December 9, 2015
Via Electronic Submission
Securities and Exchange Commission
Division of Corporation Finance
100 F Street, NE
Washington, DC 20549
| Attention: | Kathleen Collins |
| Rebekah Lindsey |
| Re: | Endurance International Group Holdings, Inc. |
| Form 10-K for Fiscal Year Ended December 31, 2014 |
| Form 10-Q for the Quarterly Period Ended September 30, 2015 |
| File No. 001-36131 |
Ladies and Gentlemen:
This letter is in response to the comments contained in the letter dated November 17, 2015 from Kathleen Collins, Accounting Branch Chief, on behalf of the staff (the “Staff”) of the Securities and Exchange Commission (the “Commission”) to Marc Montagner, Chief Financial Officer of Endurance International Group Holdings, Inc. (“we” or the “Company”), relating to the Form 10-K and Form 10-Q referenced above. For convenience, the responses are keyed to the numbering of the comments and the headings used in the Staff’s letter.
Form 10-K for Fiscal Year Ended December 31, 2014
Management’s Discussion and Analysis of Financial Condition and Results of Operations
Overview, page 51
| 1. | We note your discussion of year-over-year growth in adjusted EBITDA and the reasons for such growth in the Overview section of your MD&A. We further note your disclosure of adjusted EBITDA and adjusted net income beginning on page 54 where you discuss the factors that contributed to the year-over-year change in these non-GAAP measures. Tell us how you considered the guidance in Item 10(e)(1)(i)(A) to present a discussion of the most directly comparable financial GAAP measure with equal or greater prominence. |
Response:
In the MD&A Overview section of our Annual Report on Form 10-K for the year ended December 31, 2014, we provided the reader with a summary of key financial and operational metrics for 2012, 2013 and 2014. This section sequentially presented revenue, net loss, Total Subscribers, average revenue per subscriber (ARPS) and adjusted EBITDA, such that we presented net loss prior to adjusted EBITDA and adjusted net income.
Securities and Exchange Commission
December 9, 2015
Page 2
Our discussion of the reasons for adjusted EBITDA growth were relatively detailed because, as further discussed in our response to Comment 2 below, we believe adjusted EBITDA is important in understanding the overall health of our business. In addition, the factors contributing to the change in net income (loss) are typically the same as those discussed in the line items of the income statement discussed in each report, and including a discussion of the change in net income (loss) is typically repetitious. Nevertheless, we acknowledge the Staff’s comment, and in our future reports on Form 10-Q and 10-K will continue to disclose net income (loss) and will include further discussion regarding the factors contributing to the year-over-year change in net income (loss), each with equal or greater prominence than our disclosure of, and discussion of the factors contributing to year-over-year changes in, adjusted EBITDA.
We note for the Staff that, beginning with our Quarterly Report on Form 10-Q for the quarter ended June 30, 2015, we are no longer presenting adjusted net income as a key metric.
Non-GAAP Financial Measures and Key Metrics, page 51
| 2. | We note your disclosure on page 52 that the company believes your non-GAAP measures help management and investors gauge your ability to generate cash flow. We further note your discussion of adjusted EBITDA on page 54, where you indicate that highlighting cash collected and cash spent in a given period provides insight to an investor to gauge the overall health of your business. Considering such disclosures, it appears that your measures of adjusted net income and adjusted EBITDA may be both performance and liquidity measures. Please tell us your consideration to also provide a reconciliation of these measures to operating cash flows and in your response, clarify how you contemplated whether the adjustments would require cash settlement. Refer to Item 10(e) of Regulation S-K. |
Response:
We consider adjusted EBITDA to be a performance measure and not a liquidity measure. Similarly, we considered adjusted net income a performance measure and not a liquidity measure prior to our decision, referred to above, to discontinue reporting adjusted net income.
Securities and Exchange Commission
December 9, 2015
Page 3
Our reference to non-GAAP measures helping “investors to gauge our ability to generate cash flow” was not intended to suggest that adjusted EBITDA is a liquidity indicator. Rather, that reference represented language carried over from prior disclosures that had been intended in prior periods to help explain the utility of liquidity metrics such as free cash flow and unlevered free cash flow, which were not included in the reports addressed in this letter. We will delete this language in future reports.
