013 Business Case Femtocells Mobile BB Era

8/16/2019 013 Business Case Femtocells Mobile BB Era

1/88

www.scf.io/ www.smallcellforum.org

DOCUMENT

The business case for femtocells

in the mobile broadband era

Signals research

December 2013

013.06.02

SMALL CELL FORUM

RELEASE 6.0

Solving the HetNet puzzle

17 : 2 5

R U R A L & R E M O T E

U R B AN

E N T E R P R I S E

V I R T U AL I Z AT I O N

H O M E

scf.io

A whitepaper prepared by Signals Researchthat highlights the economics of femtocellsunder a wide range of scenarios andassumptions on a global basis

The original format of this publication has beenupdated in line with the style of Small CellForum’s Release programme


2/88

SMALL CELL FORUM

RELEASE 6.0 scf.io

If you would like more information about Small Cell Forum or wouldlike to be included on our mailing list, please contact:

Email [email protected]

Post Small Cell Forum, PO Box 23, GL11 5WA UK

Member Services [email protected]

Small Cell Forum accelerates small cell adoption to drive the wide-scale adoption of small cells and accelerate the delivery of integratedHetNets.

We are not a standards organization but partner with organizations that informand determine standards development. We are a carrier-led organization. Thismeans our operator members establish requirements that drive the activitiesand outputs of our technical groups.

We have driven the standardization of key elements of small cell technologyincluding Iuh, FAPI/SCAPI, SON, the small cell services API, TR-069 evolutionand the enhancement of the X2 interface.

Today our members are driving solutions that include small cell/Wi-Fiintegration, SON evolution, virtualization of the small cell layer, driving massadoption via multi-operator neutral host, ensuring a common approach toservice APIs to drive commercialisation and the integration of small cells into5G standards evolution.

The Small Cell Forum Release Program has now established business casesand market drivers for all the main use cases, clarifying market needs andaddressing barriers to deployment for residential, enterprise and urban small

cells. The theme of Release 6 is Enterprise, with particular emphasis on realworld and vertical market deployments, and the role of neutral host solutionsto drive the mass adoption of small cells in business environments.

Small Cell Forum Release website can be found here: www.scf.io


3/88

Report title: The business case for femtocells in the mobile broadband era Issue date: 01 December 2013

Version: 013.06.02

Scope

This document considers the impact of femtocells in an era of data-intensive

smartphones, broadband enabled laptops, and a growing number of highly capablemobile internet devices.

NB. At the time of commission Small Cell Forum was known as Femto Forum. Allreferences to Femto Forum are interchangeable with Small Cell Forum.


4/88


Version: 013.06.02

Executive summary

In 2009 Signals Research Group, LLC (SRG) authored the Femto Forum Femtocell

Business Case White paper (Small Cell Forum publication 005, June 2009) whichdescribes in detail a range of consumer propositions and operator business cases foroperators in North America, Europe, and Asia. The scenarios – while hypothetical –were intended to reflect real operator offerings. They highlighted a number ofimportant benefits: voice coverage, home-zone pricing, churn reduction, macrocellular traffic offload, emerging services, and others.

The results were broken into two categories, a “basic value proposition,” which wasrelatively concrete, and an “enhanced value proposition,” which included items whichwere more speculative and/or applied to only certain categories of operators.

In the past year wireless data usage has continued to escalate – often makingheadlines in the general press – and manufacturers have made great strides in

developing the chipsets and product architectures behind LTE and WiMAX femtocells.Finally, a growing number of large operators have launched femtocell offerings, andthe expected volume of femtocell shipments in 2010 is likely to reach critical mass(jumping from a few hundred thousand units worldwide to millions of units).

It is time, therefore, to consider the impact of femtocells in the “Mobile BroadbandEra.” In this whitepaper we jump forward in time – ostensibly to 2012 -2013 – todiscuss the likely impact of femtocells in the rapidly approaching era of data-intensivesmartphones, broadband enabled laptops, and a growing number of highly capablemobile internet devices (MIDs).

Based on the scenarios that we present in this whitepaper, we can reach several broad

conclusions, which we substantiate later in this report, including:

• The business case, while compelling in 2009, improves further as data usecontinues to increase and as operators move from purely coverage-drivenfemtocell deployments to ones based on macrocell-off load, performance

enhancement, and personalized home-zone and office-zone services. Home-zone and office-zone services, which are included only in a few of theanalyses in this report – principally in connection with the enterprise willprovide significant additional upside.

•

In all of the 3G and next generation cases studied using a customer lifetimevalue approach, the macrocell-off load network savings exceed the cost ofthe femtocell. As part of this study we identified four basic customersegments, ranging from the Socially Networked Family to theCommunications Enabled Firm. While the exact definition of each segmentvaries, they share several attributes, most notably they have widely adoptedsmartphones and other data-centric devices (e.g., embedded notebooks, USB

dongles and MIDs), and they have an appetite for generating lots of mobiledata traffic. To be specific, we assume that by 2012-2013 the averagemonthly mobile data usage of laptops is 5 GB while the average monthly datausage of smartphones is 500 MB. While 500 MBs per month of smartphone

usage represents an order of magnitude jump from our business case lastyear, some trend-watchers consider it rather conservative.

Further, we assume that the wholesale cost of the femtocell reaches moreattractive price points than those we assumed in our first whitepaper as the

market develops and as economies of scale are reached. The combination of

http://femtoforum.org/femto/index.php?id=69http://femtoforum.org/femto/index.php?id=69http://femtoforum.org/femto/index.php?id=69http://femtoforum.org/femto/index.php?id=69http://femtoforum.org/femto/index.php?id=69


5/88


Version: 013.06.02

these two factors – higher data usage, which can be offloaded via femtocells,and lower-cost femtocells – is what drives this conclusion.

• An operator offering either 3G or dual-mode LTE/HSPA femtocells is able torealize a return of 10 times on their LTE/HSPA femtocell service investment

across a wide range of mobile broadband customers. In the first whitepaper,the focus was exclusively on the customer lifetime value for a given customersegment. In reality, an operator is going to ultimately target numerouscustomer segments, perhaps going after each segment with a slightlydifferent strategy and service offering.

Since publishing the first whitepaper, we have expanded the capabilities ofthe femtocell business modelling tool so that it is possible to look at themulti-year effect of an operator targeting several customer segmentssimultaneously. As demonstrated in Chapter 6, a 10x return is entirely

realistic, and largely based on the premise that the initial femtocellinvestment is relatively modest. Said another way, unlike other network

investment decisions, an operator’s decision to deploy femtocells does notcome with the requisite huge upfront investment.

Instead, the upfront investment is relatively modest with incremental

investments quickly being offset by the subsequent revenues for eachfemtocell customer that is added to the network. As a result, the impressivereturns that we demonstrate are entirely reasonable while the risk to theoperator is relatively modest.

In the specific example that we use in this whitepaper, we look at arepresentative European operator with 10 million subscribers who deploysfemtocells to 10% of its customer base. Based on the assumptions that welay out in Chapter 6, the operator is able to pay back its initial femtocell-

related investment by Year 4 and ultimately realize a 10x return oninvestment.

