GSN + GDN

CPC - Manual CPC - Keyword

Max Clicks (Automated CPC) - campaign

Branding -
GDN -
vCPM - Viewable cost per thousand impressions
Viewable cost per Mile / thousand impressions
GSN -
Target Impression Share - Mobile, brand related keyword
100% - Impression share

Convesions - filling the form, buying, calls, message, reading the blogs
Conversion - Smart Bidding - historical data - conversion tracking should be enabled
Ecpc - Enhanced CPC - where a conversion is expected, ads will increase cpc
Max. Conversion
Max. Conversion Value
Target CPA - 200/conversion - 220, 240, 200, 180 - Target cost per Acquisition
ROAS - Return On Ad spend

CPV - video campaign - cost per view

video campaign - cost per thousand impressions (vCPM)
rich media ad - light box ad - GDN - CPE - Cost per engagement

ROAS
1. your target campaign / group has a clearly defined profit margin
2. The campaign / group receives a good number of conversions and at a good frequency - atleast 20 conversi
3. Your target ROAS is realistic

(revenue / spend ) = ROAS

SEM spend - 10000

Revenue - 40000 / 80000

ROAS - 4:1
ROAS - 8:1
200%

CPC Traffic
CPM Brand Awareness
CPV / CPE Awareness / Views
Smart Conversions
 max CPC
50/click 4th Online Food 20
Keywords
70/click or 40/click 2nd Order Food Online 30
Veg 50
Ad groups
Nonveg 30
Mobile 50
view - ad - 1 sec - 50% Desktop 30 Device
Tablet 0

Impression Share: Total Impr. / Brand Impr.

Car insurace - 100000
Impression - 20000 20% 30%

Smart bidding

15 conversion in last 30 days 3 months data

More ads Sundays - mobile - 10am - high clicks - high conversions
Less ads Wednesday - desktop - 4pm - high clicks - no conversion

CPC bid of the keyword 5 CPA Bid 5

No. of Cliks received 4500
Total Cost 22,500 Man. Cost 30
No. of Conversions 4000 Adv. 5.63
CPA (Total cost / No. of Conv.) 5.625 Sales 50
Profit 14.37

ency - atleast 20 conversions in last 45 days, higher the better

Laptop A 100000 2 -- 5
Laptop B 80000 10 -- 15
Laptop C 40000 20 -- 10
32 - 40
ks - high conversions
clicks - no conversion
 9 -8 1 7 -0 0 9
REV: NOVEMBER 12, 2019

MARK ROBERGE

TOM EISENMANN

eSig: Growth Analysis

In November of 2015, Kate Dalton, Director of Growth for the electronic signature startup eSig, was
working on eSig’s Q1 2016 plan. The plan needed to specify:

1. The number of new users eSig would acquire in Q1 2016.

2. How much to spend on each major paid marketing channel during Q1.

eSig’s team had run several paid marketing experiments during 2015 and had expanded a few
marketing channels to a reasonable scale. To evaluate options for the Q1 2016 plan, Dalton pulled data
on the results of user acquisition efforts since February 2015 from several sources.

eSig Background
eSig was founded in 2013 and launched a beta version of its product in mid-2014. The beta
designation was removed in February 2015 when the company released an improved product version
and started scaling up marketing. At that point, eSig had about 70,000 registered users, 2,000 of whom
were paying subscribers. eSig had raised $1.2M in seed funding in late 2013 and a $5M Series A in
January 2015. As of November 2015, the company had 20 employees.

eSig’s web browser-based product allowed users to electronically sign a PDF or Word document
using their personal computer, eliminating the need to print, sign, and scan the document. The product
also allowed users to send a document to a recipient with a prompt for an electronic signature. The
company planned to launch Dropbox- and Google Drive-compatible iOS and Android mobile
applications in Q2 or Q3 2016.

eSig’s founders sought to disrupt the electronic signature market with a freemium business model.
Incumbent rivals DocuSign and Adobe EchoSign offered a 30-day free trial. However, neither
incumbent offered a permanently free version with limited capacity. eSig’s team believed that a large
segment of the market needed electronic signature capabilities only a few times per year and a free
offering could capture this segment. If enough free users upgraded to premium subscriptions, eSig
could offset the cost of acquiring and supporting a large free user base.