Our statement that “we believe highlighting cash collected and cash spent in a given period provides insight to an investor to gauge the overall health of our business” helps to explain that adjusted EBITDA serves as an operational performance indicator by measuring cash billings and expenses. Like many subscription-based businesses, we use cash billings and expenses, as well as GAAP revenue and expenses, to measure our performance. We believe that evaluating our business on a cash billings basis provides a valuable indicator of the “real time” health of our business, in part because it reflects expansions and contractions in our business more promptly than GAAP measures. Unlike adjusted EBITDA, GAAP revenue only reflects trends in our business over a relatively long period of time, because, under GAAP, revenue is recognized over the term of a customer’s subscription, which is typically 12 months or longer. Adjusted EBITDA provides insight into our cash billings because we add changes in our deferred revenue back to GAAP net income in order to derive adjusted EBITDA. Adjusted EBITDA also allows us to evaluate how our cash collections in a period align with the associated cash expenses during the same period, as opposed to GAAP measures that require us to recognize expenses over the term of the applicable subscription.
We note that we do present three non-GAAP liquidity measures – free cash flow, unlevered free cash flow and unlevered free cash flow, as reported – in our earnings materials, which are not filed with the Commission. (See, for instance, our earnings release furnished to the Commission on Form 8-K under Item 2.02 and 9.01 on November 2, 2015.) For these liquidity measures, we have included corresponding disclosure of GAAP operating cash flow with equal or greater prominence, a reconciliation to operating cash flow, and a description of the limitations of these measures.
Nevertheless, we acknowledge the Staff’s comment, and in our future reports on Form 10-Q and 10-K, we will clarify our disclosure that we view adjusted EBITDA as a performance measure and not as a liquidity measure.
Securities and Exchange Commission
December 9, 2015
Page 4
Results of Operations
Income Tax Expense (Benefit), page 67
| 3. | We note on page 114 that you experienced volatility in your foreign rate differential during 2014. Please tell us your consideration to disclose the impact of changes in foreign earnings on your effective tax rate. Refer to Item 303(a)(3) of Regulation S-K. |
Response:
Our foreign rate differential is affected by the level and mix of income and the statutory tax rates in six jurisdictions in which we operate. Our subsidiaries organized in the United Kingdom and the United Arab Emirates regularly represent the most significant portion of our foreign losses that are subject to foreign tax rates that are lower than the U.S. statutory rate and thus reduce the expected benefit. The statutory tax rate for the United Kingdom is 20% and for the United Arab Emirates is 0%. We acquired our interest in the U.K. subsidiary in the fourth quarter of 2013 and our interest in the United Arab Emirates subsidiary in the first quarter of 2014. As such, the increase in our GAAP net foreign losses in 2014 ($27.6 million) from 2013 ($2.9 million) was primarily due to the timing of these acquisitions.
We considered and disclosed on page 114 of our 2014 Form 10-K, in the tax rate reconciliation, the factors that had an unusual, infrequent and significant impact on our benefit for income taxes for 2014, 2013 and 2012, which included the differential between lower net effective foreign rates and the higher U.S. rate, state taxes, changes in valuation allowances, tax credits, nondeductible expenditures, and legislative changes.
Nevertheless, we acknowledge the Staff’s comment, and in our future filings on Form 10-Q and 10-K, we will include in MD&A, where material, additional information regarding our foreign tax rate differential and the primary jurisdictions affecting such rates, along with a cross-reference to our tax footnote, in substantially the following form (based on the disclosure in our 2014 Form 10-K):
“A significant amount of our GAAP foreign losses are generated by our subsidiaries organized in the United Kingdom and the United Arab Emirates. In 2014 and 2013, the foreign rate differential predominantly relates to these jurisdictions. Our foreign rate differential in 2014 has a negative impact on our expected benefit since the majority of the foreign losses are generated in jurisdictions where the statutory tax rate is lower than the U.S. statutory rate – specifically the United Kingdom, which has
Securities and Exchange Commission
December 9, 2015
Page 5
a statutory tax rate of 20% and represents $22.5 million of our foreign losses, and the United Arab Emirates, which has a statutory tax rate of 0% and represents $6.2 million of our foreign losses.”
In addition, we will describe any known trend or uncertainty pertaining to any such foreign jurisdiction that we believe would have a material impact on our results of operations, and if any additional foreign jurisdiction impacts our foreign rate differential, we will add that jurisdiction to the above disclosure.