•

A business case exists even for a regional operator deploying only a fewthousand femtocells. Many smaller regional operators, in addition to the largemulti-national operators, are interested in deploying femtocells. However, forthe smaller regional operators, the upfront investment can be more criticalsince the number of femtocell subscribers that they can sign up will obviouslybe fewer than what it would be for a large operator.

As we demonstrate in Chapter 7, these smaller regional operators can alsoimplement a successful and profitable femtocell business strategy. The key to

their success is to delay or even minimize some of the upfront integrationactivities that a larger operator would normally pursue before selling its firstfemtocell. Instead, the smaller regional operator would complete only themost essential integration activities and would instead bear additionaloperational expenses with each incremental femtocell subscriber. The netresult – counterintuitive to large operators – is that the small operatorminimizes the potential disadvantages of small scale.

• A new entrant or incumbent deploying LTE or Mobile WiMA X can dramatically

improve the customer experience by providing each new subscriber with afemtocell. One of the challenges associated with deploying LTE or MobileWiMA X is that it requires the deployment of a brand new network, and

typically at a higher frequency band, which necessitates additional cell sites


6/88


7/88


Version: 013.06.02

Appendix I points to a number of important external references. Appendix II describesthe customer lifetime value (CLV) methodology and re-familiarizes the reader withhow to interpret the waterfall charts that we use throughout this whitepaper. Appendix

III provides critical background information on our approach to macro networkeconomic modelling.

Summary of Results

Table 1 on page 12 provides a useful reference, summarizing key results, which we

will discuss throughout the rest of the whitepaper. In the coming pages we willconsider four different segments of users. We will look at the economics of HSPAfemtocells and LTE/HSPA femtocells and we will look at the economics of operators ofdifferent sizes. The results in Table 1 quantify each of these scenarios. To make the

scenarios in Table 1 as comparable as possible we have given them similar adoptioncurves. In each case we assume that 10% of the total subscribers of an operator haveaccess to a femtocell along the lines of that service concept. In reality, operators are

likely to serve multiple segments simultaneously, as described in Chapter 6. Our smalloperator study is based on the Technology Savvy Family (to be defined in a fewpages), which is why the numbers in the small operator study look similar to those ofthe Technology Savvy Family. In most cases we focus on the “Basic Value Proposition”

because it includes the most measurable effects. In the case of the CommunicationsEnabled Firm (also defined in a few pages), we show the Enhanced Value Proposition,because many of the bars it contains are – for that segment – significant in theirimpact and extremely likely, given the customer proposition.

Finally we include a “return on investment” calculation. In a textbook ROI calculationyou might have a large upfront investment (the “I”) at time zero followed by a stringof positive cash flows (the “R”). The real world is often much more complex. In afemtocell business case the operator makes a significant upfront investment,principally in provisioning and billing system integration and in femtocell-specificplatforms. The operator then continues to invest with each additional femtocell. Thatincremental investment – which we call a subsidy is the cost of the femtocell plus along list of associated expenditures (allocated platform costs, marketing, sales

incentives, the cost of future upgrades, etc), less the amount the customer pays forthe femtocell. One might calculate “investment” as the maximum negative cash flowof the business, relative to a business without femtocells. Such an approach would beconsistent with the practices of many financial professionals. The results, however,would be highly sensitive to the assumed rate of adoption of femtocells.

We have taken a different and slightly more conservative approach which we hope willbe more intuitive to many readers. We have included in “investment” all of the

expenditures (upfront and incremental) needed to bring femtocells to market and tosupport the proposition long-term. The incremental expenditures are essentially the

“net subsidy” required to support femtocells as a product line. Over time, some of theearly femtocell users will churn off. The infrastructure that was purchased to supportthem will become idle and available for other femtocell users. In a similar fashion, the

initial investment in platforms will result in pre-purchased capacity, which means thatincremental expenditures for platform hardware will be minimal in the early years. Thesystem integration costs which the operator bears initially to launch the business willnot be repeated in later years. At the same time, periodic femtocell upgrades willcontinue. Since we assume that femtocells reduce churn the acquisition costs of theoverall business diminishes slightly, because the number of gross adds needed tomaintain the operator’s market share (which we conservatively assume does not

increase) is reduced.


8/88


Version: 013.06.02

Through a fairly elaborate process with a large number of assumptions we look at allthe cash flows of the business (revenue and expense impacts) and compare theprofitability of the business with femtocells to the profitability of the business without

femtocells. The value of femtocells is then the incremental profit contribution, thedifference between these two hypothetical businesses. Of course, there are many

subscribers who never purchase or have access to a femtocell, so their profitabilityremains the same. Finally, we calculate the ratio of return (the incremental profit

contribution) to investment (in our construct, the sum of the up-front and incrementalinvestments over time). This produces a ratio which is not discounted for the timevalue of money. It is like someone who purchases a house then sells it 30 years later.They will have a huge return on investment (in nominal dollars or Euros), even thoughthe value of the asset, after adjusting for inflation or – even worse – after applying amarket cost of capital, might have decreased. In a similar fashion, we calculate anumber of other financial KPIs. We calculate the present value of femtocell-related

investments, the present value of the incremental profit contribution, and the presentvalue of the effective subsidies based on a 10% WACC. We also calculate an internalrate of return for the femtocell investment.

We have discovered that the return on investment (an undiscounted figure) and theIRR (reflecting the time value of money) are big numbers. In most scenarios there is a

10 times return on investment. One might ask, “How is this possible?” Unlike anindustry sector which can independently attract capital and produce investor returns,femtocells represent a puddle of profitability adjoining a much larger lake, which is the

composite financials of the operator. An operator can sell only so many femtocells to acertain profitable segment before that opportunity is exhausted. While the return oninvestment is extremely high, the opportunity for investment is finite.

If we look at other investment opportunities within an operator environment we see asimilar pattern. If we consider the revenue an operator receives for text messaging,

subtract direct expenses (termination costs, sales and marketing expense, billingexpense, etc), then divide the profit contribution by the capital investment an operatormakes in SMSCs (the platform infrastructure needed to deliver short messaging) thereturn on capital is stunning. These returns are very real. If investors could invest onlyin SMSCs and magically capture all the associated text messaging revenues, theywould. Unfortunately, they also need to invest in the “other” parts of a mobileoperator. That operator needs to purchase spectrum, build a costly cellular network,

subsidize handsets, etc. Text messaging is therefore a “pool” of profitability within alarger business entity. Voice messaging is another pool of profitability. It is hard toimagine a modern mobile operator debating whether or not to offer text messaging(although people had those debates 20 years ago) or debating whether or not to offer

voice mail. Each is considered an integral and inseparable part of the mobile franchiseand each – on a standalone basis – continues to be immensely profitable. Fifteenyears from now femtocells might be viewed in a similar light, as an integral part of theoperator franchise, a profitable platform investment, the wisdom of which few woulddebate.

