investment firms, including the early stage ones have changed the way they approach investing. These days, Seed stage investing is more or less like the Series-A/ B a couple of years ago. The investors look for genuine traction – Revenues and focus more on profitability rather than GMVs. Gone are the days, when all it took for a startup to raise money was to have a few thousand free users sign up on the platform.

In an ideal world, bootstrapped, cash-flow positive startups would be the rockstars. But there are times when a startup needs fund to survive and grow, the cost of running a startup is often not understood well enough by founders. From my experience, the investors look for the following in a startup:

 

  1. Ability to generate revenue (an existing portfolio of a few paying customers). This is also a confirmation on product/market fit.
  2. Ability to scale – Addressable market
  3. Identified a Channel to Acquire customers – A process that is repeatable
  4. The Team – A balanced set of skillset
  5. Ability to hustle

 

If you feel your product or startup fits the bill and is in need for funding, then go for it. Remember, fundraising is a painfully long process that will drain you out. So ensure you have a support system in place to continue running the business while the founder focuses on raising money. If your startup does not tick off the 5 criteria listed above, then put your foot down and get cracking on these 5 items.