Consolidated Financial Statements
Note 2. Summary of Significant Accounting Policies
Revenue Recognition, page 89
| 4. | We note that revenue from the sale of a domain name purchased from a third-party registrar is recognized on a gross basis. Please provide us with your analysis that supports gross revenue recognition and specifically address how you considered each of the factors in ASC 605-45-45. In addition, you state that revenue is recognized upon billing as you have no remaining obligations. Please clarify whether the third-party registrar has any further obligations to the subscriber and tell us how you determined who the primary obligor is in these arrangements. |
Response:
For a significant majority of our domain name sales, we act as the registrar of record, or registrar, which requires us to maintain the named server information and contact names associated with the domain, process renewal requests and pay ICANN quarterly registration fees. When we are the registrar, we spread the revenue and related registration fee expenses over the term of the domain subscription, which is usually a 12-month period or longer. We record the revenue on a gross basis, and we record the various costs to register and maintain the domain as cost of revenue.
In 2014, we recognized approximately $19 million of revenue, or 3% of our consolidated net revenues, from domain sales for which we were not the registrar, which we refer to as third-party domain sales. In the case of third-party domain sales, we resell domains primarily from two different parties, and those parties perform all post-sale registrar services. For these transactions, revenue and the fee paid to the registrar are recognized up front. We record this revenue at the time that we bill the customer, and we record the registrar fee as cost of revenue.
Securities and Exchange Commission
December 9, 2015
Page 6
We have assessed the proper revenue recognition for all of our domain sales under ASC 605-45-45 to determine whether revenue should be recorded on a gross or net basis. We have concluded that revenue for third-party domain sales should be recorded gross for the following reasons: (i) we are the contracting party with the customer; (ii) we have latitude in establishing customer price; (iii) we acquire and deliver the product to the customer; (iv) we have discretion in supplier selection; and (v) amounts we earn are not fixed. In addition, as we have no obligations to perform any services after the initial third-party domain sale, and the customer has all the rights and benefits of ownership after the sale, we believe revenue and related costs are appropriately recorded at the time of sale.
Our assessment of the following criteria set forth in ASC 605-45-45 support recording third-party domain sales revenue on a gross basis:
The Company Is the Primary Obligor
Criterion: Whether the Company or the supplier is responsible for providing the product or service. Representations (written or otherwise) made by the entity during marketing and terms of sales contracts can provide evidence to support who is responsible for fulfillment of the product or service.
Discussion: Customers purchasing third-party domains from us do so through our service platform, pay us via our billing system and agree to our terms of service with us when making their domain purchase.
At the same time, the customer is made aware of the relationship with the third-party domain provider through links in our terms of service. The customer is advised that he or she must adhere to the terms of service of the ultimate registrar. The registrar is responsible for maintaining ownership details for the domain name, paying quarterly ICANN fees, sending various notices to the domain owner regarding contact information and renewal reminders, and maintaining data retention and backups required by ICANN.
Securities and Exchange Commission
December 9, 2015
Page 7
Accordingly, from the customer’s perspective, the actual primary obligor may not be clear. We believe that it is more appropriate to consider other criteria under ASC 605-45-45 to determine whether revenue recognition should be gross or net.
The Company Has General Inventory Risk
Criterion: Whether the Company or supplier has inventory risk before sale or after sale (returns).
Discussion: Generally, in accordance with our terms of service, domain sales are not refundable, and the third-party domain provider will not refund fees that we have paid. However, in some instances we will issue a refund of a domain sale to enhance customer goodwill, or we will incur refunds initiated by a customer through its credit card provider, which are known as chargebacks. When these refunds or chargebacks occur, we are not able to obtain a refund of the domain fees we previously paid to the third-party provider, so we do have refund risk, albeit small. Given the small amount of inventory risk, we do not believe this factor provides clarity on either gross or net revenue recognition.
The Company Has Latitude in Establishing Price
Criterion: Whether the Company or supplier has the latitude, within economic constraints, to establish the exchange price with the customer.
Discussion: We have complete latitude in pricing decisions for all domain sales, and are not bound by any contractual restrictions in how we offer prices for these products. Accordingly, we believe this factor supports revenue recognition on a gross basis.
The Company Changes the Product or Performs Part of the Service
Criterion: Whether the Company changes the product or performs part of the service ordered by the customer.
Discussion: We are responsible for selecting the appropriate supplier, purchasing the requested domain name, paying the supplier and delivering the domain name for use by the customer. After acquiring the domain, any post-sale activities are the responsibility of the third-party domain provider. Both we and the third-party provider have responsibilities in the sale of the domain to the end customer. We believe our involvement in the sale transaction is meaningful, and supports revenue recognition on a gross basis.