Does this mean that femtocells will significantly impact operator financials? Itdepends. Our analyses in this whitepaper (excluding Chapter 8) assume thatfemtocells are used by 10% of subscribers at the end of 5 years. The present value of

profit contributions from femtocells – given such an assumption falls in the range of1% to 3% of the present value of an operator’s total revenue. They represent a verynice uplift – as did SMS or voicemail in the early days – but, based on the numbers wehave been running, will not singlehandedly transform an operator’s P&L. If femtocellswere to penetrate faster – which seems entirely possible – or if they were to reach

much higher levels of penetration (e.g. 30%-40%), as some operators now envision,


9/88


Version: 013.06.02

the impact could be profound. Of course, as penetration increases and as femtocellsbecome mainstream, the incremental return on investment is likely to diminish, sinceoperators will have to become less selective. Even with very conservative penetration

assumptions, the required level of investment is so small and the expected return oninvestment so high, that enabling femtocells as a platform is a decision most operators

should seriously consider.

NB. At the time of commission Small Cell Forum was known as Femto Forum. Allreferences to Femto Forum are interchangeable with Small Cell Forum.


10/88

Report title: The business case for femtocells in the mobile broadband era Issue date: 01 December 2013Version: 013.06.02

Study Market Segment Technology STARTINGValue

FemtocellCOSTS

NetworkSAVINGS

CoverageVoice

USAGE

FREECalling

Plan

MonthlyFEE

ReducedCHURN

HSPAScenarios

The Technology SavvyFamily

HSPA €913.83 €(130.95) €950.22 €370.84 €- €275.53 €263.07

The Socially NetworkedFamily

HSPA €2,387.09 €(130.95) €373.71 €741.68 €- €275.53 €387.05

The Mobile Executive HSPA €242.76 €(130.95) €729.56 €185.42 € €275.53 €145.38

The CommunicationsEnabled Firm

HSPA €28,205.51 €577.12 €3,103.66 €- €4,700.03 €- €3,221.60

LTEScenarios


LTE/HSPA €1,159.28 €(230.03) €839.76 €370.84 €- €275.53 €273.57


LTE/HSPA €2,457.21 €(230.03) €342.16 €741.68 €- €275.53 €390.04

The Mobile Executive LTE/HSPA €435.62 €(230.03) €642.78 €185.42 €- €275.53 €153.63


LTE/HSPA €29,502.13 €(55.82) €2,779.50 €- €4,700.03 €- €3,294.98

SmallOperatorScenarios

Very Large Oper Full+ HSPA €913.83€(127.

32)€950.22 €370.84 € €275.53 €263.07

Large Oper Full HSPA €913.83 €(135.77) €950.22 €370.84 € €275.53 €263.07

Small Oper 3G Full HSPA €913.83 €(266.14) €950.22 €370.84 €- €275.53 €263.07

Small Oper 3G Limited HSPA €913.83 €(161.32) €950.22 €370.84 €- €275.53 €263.07

Very Small 3G Limited HSPA €913.83 €(267.99) €950.22 €370.84 € €275.53 €263.07


11/88


Study Market Segment TechnologyBASICValue

AddlFamily /

Employee

EnhancedServices

ExistingServices

(TV)

OtherBusiness

Churn

ENHANCEDValue

HSPA Scenarios


HSPA €2,642.54 €133.30 €296.95 €296.95 €141.72 €3,511.46


HSPA €4,034.10 €133.30 €296.95 €296.95 €141.72 €4,903.02

The Mobile Executive HSPA €1,447.70 €133.30 €296.95 €296.95 €141.72 €2,316.62


HSPA €39,807.92 €9,826.93 €3,756.77 €2,504.51 €- €55,896.14

LTE Scenarios


LTE/HSPA €2,688.96 €137.08 €296.95 €296.95 €141.72 €3,561.66


LTE/HSPA €3,976.59 €137.08 €296.95 €296.95 €141.72 €4,849.30

The Mobile Executive LTE/HSPA €1,462.94 €137.08 €296.95 €296.95 €141.72 €2,335.64


LTE/HSPA €40,220.83 €9,946.45 €3,756.77 €2,504.51 €- €56,428.57

Small Operator Scenarios

Very Large Oper Full+ HSPA €2,646.17 €133.30 €296.95 €296.95 €141.72 €3,515.09

Large Oper Full HSPA €2,637.72 €133.30 €296.95 €296.95 €141.72 €3,506.64

Small Oper 3G Full HSPA €2,507.35 €133.30 €296.95 €296.95 €141.72 €3,376.27

Small Oper 3G Limited HSPA €2,612.17 €133.30 €296.95 €296.95 €141.72 €3,481.09

Very Small 3G Limited HSPA €2,505.50 €133.30 €296.95 €296.95 €141.72 €3,374.42


12/88


Study Market Segment TechnologyB/S1 E/S2

Return onInvestment

Basis for ROICalculations

HSPA Scenarios The Technology Savvy Family HSPA 289% 384% 16.92x Basic Value

The Socially Networked Family HSPA 169% 205% 15.22x Basic Value

The Mobile Executive HSPA 596% 954% 12.10x Basic Value

The Communications Enabled Firm HSPA 141% 198% (72.89x) Enhanced Value

LTE Scenarios The Technology Savvy Family LTE/HSPA 232% 307% 9.04x Basic Value

The Socially Networked Family LTE/HSPA 162% 197% 8.49x Basic Value

The Mobile Executive LTE/HSPA 336% 536% 6.27x Basic Value

The Communications Enabled Firm LTE/HSPA 136% 191% 174.43x Enhanced ValueSmall Operator Scenarios Very Large Oper Full+ HSPA 290% 385% 16.76x Basic Value

Large Oper Full HSPA 289% 384% 16.14x Basic Value

Small Oper 3G Full HSPA 274% 369% 10.17x Basic Value

Small Oper 3G Limited HSPA 286% 381% 13.92x Basic Value

Very Small 3G Limited HSPA 274% 369% 8.67x Basic Value

Table A Summary of Results

1 B/S = Basic Value Proposition / Start Source: Signals Research Group, LLC2 E/S = Enhanced Value Proposition / Start


13/88


Version: 013.06.02

Contents

1. Methodology .....................................................................5

1.1 Customer Lifetime Value ...................................................... 5

1.2 Reading a Waterfall Chart ..................................................... 5

2. The Mobile Broadband Era ................................................7

2.1 Exponentially Increasing Wireless Data Traffic ......................... 7

2.2 Data Offload Is Real and Measurable ...................................... 7

3. Introducing Four Segments, Circa 2012-2013 ..................9

3.1 The Technology Savvy Family ............................................... 9

3.2 The Socially Networked Family .............................................. 9

3.3 The Mobile Executive ......................................................... 10 3.4 The Communications Enabled Firm ...................................... 11

3.5 Details: Devices, ARPU, Usage ............................................ 12

3.6 The Impact of LTE ............................................................. 16

4. An Incumbent European Operator with HSPA ................. 17

4.1 Starting Point: Last Year’s Business Case ............................. 17

4.2 A Journey into the Future ................................................... 18

5. Operator View: Multiple Segments ................................. 25

5.1 Rationale / Economies of Scale across Segments ................... 25

5.2 Key Assumptions ............................................................... 25

5.3 Summary Results .............................................................. 28

6. Operator View: Large and Small Operators / Economiesof Scale .......................................................................... 32