Senior Lecturer Mark Roberge and Professor Tom Eisenmann prepared this case. It was reviewed and approved before publication by a company
designate. Funding for the development of this case was provided by Harvard Business School and not by the company. Senior Lecturer Roberge
has an interest, including a financial interest, in the startup on which eSig is based. The company's name and certain details in the case and
accompanying spreadsheet have been disguised. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve
as endorsements, sources of primary data, or illustrations of effective or ineffective management.

Copyright © 2016, 2019 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-
7685, write Harvard Business School Publishing, Boston, MA 02163, or go to www.hbsp.harvard.edu. This publication may not be digitized,
photocopied, or otherwise reproduced, posted, or transmitted, without the permission of Harvard Business School.

This document is authorized for use only in Dr. Tejas Shah's MBA-FT2_DM_TS_29102020 at Nirma University from Nov 2020 to Apr 2021.
817-009 eSig: Growth Analysis

In line with the freemium strategy, eSig’s target market was intentionally broad, and included
anyone with a need to sign documents. Common professionals adopting the product were real estate
agents, lawyers, salespeople, small business owners, and government officials. There were two
versions of eSig: eSig Free and eSig Pro. eSig Free allowed up to three signatures per month, with no
limit during the first month after registration. eSig Pro cost $10 per month and allowed an unlimited
number of signatures.

Dalton’s analysis showed similar retention rates and margins, over time, for user cohorts acquired
through different marketing methods. eSig’s overall gross margin was 79%. The annual retention rate
for Pro users was projected to be 87%. Activated eSig Free users had a projected 55% annual retention
rate, with retention measured as any use of the Free product over the prior three months.

eSig User Experience

To register to use eSig, users navigated to an eSig landing page with a signup form asking for their
name and email address. Once they submitted this information, users were in the “created” user state.
After submitting the form, new users received an email with an embedded activation link. They clicked
on the link to confirm their email address and activate their account. Once users clicked on the
activation link AND sent at least one document using eSig, users were in the “activated” user state. If,
after their first month, users surpassed the maximum 3 signatures per month, they were prompted to
upgrade to the Pro version of eSig for $10 per month. Users who upgraded to the Pro version were in
the “activated-upgraded” user state.

Board Meeting
Every month, Dalton sent eSig cofounder/CEO Tanya Tsai a spreadsheet that collected eSig user
acquisition data from several sources, along with her analyses of the data. Descriptions of each tab in
the spreadsheet are included in Appendix A.

After Tsai reviewed Dalton’s spreadsheet, Tasai sent Dalton the following email:

To: Kate Dalton

From: Tanya Tsai
Date: November 7, 2015
Subject: Analysis for board meeting

At the upcoming board meeting, we’ll be discussing our growth plan for Q1. To get ready
for the meeting, I need your input on a bunch of questions:

1. For each major marketing medium — direct traffic, email, organic search, and paid
ads — how have conversion rates from accounts created to paid upgrades been
trending? What are your hypotheses about factors behind conversion rate trends?
2. What might explain changes in the number of accounts created via direct traffic over
time? In accounts created via organic search?
3. Does eSig benefit from virality? What is your estimate of the viral coefficient for an
eSig Free account? What can we do to increase virality?
4. Have our investments in paid marketing been yielding a good return?

This document is authorized for use only in Dr. Tejas Shah's MBA-FT2_DM_TS_29102020 at Nirma University from Nov 2020 to Apr 2021.
eSig: Growth Analysis 817-009

5. Our contractors have produced a steady stream of 1–2 blog posts per week over the
past year at a total cost of $5,000. Was this a good investment?
6. What’s your current estimate for the lifetime value of an eSig Pro customer? What’s
the most we should spend on paid marketing to acquire a Pro customer?
7. How confident are you that we can profitably triple our account base in 2016?
8. What additional research, analysis, and experiments do you plan to do over the
coming weeks to boost confidence in our growth projections?

It’d be great if you get me this input ASAP. I know it’ll take you more time to complete the
Q1 projections we’ll present to the board. As a first step, I’d like you to structure your
projection into two parts:

1. Generate a baseline projection that shows the number of Pro customers we’ll have on
March 31, 2016 if we cease all paid marketing, effective immediately.
2. One of our board members thinks we should start fundraising again at the end of Q1
and thinks that the more Pro customers we have when we start, the more we’ll boost
our valuation. To give the board a feel for what we could do if we follow this
strategy, I’d like a plan from you that maximizes the number of Pro customers we’ll
have as of 3/31/16, assuming we spend exactly $1 million on paid marketing
between now and the end of Q1. Show me how you’d allocate the $1 million, by
marketing method and by month. Don’t worry for now about whether the marketing
investments have a positive NPV—I’ll sort out with the board whether we want to
try to boost valuation with a negative NPV growth strategy.