Securities and Exchange Commission
December 9, 2015
Page 8
The Company Has Discretion in Supplier Selection
Criterion: Whether the Company has multiple suppliers for the product or service.
Discussion: We currently use two primary suppliers for third-party domain sales, have 11 lower volume suppliers, and could obtain more suppliers. We therefore have discretion in supplier selection, and this factor supports revenue recognition on a gross basis.
The Company Is Involved in the Determination of Product or Service Specifications
Criterion: Whether the Company must determine the nature, type, characteristics or specifications of the product or service.
Discussion: Given the nature of domain sales, we determined this criterion to be non-applicable.
The Company Has Physical Loss Inventory Risk
Criterion: Whether the Company or supplier has inventory risk during delivery.
Discussion: Given the nature of domain sales as web-based, non-physical products, we determined this criterion to be non-applicable.
The Company Has Credit Risk
Criterion: Whether the Company or supplier has risk of credit loss due to failure of customer to pay the amount invoiced.
Discussion: We are generally paid in advance, via credit card or PayPal, which eliminates nearly all credit risk for both us and the domain provider. There is some risk of refund or chargebacks to us, albeit small. The supplier has no credit risk from the end customer and the supplier is generally paid in advance by us. Given the small amount of credit risk, we do not believe this factor provides clarity on either gross or net revenue recognition.
Securities and Exchange Commission
December 9, 2015
Page 9
We also assessed the following criteria set forth in ASC 605-45-45 that suggest recording third-party domain sales revenue on a net basis:
The Supplier is the Primary Obligor
Criterion: Whether the supplier or the Company is responsible for providing the product or service. Representations (written or otherwise) made by the entity during marketing and terms of sales contracts can provide evidence to support who is responsible for fulfillment of the product or service.
Discussion: As discussed above, customers purchasing third-party domains from us do so through our service platform, pay us via our billing system and agree to our terms of service with us when making its domain purchase. In addition, as noted above, the third-party registrar performs all post-sale services.
Given these factors, from the customer’s perspective, the actual primary obligor may not be clear. We believe that it is more appropriate to consider other criteria under ASC 605-45-45 to determine whether revenue recognition should be gross or net.
The Supplier has the Credit Risk
Criterion: Whether the supplier or the Company has risk of credit loss due to failure of customer to pay the amount invoiced.
Discussion: We are generally paid in advance, via credit card or PayPal, which eliminates nearly all credit risk for both us and the domain provider. There is some risk of refund or chargebacks to us, albeit small. The supplier has no credit risk from the end customer and the supplier is generally paid in advance by us. Given the small amount of credit risk, we do not believe this factor provides clarity on either gross or net revenue recognition.
The Amount the Company Earns is Fixed
Criterion: Whether the Company earns a fixed fee or fixed percentage on each sale.
Securities and Exchange Commission
December 9, 2015
Page 10
Discussion: We do not earn a fixed fee or fixed percentage on each sale. Rather, we have complete latitude in setting prices, and we pay a fixed per-unit price to the third-party provider for the domain purchase and services. Accordingly, we believe this factor supports revenue recognition on a gross basis.
Based on our assessment of the criteria for gross or net revenue reporting in ASC 605-45-45, we have concluded that the preponderance of the criteria support recording third-party domain sales revenue on a gross basis, and further, since we have no obligations to perform any services after the initial third-party domain sale, and the customer has all the rights and benefits of ownership after the sale, we believe revenue and related costs are appropriately recorded at the time of sale.
| 5. | Please clarify for us which of your products or services are subject to refund or cancellation. Also, to the extent material, tell us your consideration to disclose the impact of refunds or cancellations on revenue for each period presented and to include a discussion of your accounting policy for estimating refunds in the Summary of Significant Accounting Policies footnote. |
Response:
Our refund policies vary across our several brands. Most of our brands, however, provide for refunds during the first 30 to 45 days following purchase, with only the hosting portion of the purchased services eligible for refund thereafter, and on a prorated basis. This prorated refund applies only to the full months remaining on the customer’s hosting contract, and these prorated refunds are usually subject to a cancellation fee of about $35. Add-ons and other services purchased are not subject to refund.