6.1 Important Up-Front Costs ................................................... 32

6.2 Small Operator: An Entry Strategy ...................................... 34

6.3 Large versus Small Operators: Cost Comparison ................... 34

6.4 Large versus Small Operators: CLV Comparison .................... 36

7. LTE/WiMAX Network Deployment Economics Asia ......... 41

7.1 The Concept ..................................................................... 41

7.2 Key Assumptions ............................................................... 41

7.3 Market Demand for LTE ...................................................... 48

7.4 The Femtocell Market in Japan ............................................ 49

7.5 LTE Economics .................................................................. 51

7.6 Mobile WiMAX Economics.................................................... 51

7.7 Optimizing the Mobile WiMAX Business Case ......................... 56

7.8 Incumbent Operator Deployment......................................... 56


14/88


Version: 013.06.02

7.9 Conclusions – LTE/Mobile WiMAX Network DeploymentEconomics in Japan............................................................ 60

8. Summary of Report ........................................................ 62

Appendix I: Additional Resources ............................................. 64 Appendix II: Customer Lifetime Value Methodology /

Waterfall Chart ............................................................... 65

9. Customer Lifetime Value Methodology. .......................... 65

9.1 Customer Life Value Methodology / Bars in the WaterfallChart ............................................................................... 65

9.2 Start Value ....................................................................... 66

9.3 Add Femtocell Costs .......................................................... 66

9.4 Network Cost Savings ........................................................ 66

9.5 Additional Network Costs Due to Increased Data Usage .......... 66 9.6 Add Coverage-Related Increase in Voice Usage ..................... 66

9.7 Add Free Calling Plan ......................................................... 66

9.8 Add Basic Monthly Fee ....................................................... 67

9.9 Add Reduction in Churn ...................................................... 67

9.10 Add Basic Value Proposition ................................................ 67

9.11 Add Value of Additional Family Member ................................ 67

9.12 Add Enhanced Services Revenue ......................................... 68

9.13 Incremental Revenue from Existing Services ......................... 68

9.14 Reduce Other Business Churn ............................................. 68

9.15 Enhanced Value Proposition ................................................ 68

10. Macro Network Economics Model ................................... 69

10.1 RAN Modelling Coverage Methodology – Summary ................. 69

10.2 Network Cost Modelling Methodology ................................... 71

10.3 Femtocell Core Network ..................................................... 75

10.4 Network Cost Savings ........................................................ 75


15/88


Version: 013.06.02

Tables

Table 3-1 Number of Devices and Femtocell Pricing .......................................... 14

Table 3-2 ARPUs, MOU, and MBs .................................................................... 15

Table 5-1 Distribution of Femtocells and Devices in a Multi-Segment Market ....... 25

Table 6-1 Key Expenditures by Operator Size / Integration Strategy ................... 34

Table 7-1 Number of Sites by Deployment Scenario ......................................... 58

Table 7-2 Network Cost Savings by Deployment Scenario ................................. 60

Table 10-1 Primary RAN Coverage Assumptions ................................................. 71

Table 10-2 Link Budget Propagation Characteristics ........................................... 71

Table 10-3 Uplink Link Budget Margins ............................................................. 71

Table 10-4 Capacity-Constrained Cell Site Capital and Expense ........................... 73

Figures

Figure 2-1 Expected Growth in Global Wireless Data Usage ................................. 7

Figure 2-2 Expected Macro Cellular Data Offload by Country ................................ 8

Figure 3-1 The Technology Savvy Family ........................................................... 9

Figure 3-2 The Socially Networked Family ........................................................ 10

Figure 3-3 The Mobile Executive ...................................................................... 11

Figure 3-4 The Communications Enabled Firm ................................................... 12

Figure 4-1 European Coverage-Challenged Household (2009 Business CaseWhitepaper) .................................................................................. 17

Figure 4-2 Technology Savvy Family, HSPA Femtocell ........................................ 20

Figure 4-3 Technology Savvy Family, LTE/HSPA Femtocell.................................. 21

Figure 4-4 The Socially Network Family, HSPA Femtocell .................................... 21

Figure 4-5 The Socially Networked Family, LTE/HSPA Femtocell .......................... 22

Figure 4-6 The Mobile Executive, HSPA Femtocell .............................................. 23

Figure 4-7 The Mobile Executive, LTE/HSPA Femtocell........................................ 24

Figure 4-8 The Communications Enabled Firm, LTE/HSPA Femtocell .................... 24

Figure 5-1 Number of Devices with Access to an LTE/HSPA Femtocell .................. 26

Figure 5-2 Number of LTE/HSPA Femtocells by Year .......................................... 27

Figure 5-3 Number of Devices with Access to an LTE/HSPA Femtocell .................. 28

Figure 5-4 Present Value of Femtocell-Related Expenditures, Retail Sales,Subsidies, and Profit Contributions ................................................... 29

Figure 5-5 Femtocell-Related Expenditures by Category by Year ......................... 30

Figure 5-6 Profit Contribution by Year .............................................................. 31

Figure 5-7 Cumulative Profit Contributions / Breakeven Analysis ......................... 31

Figure 6-1 Femtocell-Related Expenditures by Operator Size and IntegrationStrategy ....................................................................................... 35

Figure 6-2 Femtocell-Related Expenditures per Femtocell by Operator Size andIntegration Strategy ....................................................................... 36

Figure 6-3 Very Large Operator (Scenario I) Customer Lifetime Value .................. 37

Figure 6-4 Large Operator (Scenario II) Customer Lifetime Value ........................ 37

Figure 6-5 Small Operator, Full Integration (Scenario III), Customer LifetimeValue ........................................................................................... 38


16/88


Version: 013.06.02

Figure 6-6 Small Operator, Minimal Integration (Scenario IV), Customer LifetimeValue ........................................................................................... 39

Figure 6-7 Very Small Operator, Minimal Integration (Scenario V), CustomerLifetime Value ............................................................................... 40

Figure 7-1 Primary Coverage Area – Probability vs. Expected Data Rates ............. 42 Figure 7-2 Primary Quality of Coverage – Required Number of Sites .................... 43

Figure 7-3 Secondary Coverage Area – Probability versus Expected Data Rates .... 44

Figure 7-4 Secondary Quality of Coverage – Required Number of Sites ................ 45

Figure 7-5 Alternative Quality of Coverage – Required Number of Sites ................ 46

Figure 7-6 Total Number of Macro-Cellular + In-Building Sites with AlternativeQuality of Coverage versus Primary Quality of Coverage ..................... 46

Figure 7-7 Network Cost Savings due to Slightly Slowed Deployment .................. 47

Figure 7-8 LTE Subscribers / Market Penetration by Year in Japan ....................... 48

Figure 7-9 Assumption: LTE Operator Market Shares ......................................... 49

Figure 7-10 LTE Subscribers by Operator ........................................................... 49 Figure 7-11 Fully Allocated Cost of LTE/HSPA Femtocells in Japan ......................... 50

Figure 7-12 Fully Allocated Femtocell Costs vs. Network Cost Savings ................... 51

Figure 7-13 Primary Quality of Coverage – Required Number of Sites (WiMAX) ....... 52

Figure 7-14 Secondary Quality of Coverage – Required Number of Sites (WiMAX) ... 53

Figure 7-15 Alternative Quality of Coverage – Required Number of Sites (WiMAX) .. 53

Figure 7-16 Total Number of Macro Cellular + In-Building Sites with AlternativeQuality of Coverage vs. Primary Quality of Coverage (WiMAX) ............. 54

Figure 7-17 Network Cost Savings due to Slightly Slowed Deployment (MobileWiMAX) ........................................................................................ 54

Figure 7-18

Fully Allocated Cost of WiMAX/1X/EV-DO Femtocells in Japan .............. 55 Figure 7-19 Fully Allocated Femtocell Costs vs. Network Savings (WiMAX) ............. 55

Figure 10-1 Marginal Cost per GB by Technology ................................................ 75


17/88


Version: 013.06.02 5

1. Methodology

We refer readers to the appendix for a more detailed discussion of our methodology,

including additional insight into how we calculate macro network economics. Last

year’s business case white paper (pub 005) also describes the importantmethodologies in detail.