As Dalton finished reading Tsai’s email, she was simultaneously overcome with the mixed emotions
of pride and anxiety. With nearly a year of data captured on the eSig funnel, the countless nights she
spent establishing a framework to analyze the data was finally being put to use at the board level. This
exposure was exactly the impact she hoped to have on eSig. At the same time, she realized how critical
an accurate forecast was on the business. Dalton cleared her calendar for the rest of the day and got to
work on Tasi’s request.

This document is authorized for use only in Dr. Tejas Shah's MBA-FT2_DM_TS_29102020 at Nirma University from Nov 2020 to Apr 2021.
817-009 eSig: Growth Analysis

Appendix A: Description of each tab in eSig spreadsheet

All data in the tabs described below is from eSig users created between 2/1/15 and 11/18/15. The
current state of the data is as of 11/18/15.

Tab 1 Funnel By Month (Users): This tab illustrates all users created, activated, and upgraded
between 2/1/15 and 11/18/15. The data is organized by the month the user was created in. For
example, in March, 35,402 users created an eSig account. At the time of the case, 27,225 of those users
had activated their accounts and 1,469 of those users had upgraded to eSig Pro. If more users from the
March cohort activate or upgrade their account, this row of data will be updated.

Tab 2 Funnel By Month (Conv%): This tab illustrates the conversion rates between user states.
The conversion rates are as of the time of the case. The data is organized by the month the user was
created in. For example, of the 34,949 users created in April, 82.2% had activated their accounts and
4.29% had upgraded their accounts, as of the time of the case. Of the April users that had activated
their accounts, 5.21% upgraded their accounts. If more users from the April cohort activate or upgrade
their account, this row of data will be updated.

Tab 3 Funnel By Month (Cohorts): This tab illustrates the time it takes from the date an account is
created to the date an account upgrades to Pro. The data is organized by the month the user was
created as well as the tenure month in which the user upgraded to Pro. Both the numerical and
percentage results are illustrated. The numerical, non-cumulative percentage, and cumulative
percentage results are illustrated. For example, in May, 45,612 users were created. In the same month
, 256 (0.56%) of those users upgraded to eSig Pro. In June, 492 (1.08%) of the users created in May
upgraded to eSig Pro. In July, 213 (0.47%) of the users created in May upgraded to eSig Pro. The final
column of each row illustrates partial data from November and will update as November progresses.

Tab 4 Funnel By Medium (Users): This tab illustrates all users created, activated, and upgraded.
The data is organized by the month the user was created in as well as the Medium through which the
user was acquired. For example, in June, 5,518 created users were acquired through Email Marketing.
At the time of the case, 3,824 of those users had activated their accounts and 138 of those users had
upgraded to eSig Pro. If more users from the June cohort of users acquired through Email Marketing
end up activating or upgrading their accounts, this data will be updated. Descriptions of the Mediums
through which users were acquired are as follows:
 BLOG: Visitors from the eSig Blog.
 CONTENT: Visitors from links embedded in eBooks, slideshares, webinars, and other high
value content marketing available for download on eSig.com.
 CPC: Visitors from pay per click campaigns on Facebook and Stumbleupon
 DIRECT TRAFFIC: Visitors that navigated to eSig.com directly by typing in the URL into a
web browser or visitors from a medium that the analytics software was unable to detect.
 EMAIL_MARKETING: Visitors from email marketing campaigns. eSig had not paid for
any email lists; most emails were addressed to the recipients of documents sent by eSig
users. Email addresses were also acquired via content downloads described above. A
source of "esig_automation" denotes visitors from lead nurturing campaigns. A source of
"esig_email" denotes visitors from one-time email marketing campaigns.
 IN_APP_NOTIFICATION: Visitors who signed up immediately in response to a call to
action upon receiving a document signature request from an existing eSig user.
 PAID: Visitors from paid ad campaigns. The various sources are paid networks the visitors
originated from.