Our revenues are affected by refunds to some extent. Most of our revenues are generated by subscriptions lasting a year or longer, which requires us to recognize these revenues over the term of the subscription, even though customers pay in advance for all of our subscriptions. We have established a reserve for refunds, which represents the amount of recognized revenue that we expect to refund, and which amounted to $0.6 million at December 31, 2013 and December 31, 2014.
Approximately 20% of our refunds are for amounts that have already been recognized as revenue. The balance of our refunds take place only days after initial payment (e.g., shortly after an automatic renewal paid by credit card), and, accordingly, a significant portion of our refunds have no impact at all on recognized revenue. Our refund
Securities and Exchange Commission
December 9, 2015
Page 11
experience indicates that a significant majority of our refunds occur within the first 30 to 45 days of the contract start date. More specifically, 75% of all refunds processed occur within the same month that a customer contract begins, and approximately 90% occur within 45 days of the contract start date.
Generally, we do not reserve for refunds that take place in the same month as the customer payment because such refunds occur before any revenue is recognized.
We only reserve for refunds that we expect will reduce recognized revenue. In order to calculate our reserve for refunds, we evaluate our history of actual refunds and use this history to estimate the number of refunds that will occur in a given fiscal year. We then estimate the portion of refunds that we expect to occur after the month in which the customer payment is made and would therefore reduce recognized revenue.
We did not include a discussion of the impact of refunds on revenue for any of the periods presented in our 2014 Form 10-K because the impact of the change in the reserve for refunds was immaterial. We did provide a discussion of the reserve for refunds in our Form 10-Qs for the second and third quarters of 2015, in Note 2 to our unaudited consolidated financial statements. We intend to continue providing this disclosure in our future filings as appropriate in our consolidated financial statements and notes and, if and when material, in the discussion of critical accounting policies included in MD&A.
Note 19. Information about Geographic Areas, page 120
| 6. | We note your disclosure that it is impracticable for you to provide revenue information by geography due to the unavailability of certain geographic information. However, we note in your earnings call held on August 4, 2015 for the period ended June 30, 2015, you discuss the percentage of revenue from international geographies and provide comparative data for the same period in fiscal 2014. Please address the following: |
| • | Tell us your consideration to disclose this information in your financial statement footnotes. Refer to ASC 280-10-50-41; and |
| • | To the extent that the impact of foreign revenues materially impacts trends in your results of operations, please also tell us your consideration to disclose the levels of and mix in international revenues separately from your domestic revenues. Refer to Item 303(a)(3) of Regulation S-K. |
Securities and Exchange Commission
December 9, 2015
Page 12
Response:
Our financial reporting systems, which convert billing data to recognized revenue data, have not historically accumulated the geographic location of the customer. As such, it was historically impracticable for us to determine or disclose revenue information by geography.
During 2015, however, we have worked to enhance our financial reporting systems to begin to provide this geographic breakdown of recognized revenue for the 2013 fiscal year and all periods after 2013. Accordingly, on quarterly earnings teleconferences, we have recently begun to provide the total amount of revenue from international customers. We intend, subject to appropriate review of these data, to disclose geographic revenue data in our 2015 Form 10-K, in compliance with ASC 280-10-50-41.
Our initial data have indicated that foreign revenues have not materially impacted trends in our results of operations. To the extent such a material impact occurs in the future, we will include a discussion of foreign revenues in our discussion of our results of operations in MD&A in our future filings.
| 7. | In addition, please tell us what consideration you gave to disclosing revenues for each group of similar products or services or tell us why providing the information is impracticable. Please refer to ASC 280-10-50-40. |
Response:
We sell web hosting services and related products, such as domains and security services, to the same customer base, and these services and products are supported by the same cost infrastructure. Revenues from these similar services and products were approximately 96% of net revenues for 2014. We view these services as similar because they are delivered by a shared network infrastructure and customer support team, targeted to the same customer base through similar marketing programs and subject to similar regulatory environments. We believe these characteristics meet the definition of “similar” as noted in ASC 280. Accordingly, we have not provided additional breakdowns of revenues.
We do earn small portions of our revenues from various domain activities, including the sale of premium domain names and domain-related advertising revenue, and from incentives from third parties that sell products or services to our customers. These activities are generally targeted to different customers and marketed differently than our
Securities and Exchange Commission
December 9, 2015
Page 13
web hosting services and related products, and therefore not considered to be “similar.” The total amount of these ancillary revenues for 2014 was $28.3 million, which was just 4% of our consolidated revenues for that year. Given the immaterial amount of these ancillary revenues, we did not provide the revenue breakdowns under ASC 280-10-50-40.