1.1 Customer Lifetime Value

We use throughout this whitepaper a methodology called “customer lifetime value,”which is widely used in many service-oriented industries. It views the world throughthe eyes of an individual customer or – more commonly – a segment of customers. It

calculates the present value of revenues and expenses over the expected life of thecustomer. Mobile subscribers can be segmented along many different dimensions,including by prepaid/postpaid, by distribution channel, by usage patterns/pricingplans, by psychographic characteristics, and by sophisticated multi-dimensional

clustering algorithms. In this whitepaper we construct and discuss a number ofhypothetical profiles – individuals, families or corporations with certain wireless usagepatterns. The suggestion is that mobile operators should analyse their subscriber

base, designing propositions to target those segments for which femtocells cansignificantly increase customer lifetime value.

1.2 Reading a Waterfall Chart

In this whitepaper customer lifetime value results are typically presented in the formof a waterfall chart. The first bar shows the value of the customer segment beforefemtocells are introduced. Each subsequent bar shows the impact of a specific change.One bar represents the cost of the femtocell (including various overheads and the cost

of future upgrades). Another bar shows the impact of macro cellular network offload.A waterfall chart is cumulative. Each successive bar is floating in space. The top of thebar (for one which is positive) represents the new sum. Bars which are positive arecolored green and bars which are negative are colored red.

Each waterfall chart begins with a “starting” NPV, representing the customer before

femtocells. Various bars appear, leading up to a “basic value proposition,” whichrepresents a conservative view of the value created by the femtocell. The “basic valueproposition” excludes effects which are somewhat speculative or futuristic or whichmight apply only to a subset of operators.

Appendix II describes each bar in detail. Appendix III describes in some detail thesophisticated mobile operator network cost modelling which undergirds many of these

results. In additional to analyzing femtocell-specific elements our models characterizethe economics of mobile operator networks by technology / geography / spectrum

band, etc. We have developed a detailed cost model with a worldwide database ofgeo-coded information, built from satellite imagery, national demographic andeconomic data, and other reliable global sources. In studies which consider coverageas many do – we model operator build plans using a portion of these GIS databasesalong with industry-standard COST-231 coverage models.

The final study we present focuses on Japan. We broke the country into fivemorphologies (dense urban, urban, suburban, rural clutter, and rural open) andmodelled the number of sites needed for each, based on a statistical indoor coveragecriterion. In discussing the study results we compared our analytically-derived results

with the current site count of the dominant mobile operator in Japan and the reported

http://femtoforum.org/femto/index.php?id=69http://femtoforum.org/femto/index.php?id=69http://femtoforum.org/femto/index.php?id=69


18/88


Version: 013.06.02 6

future LTE site counts of other operators. The methodology behind these analyses isdescribed in greater detail in Appendix III and in much greater detail in SRG’s 1000page report on next generation infrastructure, The Dollars and Sense of BroadbandWireless (http://www.signalsreasearch.com).

http://%28http/www.signalsreasearch.com).http://%28http/www.signalsreasearch.com).http://%28http/www.signalsreasearch.com).http://%28http/www.signalsreasearch.com).


19/88


Version: 013.06.02 7

2. The Mobile Broadband Era

A central theme of this whitepaper is “mobile broadband,” and how wireless data

traffic is increasing exponentially in operator networks. This trend is likely to impact

the business case for femtocells in the coming years. In this whitepaper we use theterm “mobile broadband” liberally, to connote not only laptops / mobile internetdevices but also smartphones – since smartphone data usage is also rapidlyincreasing.

The reader will discover that data offload – an ancillary piece of the analysis in lastyear’s femtocell business case whitepaper – is poised to become a central piece in anoperator’s decisionmaking process in the years ahead.

2.1 Exponentially Increasing Wireless Data Traffic

Wireless data traffic has been rapidly increasing for several years, but it has recentlyreached a tipping point, in which managing the exponential growth in data traffic isincreasingly becoming a critical issue for operators.

Cisco, in collaboration with a handful of other partners, has mapped the growth ofmobile broadband data. As a result of increased adoption of smartphones, increasedadoption of broadband data devices (dongles, cards, MIDs), and increased usage ofeach class of device, they forecast a 108% compound annual growth rate of mobiledata. Figure 2-1 describes this exponential growth in data over the next five years.

Figure 2-1 Expected Growth in Global Wireless Data Usage

2.2 Data Offload Is Real and Measurable

A dynamic we will discuss at length in this whitepaper is “offload.” The notion is thattraffic originating on devices with wide-area data capabilities is “offloaded” to another

physical network, relieving the mobile operator of the need to transport it. Today,

offload occurs most frequently via Wi-Fi networks, largely because virtually all laptops


20/88


21/88


Version: 013.06.02 9

3. Introducing Four Segments, Circa 2012-2013

Before we discuss the “Mobile Broadband Era” in more detail (starting in Chapter 4)

we must first introduce several new segments. Each of these new segments is a

category of consumer (in many cases families) that operators might choose to targetwith a femtocell proposition. Describing the business case in terms of such segmentsmakes our discussion very concrete. In reality, there are an infinite number of possible

variations and there are segments with entirely different attributes that might beextremely attractive.

The segments an operator chooses to target are likely to reflect unique attributes oftheir country and of the operator. We have selected these segments to illustrate thelikely impact of femtocells in a variety of scenarios – each reflective of trends we seeunderway today.

3.1 The Technology Savvy Family

The “technology savvy family” has two smartphones, one laptop, and a mobile

internet device (MID). The MID includes many other types of devices: netbookcomputers, e-book readers, the iPad (pictured in Figure 3), and other special-purposedevices.

Figure 3-1 The Technology Savvy Family

3.2 The Socially Networked Family

The “socially networked family” is all about smartphones. Smartphones have increasedfrom a niche a few years ago to 25% of the market in Europe and 32% of the marketin the US.


22/88


Version: 013.06.02 10

By 2012 they will represent 37% of the market in Europe and 44% of the market inthe US. The “socially networked family,” with 4 smartphone users is, therefore, in anincreasingly likely demographic.