This document is authorized for use only in Dr. Tejas Shah's MBA-FT2_DM_TS_29102020 at Nirma University from Nov 2020 to Apr 2021.
eSig: Growth Analysis 817-009

 REFERRALS: Visitors from non-paid links on other websites. The various sources were the
websites the visitors originated from.
 SHARE_LINK: eSig offered users a free month of eSig Pro for every new user they
influenced to sign up. To receive the referral bonus, the user had to upgrade to a Pro
account.
 SOCIAL_MEDIA: Visitors from non-paid links on social media. The various sources were
the social media websites the visitors originated from.
 OTHER: Visitors from a medium that the analytics software was unable to detect.

Tab 5 Funnel By Medium (Conv%): This tab illustrates the conversion rates between user states.
The conversion rates are as of the time of the case. The data is organized by the month the user was
created in as well as the Medium through which the user was acquired. For example, of the 5,518
created users acquired through Email Marketing in June, 69.3% had activated their accounts and 2.50%
had upgraded their accounts, as of the time of the case. Of these users that had activated their accounts,
3.61% upgraded their accounts. If more users from the Email Marketing June cohort activate or
upgrade their account, this data will be updated.

Tab 6 Funnel By Medium (Cohorts): This tab illustrates the time it takes from the date an account
is created to the date an account upgrades to Pro. The data is organized by the month the user was
created, the Medium through which the user was acquired, and the tenure month in which the user
upgraded to Pro. The numerical, non-cumulative percentage, and cumulative percentage results are
illustrated. For example, in June, 5,518 created users were acquired through Email Marketing. In the
same month , 10 (0.18%) of those users upgraded to eSig Pro. In July, 60 (1.09%) of these users created
in June upgraded to eSig Pro. In August, 26 (0.47%) of these users created in June upgraded to eSig
Pro. The final column of each row illustrates partial data from November and will update as November
progresses.

Tab 7 Funnel By Paid Source (Users): This tab illustrates all users created, activated, and upgraded
from the Paid Medium. The data is organized by the month the user was created in as well as the Paid
Source through which the user was acquired. For example, in July, 17,591 created users were acquired
through paid campaigns on Facebook. At the time of the case, 15,334 of those users had activated their
accounts and 224 of those users had upgraded to eSig Pro. If more of these users end up activating or
upgrading their accounts, this data will be updated.

Tab 8 Funnel By Paid Source (Conv%): This tab illustrates the conversion rates between user states
for all users acquired from the Paid Medium. The conversion rates are as of the time of the case. The
data is organized by the month the user was created in as well as the Paid Source through which the
user was acquired. For example, of the 17,591 created users acquired through Facebook paid
campaigns in July, 87.2% had activated their accounts and 1.27% had upgraded their accounts, as of
the time of the case. Of these users that had activated their accounts, 1.46% upgraded their accounts.
If more of these users end up activating or upgrading their accounts, this data will be updated.

Tab 9 Funnel Paid Source (Cohorts): This tab illustrates the time it takes from the date an account
is created to the date an account upgrades to Pro. The data is organized by the month the user was
created, the Paid Source through which the user was acquired, and the tenure month in which the user
upgraded to Pro. The numerical, non-cumulative percentage, and cumulative percentage results are
illustrated. The data is non-cumulative and is limited to users acquired through the Paid Medium. For
example, in July, 17,591 created users were acquired through paid campaigns on Facebook. In the same
month , 24 (0.14%) of those users upgraded to eSig Pro. In August, 119 (0.68%) of these users created

This document is authorized for use only in Dr. Tejas Shah's MBA-FT2_DM_TS_29102020 at Nirma University from Nov 2020 to Apr 2021.
817-009 eSig: Growth Analysis

in July upgraded to eSig Pro. In September, 42 (0.24%) of these users created in July upgraded to eSig
Pro. The final column of each row illustrates partial data from November and will update as November
progresses.

Tab 10 CAC by Paid Source: This tab illustrates the spend on each Paid Source in each month. The
tab also illustrates the cost per created user and cost per upgraded user, organized by Paid Source and
month in which the user was created. The analysis is as of the time of the case. For example, in August,
eSig spent $8,645.56 on Google Ad Words, which resulted in a cost per created user of $6.30 and a cost
per Pro upgrade of $125.30. If more users from the August cohort upgrade to Pro, the cost per Pro
upgrade will be updated.