Form 10-Q for Quarterly Period Ended September 31, 2015
Management’s Discussion and Analysis of Financial Condition and Results of Operations
Non-GAAP Financial Measures and Key Metrics
Average Revenue per Subscriber, page 32
| 8. | We note your disclosure that the revenue amount used to determine your measure of average revenue per subscriber includes revenue not attributable to subscribers. Please tell us what consideration was given to quantifying the amount of revenue included in this metric that is not attributable to subscribers so a reader may understand the impact of including such amounts on this metric. Alternatively, please tell us your consideration of providing a measure of average revenue per subscriber that only includes revenue amounts attributable to subscribers. |
Response:
We do not quantify the amount of revenue included in the calculation of average revenue per subscriber (“ARPS”) that is not attributable to subscribers (“Non-Subscriber Revenue”) because we are not able to quantify all of the components of Non-Subscriber Revenue. We also note that we have historically included all of adjusted revenue in our calculation of ARPS in an effort to present a model in which ARPS x number of subscribers = total adjusted revenue.
We have three principal sources of Non-Subscriber Revenue: revenue attributable to customers who purchase only a domain name from us and do not purchase any other products (“Domain-Only Customers”), domain monetization revenue, and marketing development funds. (Note that ARPS is based on adjusted revenue, as described in our disclosure, so the discussion below refers to that measure.)
Securities and Exchange Commission
December 9, 2015
Page 14
| • | Domain-Only Customers. Domain-Only Customers are expressly excluded from our Total Subscribers definition. We are unable to quantify the amount of our revenue attributable to Domain-Only Customers because our billing systems are not able to distinguish between revenue from Domain-Only Customers and revenue from customers who purchase domains from us in addition to other products. Although we are able to measure the total amount of our revenue from domains (approximately 20% of our total adjusted revenue), we cannot further break down domain revenue into revenue from Domain-Only Customers versus revenue from customers who purchase additional products from us and therefore meet our Total Subscribers definition. We are also unable to quantify the number of Domain-Only Customers, which precludes us from including them in our Total Subscribers definition. |
| • | Domain Monetization. Domain monetization revenue consists principally of revenue from our BuyDomains brand, which provides premium domain name products and services. Domain monetization revenue also includes, to a lesser extent, revenue from advertisements placed on unused domains (often referred to as “parked” pages) owned by us or our customers. Historically, the contribution of domain monetization activities to our adjusted revenue has been insignificant, but has been increasing beginning in 2014 due primarily to our acquisition of BuyDomains in September 2014. Our domain monetization adjusted revenue was $3.0 million in 2013, $19.1 million in 2014, $8.2 million for the nine months ended September 30, 2014 and $29.7 million for the nine months ended September 30, 2015, and its exclusion from our ARPS calculation would have resulted in ARPS being $0.07, $0.25, $0.43 and $0.77 lower for those periods, respectively. |
| • | Marketing Development Funds. Marketing development funds are amounts that certain of our partners pay to us to assist in and incentivize our marketing of their products. Our marketing development fund adjusted revenue was $7.5 million in 2013, $9.1 million in 2014, $6.7 million for the nine months ended September 30, 2014 and $8.8 million for the nine months ended September 30, 2015, and its exclusion from our ARPS calculation would have resulted in ARPS being $0.19, $0.20, $0.20 and $0.23 lower for those periods, respectively. |
Securities and Exchange Commission
December 9, 2015
Page 15
In our Forms 10-Q for the first and second quarters of 2015, we noted an increase in premium domains sold through our BuyDomains and Directi businesses as a factor contributing to year-over-year growth in ARPS. In our future reports on Form 10-Q and 10-K, we will continue to enhance our disclosure about Non-Subscriber Revenue contribution to ARPS, including a more detailed qualitative description of the sources of Non-Subscriber Revenue included in the ARPS calculation and an express statement that ARPS does not represent an exact measure of the amount an average subscriber actually spends with us each month. We will also highlight instances where we believe Non-Subscriber Revenue has materially contributed to period-over-period changes in ARPS.