Figure 3-2 The Socially Networked Family

3.3 The Mobile Executive

The “mobile executive,” as shown in Figure 5, with one smartphone and a wireless

broadband enabled laptop, increasingly characterizes the modern mobile knowledgeworker. Many of the readers of this whitepaper will fit this profile.

This individual has a degree of mobility, but, more importantly, a desire to stayconnected and a willingness to pay for that connectivity. An executive who makes areasonable number of trips per month within a country could be highly motivated topurchase a mobile broadband plan but would not necessarily have usage patterns thatare dramatically different from those of the other segments that we have discussed.


23/88


Version: 013.06.02 11

Figure 3-3 The Mobile Executive

3.4 The Communications Enabled Firm

The “communications enabled firm” shown in Figure 3-4 is a first glimpse at the

enterprise femtocell space. We assume that the firm has 50 employees. It has 10employees with companyprovided smartphones and 10 employees with personalsmartphones using the femtocell. The firm also has 5 wireless broadband-enabledlaptops.


24/88


Version: 013.06.02 12

Figure 3-4 The Communications Enabled Firm

3.5 Details: Devices, ARPU, Usage

In three of the consumer segments (the Technology Savvy Family, the Socially

Networked Family, and the Mobile Executive) the usage per device is the same. Inconstructing these scenarios we have simply varied the number of devices per

household. The monthly APRU per device is also the same. A contract smartphone is€40 per month, a USB data dongle is €30 per month, and a connected MID device is€15 per month. Laptops, at this future date, consume 5 GBs per month (a figure

which includes fixed and mobile usage) wirelessly, in the absence of a femtocell. AMID, similarly, consumes 1 GB per month (reflective of its more limited functionalityand reduced monthly cost). A smartphone consumes 750 MOU per month and 500MBs per month. Both of these are significantly higher than European averages todaybut – given current trends are likely to grow.

Since video has such a profound impact on usage, its likely effect on mobile

consumption is bimodal. If the amount of mobile video were to stagnate at or neartoday’s level then these estimates of usage would be very high. On the other hand, ifpeople were to consume video freely as they do on their laptops and desktops (at

increasing levels) then these assumptions would be low. Currently, many mobileoperators have restrictions in place (mobile application “approval” requirements andfair use policies which limit monthly wireless broadband consumption, etc.). If theserestrictions were removed usage would very likely explode.

The enterprise has fifty employees. Ten employees have company-providedsmartphones and use 1000 MOU per month. Another ten employees have personalsmartphones. They use 750 MOU per month, just like their consumer counterpartsdescribed above. The weighted average usage, therefore, is 825 MOU per smartphone

user per month. The consumer ARPU is €40, as in other segments, and the business


25/88


Version: 013.06.02 13

ARPU (company provided phones) is €60. Finally, we assume that the enterprise hasfive laptop wireless data users in the firm, with data usage and pricing similar to thatof the three other segments described earlier in this section. The pricing and usage

associated with these wireless data subscriptions is similar to the wireless data usageof the other segments.

We assume that the operator charges the consumer a €50 “price” for an HSPAfemtocell plus €5 per month. Today, many operators are charging the full cost of thefemtocell up-front with no monthly fee. Other operators charge a monthly fee until thefemtocell is paid off in effect (in effect a lease-to-own plan). The model of “full costup-front” is wonderfully attractive to the operator since it reduces their incrementalinvestment per household to zero. It works with the very early market where a smallnumber of consumers are hungry for a femtocell (people with lots of money and

absolutely no cellular coverage at home). As that segment is satisfied the next tier inthe market will demand more attractive terms. The other model we see is a monthlyfee until the full price of the femtocell is paid off. This is also a reasonable model. With

such an offer there is most likely no commitment on the part of the operator to

upgrade the femtocell in the future.

We have selected the model we used last year, €50 up-front + €5 per month as areasonable consumer model. We assume, unlike the other price structures describedin this whitepaper, that the operator will upgrade the consumer every 36 months for

life. Since the femtocell builds loyalty and since the operator enjoys measurablemonthly cost savings as a result of having the femtocell in the home, it makes senseto keep the consumer happy with upgrades. Also, it is likely that by the time of thefirst upgrade cycle femtocells will be so inexpensive (and the gateways and other non-

recurring costs will be amortized) that it would be a poor business decision for theoperator not to upgrade someone who had a femtocell and was using it.

There is one twist on femtocell pricing which is technology-specific. We assume thatthe operator sells the HSPA femtocell for €50 and the LTE femtocell for €100. This is

fair, since the LTE/HSPA femtocell will be future-proof and will deliver significantlyhigher speeds to those with LTE-capable end-user devices. Technology enthusiasts willbuy an LTE/HSPA femtocell, whether they need it or not, and they won’t complainabout the higher price. In the mean time, the mass market (our assumption: 75%)

will be perfectly happy with an HSPA femtocell and they will appreciate the lowerprice.

The starting business case (the “START” bar, before a femtocell is added) is strongerfor LTE, because an LTE user consumes macro cellular data at a lower cost per GB.However, the cost of the dual-mode HSPA/LTE femtocell is higher ($275 wholesale

versus $100 wholesale), even after adjusting for the additional €50 in purchase price

(€100 €50 = €50 = $68, which is less than $275 $100 = $175). The LTE case issoftened as well by the reduced impact of network offload.

Ironically, the older and the less efficient the macro cellular network the moreimpactful a femtocell will be. An operator that is still carrying a lot of traffic on

GPRS/EDGE will see huge benefits from femtocells, because the cost of carrying trafficon the macro cellular network is enormous. As the macro cellular network ismodernized (and devices are updated to use its capabilities) the impact of femtocells

from a data offload perspective are reduced. Therefore, the offload benefit of an LTEfemtocell with an LTE device is slightly less than the benefit of an HSPA femtocell withan HSPA device if usage is equal. The net result is that in most cases the LTE/HSPAfemtocell business case (the ending value minus the starting value) is slightly weaker

than the HSPA femtocell business case, although the results are extremely close. As


26/88


Version: 013.06.02 14

LTE deployments become ubiquitous and LTE/HSPA devices replace HSPA-only devicesand as LTE/HSPA femtocells are reduced in price a cross-over will occur where themass market will demand the more capable device and an updated calculation ofoperator economics will say “Give it to them!”

The key characteristics of these segments are summarized in Table 3-1and Table 3-2.

Segment Femtocell Price3 SmartPhones

Laptops MIDs


Price: €50 + €5/Month 2 1 1


Price: €50 + €5/Month 4 0 0

The Traveling

Executive

Price: €50 + €5/Month 1 1 0

The Communications

Enabled Firm

Price: €2,000 +

€5/Phone/Month

20 5 0

Table 3-1 Number of Devices and Femtocell Pricing

3

LTE/HSPA up-front pricing is €100, and €2,500 for consumer and enterprise segments, respectively, asdescribed above.


27/88


Version: 013.06.02 15

Device ARPU4 Voice Data

Smart Phone €405 750 MOU 500 MBs

Laptop €30 N/A 5GBs

MID €15 N/A 1GB

Table 3-2 ARPUs, MOU, and MBs

As each subscriber adds a femtocell what happens? Our assumption is that usagepatterns change in a way that is proportional to the “pre-femtocell” usage. Weassume, as we did last year, that voice usage occurs 36% in the home, 24% in theoffice, and 40% elsewhere. We similarly assume that data usage occurs 45% in the

home, 25% in the office, and 30% elsewhere. When we quote a figure for monthlyusage (e.g. “750 MOU” or “5 GB”) this figure represents the combined usage in allvenues.