11 (RAW DATA) User Data: This tab illustrates the current state of all users. Because the data from
this tab is summarized in the other tabs, analysis in this tab is not necessary for the questions posed in
the case. The rows in the sheet represent users created on the same day via the same Medium and
Source that are currently in the same user state. For example, on 4/12/15, there were 2 users that found
eSig via a Medium of Paid and a Source of Facebook that ended up activating their accounts 1 day later
and upgrading their accounts 67 days later. This user cohort is represented in row 36,004 of the User
Data Sheet. A description of the columns in the tab is as follows:
 Column A, labeled “create_month”, showed the month the user account was created.
 Column B, labeled “create_day”, showed the day the user account was created.
 Column C, labeled “source”, represented different points of origin for each medium.
 Column D, labeled “medium”, showed the marketing method through which the users
originated. Mediums include Paid, Referral, Organic, etc.
 Column E, labeled “days_until_activated”, showed the number of days between the date
the user created an eSig account and the date the user activated the account.
 Column F, labeled “days_until_upgraded”, showed the number of days between the date
the user created an eSig account and the date the user upgraded to eSig Pro.
 Column G, labeled “end-status”, showed the user’s state as of 11/18/15. End-status did
NOT reflect attrition in the case of Pro users who had terminated their subscriptions; these
former users were still included in total users counts for “2-activated-upgraded” cohorts.
Likewise, inactive Free users were still included in total users counts for “1-activated”
cohorts. A separate eSig system tracked user retention, and Dalton had not yet integrated
that system’s output into the data and analyses presented in her Excel sheet.
 Column H, labeled “count”, showed the number of users in the cohort.

11 (RAW DATA) Spend Data: This tab illustrates the daily spend and results for each paid
marketing campaign. A “campaign” was defined as the use of a specific paid marketing source on a
given day. The end-dates for “Cost Data” (10/25) and “User Data” (11/18) differed because Dalton had
exported the respective data from different sources; lags in receiving marketing spend figures from
some ad networks delayed the availability of data for recent campaigns. Dalton noted the need to take
care when examining results for November derived from “User Data” and for October using “Cost
Data,” respectively. Because the data from this tab is summarized in the other tabs, analysis in this tab
is not necessary for the questions posed in the case. A description of the columns in the tab is as follows:
 Column A, labeled “Date”, showed the day of the campaign.
 Column B, labeled “Impressions”, showed the number of times that ads associated with
the campaign were viewed.
 Column C, labeled “Clicks”, showed the number of times the ads associated with the
campaign were clicked upon, sending the visitor to eSig’s site.
 Column D, labeled “Spend”, showed the total dollars spent on a given campaign.

This document is authorized for use only in Dr. Tejas Shah's MBA-FT2_DM_TS_29102020 at Nirma University from Nov 2020 to Apr 2021.
eSig: Growth Analysis 817-009

 Column E, labeled “Created Users”, showed the number of eSig users created from the
campaign; account creation always occurred on the same day as the campaign
 Column F, labeled “Activated Users”, showed the number of eSig users created through
the campaign who eventually activated accounts.
 Column G, labeled “CTR”, showed the click through rate for the campaign, calculated by
dividing clicks by impressions.
 Column H, labeled “CPC”, showed the cost per click for the campaign, calculated by
dividing spend by clicks.
 Column I, labeled “CPM”, showed the cost per thousand impressions for the campaign,
calculated by dividing spend by the number of thousands of impressions.
 Column J, labeled “CVR”, showed the conversion rate of clicks to users created, calculated
by dividing users created by clicks.
 Column K, labeled “AVR”, showed the conversion rate of users created to activated users,
calculated by dividing activated users by users created.
 Column L, labeled “CPCU”, showed the cost per created user, calculated by dividing
spend by created users.
 Column M, labeled as “CPAU”, showed the cost per activated user, calculated by dividing
spend by activated users.
 Column N, labeled “Source”, showed the channel through which the marketing campaign
was executed.
 Column O, labeled “Day of Week”, showed the day of the week on which the marketing
campaign was executed.
 Column P, labeled “Month”, showed the month in which the marketing campaign was
executed.
 Column Q, labeled as “Year”, showed the year in which the marketing campaign was
executed.
Source: Company documents.

This document is authorized for use only in Dr. Tejas Shah's MBA-FT2_DM_TS_29102020 at Nirma University from Nov 2020 to Apr 2021.