| 9. | Please clarify your disclosure on page 32 that “total subscribers for a period reflects adjustments to add or subtract subscribers as we integrate acquisitions and/or are otherwise able to identify subscribers that meet the definition of total subscribers.” In this regard, clarify whether the 9% total increase in subscribers attributable to “these adjustments” was due to acquisitions or whether you identified subscribers other than those acquired through business combinations during the period that were previously excluded from the metric. If the latter is true, please also tell us how you considered the impact of those additional subscribers on previously reported average revenue per subscriber metrics. |
Response:
Our Total Subscribers figure for a period reflects adjustments to add or subtract subscribers as we integrate acquisitions and/or are otherwise able to identify subscribers that meet the definition of Total Subscribers. These adjustments amounted to 9% of the increase in Total Subscribers during the third quarter of 2015 and consisted of subscribers who we identified in the course of routine data analytics activities as meeting the definition of Total Subscribers during the quarter, rather than subscribers who were identified as part of acquisition integration activities.
Specifically, during the third quarter of 2015, the number of Total Subscribers on our platform increased by 87,233. Of this increase, 7,682 subscribers, or approximately 9% of the increase, were existing subscribers that were added to our Total Subscribers figure in the third quarter because they were identified during the quarter as meeting our Total Subscribers definition.
Securities and Exchange Commission
December 9, 2015
Page 16
Of these 7,682 subscribers, 6,455 were existing subscribers of our Netfirms brand and 1,227 were existing subscribers of our HostGator Brazil brand. The explanations for the adjustments are as follows:
| • | The Netfirms subscribers had been incorrectly identified by our billing systems as Domain-Only Customers and, as a result, they had been excluded from our Total Subscribers figure. In the second quarter of 2015, we determined as part of our ongoing review of our subscriber data that approximately 40,000 of our Netfirms customers were in fact purchasing a hosting package from us in addition to a domain and therefore should have been counted in Total Subscribers. Since that determination, we have added these customers to our Total Subscribers figure if and as they renew and in the period of such renewal. Accordingly, the 6,455 Netfirms subscribers added in the third quarter of 2015 represented the number of renewals from such subscribers during that quarter. |
| • | The HostGator Brazil subscribers had been previously omitted due to an error in the calculation of subscriber cancellations for HostGator Brazil in the second quarter of 2015. We identified this error during the third quarter of 2015 and adjusted the Total Subscriber balance accordingly. |
The number of subscribers added by the adjustments in the third quarter of 2015 was very small (approximately 0.17%) relative to our Total Subscriber base. Including these subscribers in the denominator for the calculation of ARPS had the effect of reducing ARPS for the third quarter by $0.02 from $14.31 to $14.29. We determined this difference to be immaterial.
| 10. | Given that acquisitions are a material part of your growth strategy, please tell us what consideration was given to separately quantifying the number of subscribers added due to acquisitions during the reported periods. In this regard, you cite numerous factors that drove the increase in subscribers during the period; however it is unclear how much growth is attributable to acquisitions and/or ongoing operations. |
Response:
We quantify the number of subscribers added due to acquisitions in our periodic reports for quarters during which we have completed acquisitions that involve the acquisition of subscribers. For instance, we separately quantified the number of subscribers contributed by our Verio and Site5 acquisitions, which occurred during the second quarter of 2015, in
Securities and Exchange Commission
December 9, 2015
Page 17
our Quarterly Report on Form 10-Q for the quarterly period ended June 30, 2015. We did not include a similar subscriber breakdown in our Quarterly Report on Form 10-Q for the quarterly period ended September 30, 2015 because we did not acquire any subscribers through acquisitions during that quarter. Nevertheless, we acknowledge the Staff’s comment, and in future filings, we will separately quantify the contribution of acquired subscribers to subscriber growth for the applicable reporting period.
* * * *
Securities and Exchange Commission
December 9, 2015
Page 18
The Company hereby acknowledges the following:
| • | the Company is responsible for the adequacy and accuracy of the disclosure in the filing; |
| • | Staff comments or changes to disclosure in response to Staff comments do not foreclose the Commission from taking any action with respect to the filing; and |
| • | the Company may not assert Staff comments as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States. |
If you have any further questions or comments, or if you require any additional information, please contact the undersigned by telephone at 781-852-3464 or electronically at marc@endurance.com. Thank you for your assistance.
Very truly yours,
/s/ Marc Montagner |
| Marc Montagner |
| Chief Financial Officer |
| cc: | David Bryson,Endurance International Group Holdings, Inc. |
| Lara Mataac,Endurance International Group Holdings, Inc. |