When a subscriber acquires a femtocell, we assume that usage changes, for a varietyof reasons. If the subscriber is in a coverage-challenged environment (which we

assumed here as our starting point) then their voice usage is likely to increase in theirhome (or office, in the case of the enterprise femtocell) because their phone willsuddenly ring while previously it may not have rung due to poor coverage. Similarly,the subscriber is more likely to make a phone call using his or her mobile phone whileat home if the subscriber knows that reliable coverage exists.

We assume that in coverage-challenged environments 33% of the voice traffic thatwould exist under “normal” conditions, does not exist (calls are missed or calls are notoriginated or calls are ended early due to poor quality). When coverage is restored we

assume that those minutes “return.” With the ratios described above, a consumer withpoor coverage at home would see a 36% (portion of usage at home) x 33% (uplift dueto coverage) = 12% increase in voice usage. In a calling party pays environment, theoutbound calls would be billable (increasing ARPU). The inbound calls would increase

operator profits (due to termination fees) but would not affect the subscriber’smonthly bill, because they would be paid by someone else.

In a mobile party pays environment the logic is a bit more complex. The subscriberdoes ultimately pay for the usage but we assume, because of the dynamics of largebucket plans common in the United States, that the revenue effect is somewhatmuted.

With data usage we have a parallel set of assumptions. We assume, however, thatdata is free (or at least included in the service bundle) in any femtocell environment.The jump in data usage results from a combination of several things (1) the realization

that it is free, (2) the increase in usage due to improved coverage (if you have no

macro cellular signal you are unlikely to use data applications, and (3) the jump inusage due to dramatically improved data speeds and reduced latencies. We assumethat the impact of the femtocell on the mobile phone is the greatest, since phones

don’t have Ethernet ports and only a modest percentage have Wi-Fi. We assume thatsmartphone data usage triples (e.g. 500 MB of usage jumps to 1500 MB) and thatlaptop usage increases by 50%.

4 Before introduction of the femtocell5 The Communications Enabled Firm has 50 employees, with 10 company-provides phones (1,000 MOU) and10 personal phones (750 MOU). Its composite usage is 875 MOU. The enterprise ARPU is €60 per

subscriber. The enterprise femtocell bundle includes free fixed and on-net mobile-to-mobile calling from theoffice.


28/88


Version: 013.06.02 16

In a non-femtocell world laptops users have many choice (including simply plugginginto a wired connection), so the likely impact on their behaviour is less. Our estimateof “pre-femtocell” usage only includes wireless connectivity. A mobile broadband userwithout a femtocell who has

5 GB of usage per month would almost certainly have additional usage in a wiredenvironment. When that user purchases a femtocell it is likely that much of the homeusage becomes wireless over the femtocell because it is suddenly “so easy” (no cablesto connect, no Wi-Fi radio to turn on or off, and performance which is most likelylimited only by the speed of the fixed broadband connection).

3.6 The Impact of LTE

LTE will be deployed very differently in different regions of the world. In the United

States it will roll out quickly and have near-ubiquitous coverage, because of the majoroperators who have invested heavily in 700 MHz spectrum. In Europe, the story is abit more complex. On the one hand, operators, who have aggressively promoted

mobile broadband, are experiencing the flipside of success: a need to rapidly growcapacity. These operators will be motivated to deploy LTE in the 2600 MHz band, eventhough it is a very expensive band in terms of cost of coverage. Other operators witha bit more headroom might choose to deploy lightly at 2600 MHz then more

aggressively in the digital dividend bands or 900/1800MHz when they becomeavailable. Our assumption, therefore, is that LTE will roll out more slowly in Europeand that the new technology will not enjoy the ubiquity of HSPA for some time.

A key question, then, in calculating femtocell economics is “What percentage of thetotal macro cellular data usage of LTE-capable devices will be carried on LTE?” A

parallel question is “What percentage of all devices will be LTE-capable?” LTE/HSPAdongles are most likely to lead the charge with LTE. MIDs (which demand less

performance and are more cost-sensitive) might be next. Phones (which must, bydefinition, support voice) are likely to be last.

We assume in our calculations that 70% of the traffic that an LTE/HSPA femtocell

offloads would have been carried on HSPA and that 30% would have been carried onLTE. Thus, the offload benefit of an LTE/HSPA femtocell with a mix of LTE-capable andHSPA-only devices is moderately less than that of an HSPA-only femtocell with HSPAdevices.


29/88


Version: 013.06.02 17

4. An Incumbent European Operator with HSPA

In this chapter we look at the impact of HSPA or LTE/HSPA femtocells on the foursegments discussed in the previous chapter.

4.1 Starting Point: Last Year’s Business Case

The scenarios we will soon describe in detail are extensions of a European consumer

scenario we discussed in our business case whitepaper last year. That scenario, calledthe European Coverage-Challenged family, portrayed a family with two mobile phones.We assumed an average contract voice usage of 400 MOU per month and an averagedata usage of 50 MBs per phone. We assumed that the family had no other datadevices.

In today’s environment, one can enjoy a lot of data services with that bandwidthallocation. As we look toward the future we believe that usage will change and theintroduction of other datahungry devices is likely to significantly change theconsumption pattern of the typical family.

The waterfall chart depicting the customer lifetime value of this family using the usageassumptions from the first whitepaper is shown in Figure 4-1. One can quickly observethat the network savings due to the femtocell is small – in large part because theamount of traffic flowing over the macro cellular network is small, relative to itscapacity. Modern networks, after all, can carry a vast amount of voice traffic very costeffectively.

Figure 4-1 European Coverage-Challenged Household (2009 Business Case Whitepaper)

The combination of the network cost savings, a monthly fee, and reduced churnresults in a basic value proposition of €2,504, an increase of €924 (58%) over thestarting value of €1,584.


30/88


Version: 013.06.02 18

4.2 A Journey into the Future

In this section we explore the deltas that we believe will exist between the presentday and the future, circa 2012-2013.

4.2.1

Changes in Behavior

As we journey into the future, what happens? Voice usage almost certainly increases.In Europe voice usage has been limited by relatively high tariff structures and by acalling party pays regulatory environment which – while extremely equitable andinclusive of all economic segments of society – does not lend itself to low tariffs. Astermination rates fall we can expect greater average levels of usage for similar ARPUs.

In the United States some operators have average usage levels over 2,000 MOU per

subscriber. The largest operators have recently starting offering and heavily promoting “all you can eat” tariffs. It is likely that usage in Europe will also move in an upwarddirection. We assume, therefore, that by 2012 usage increases from 400 MOU to 750

MOU per contract subscriber, with the same €40 ARPU we assumed in our startingscenario.

We similarly assume that consumers increasingly move from feature phones tosmartphones and that smartphone usage continues to climb. Usage across the internetis increasingly dominated by video. As mobile broadband speeds improve and as

smartphone displays and processors become more capable it is only natural that agreater and greater amount of video will make its way to mobile devices. Wirelessdata usage overall (reflecting increased adoption and increased usage per device) isgrowing at well over 100% per year. We believe it is reasonable, therefore, to assumeusage levels of 500 MBs per month in 2012 for smartphone-enabled users.

Finally, we can expect to see entirely new categories of devices. Apple has recentlyannounced the iPad. It will run mobile applications, browse the web, play music andvideo, display books, and perform a host of other functions – many data intensive. Itis likely to be owned by people who already own other wireless data devices, such assmartphones and laptops.

We have also seen an explosion in the past of netbooks, a new class of computingdevice that meets the need for portability and for a very low price point, for those whowould like to be connected, but who do not need the capabilities or who do not wish tocarry the bulk of a traditional laptop. A host of other data-centric devices will hit the

market as well. Some of the segments we described above include MIDs. We typicallyassume that the MIDs consume 1GB per month of data (versus the 5 GBs per monthof data we assume for laptops in 2012) and that the operator provides tariffscommensurate with that level of usage.

4.2.2

Changes in Cost

Network costs will also change as we move into 2012-2013. Operators who todayhave HSPA networks with traditional core networks will increasing move to flat IP core

networks. Many operators will launch LTE networks for additional capacity andadditional performance.

Femtocells will migrate from today’s traditional architecture, which routes all trafficthrough the operator’s core to a local breakout architecture, eliminating theincremental expense of data usage within the home femtocell, except for those

services which specifically originate within the operator’s network.


31/88


Version: 013.06.02 19

Finally, as mobile broadband becomes more common the cost per GB will fall andacquisition and retention costs will also decline.

The cost of a femtocell will also decrease, like any consumer product. We assumed inour original business case that the average wholesale price of a femtocell was $200.

By 2012 we expect HSPA femtocells to decline to $100, dual-mode LTE/HSPAfemtocells to sell for $275, and dualmode Mobile WiMA X / 1X / EV-DO femtocells tosell wholesale for $350. Arriving at a future price point for each category of femtocellwas not easy. The range of viewpoints among informed individuals in the industry iswide. The figures we cite are not the lowest or the highest.

Most agree that femtocells will follow a predictable path of declining prices, like anyconsumer product. They key question is “how quickly.” The femtocell value chaincurrently has many layers. As volumes increase the natural forces in consumerelectronics will push for a compression of the value chain. In the end the price point of

a femtocell may not be as low as a Wi-Fi access point but it will be much closer than itis today, thus making it more affordable to most consumers, even without a subsidy.

Just as Wi-Fi access points vary widely in price, as they target various segments,bundle various other functionality, and are embodied with “basic” or advancedstandards, we expect femtocells to do the same. The difference is that femtocells can

play a central role in managing operator costs. As they drop in unit cost, asprovisioning systems make deployment effortless, and as operators’ up-front costs getamortized among early adopters, we could see a growing force in the operator

community to make femtocells widespread and to remove many of the obstacles thatexist today which are curtailing consumer adoption.

4.2.3

Changes in the Femtocell Market

Some key beliefs about the femtocell market will also change. A key assumption in our

original business cases – supported by operator experiences – was that femtocellswould significant improve customer retention (or, stated differently, would reducechurn). As the market matures, as femtocells become less “novel,” and as competingoperators in a market start to offer femtocell solutions, we believe the impact offemtocells on churn will decrease. We have assumed, therefore, in our 2012 businesscases that femtocells will reduce churn from their original level by 10%, not the 20%we originally assumed.

4.2.4

The Technology Savvy Family

The Technology Savvy Family, with both a laptop and a MID, shows huge cost savingsto the operator from traffic offload plus some uplift for voice.

In this particular scenario, the network savings (€950.22) from offloading voice anddata traffic from the macro cellular network onto the femtocell network more than

offsets the initial cost of the femtocell (€130.95 = €100.00 + 80.95 in allocated costs,less €50 up-front fee). Additional revenue from increased in-home voice usage(€370.84), the monthly fee (€275.53 assuming

€5.00 per month) and reduced churn (€263.07), results in a basic value proposition of€2,642.54 versus the starting value of €913.83 – an increase in the customer lifetimevalue of €1,728.71.


32/88


Version: 013.06.02 20

Figure 4-2 Technology Savvy Family, HSPA Femtocell

Figure 4-3 shows the same family with an LTE/HSPA femtocell. The differencesbetween Figure 4-2 and Figure 4-3 are subtle. The starting value of Figure 4-3 isslightly greater. The low cost per GB of an LTE network is reflected in a slightly higher

pre-femtocell “START” value. The net cost of the LTE-capable femtocell is higher thanthat of an HSPA-only femtocell, even after adjusting for the difference in retail price.The network savings associated with the LTE-capable femtocell (assuming it issupporting some LTE capable devices and that they, in turn, would, in the absence of

a femtocell, send a portion of their traffic to LTE macro cellular sites) is less per GBthan that on an HSPA-only femtocell with HSPA-only devices (remember, legacytechnologies increase the value of a femtocell). Most of the other bars are the same.

The ending value (“BASIC VALUE PROPOSITION”) of Figure 4-3 is slightly higher thanthe ending value of Figure 4-2 due to the higher starting value but the gain in value(the impact of the femtocell) is slightly less, assuming identical usage. We will see asimilar pattern with most of the other segments.


33/88


Version: 013.06.02 21

Figure 4-3 Technology Savvy Family, LTE/HSPA Femtocell

4.2.5

The Socially Networked Family

The Socially Networked Family, with 4 smartphones, shows some uplift due to dataoffload and a lot of uplift due to voice (reference Figure 4-4).

In this scenario the network cost savings is not as great since the data usage is muchlower than it was in the first scenario. Still, the combination of the increased revenue

from the higher in-home voice usage (€741.68), the incremental revenue from themonthly fee (€275.53) and the reduced churn (€387.05), results in a basic valueproposition of €4,034.10, an increase of €1,647.01.

Figure 4-4 The Socially Network Family, HSPA Femtocell


34/88


Version: 013.06.02 22

The increase in value with LTE/HSPA femtocells is similar. The femtocells have ahigher wholesale cost (€275 vs. €100) but are purchased by early adopters at a pricepoint (€50 vs. €100) that reflections their greater functionality. The percentage of dual

mode (LTE/HSPA) vs. single mode (HSPA) femtocells will increase over time, as thedifference in price narrows, as LTE devices become more widespread, as residential

internet connection speeds increase, and as the additional capabilities of LTEincreasingly become a consumer expectation.

Figure 4-5 The Socially Networked Family, LTE/HSPA Femtocell

4.2.6 The Mobile Executive

The Mobile Executive is a “one person family” from the femtocell perspective. Weassume that the data package has been very aggressively priced. It is possible,therefore, that the starting lifetime value of the Mobile Executive is modest. We see inthe graph a huge uplift due to data offload. While the femtocell is not “leveraged” withthe Mobile Executive in the way that it is with a four-person family, the impact is still

large. Mobile executives are easy to find and likely to be early adopters of femtocelltechnology.


35/88


Version: 013.06.02 23

Figure 4-6 The Mobile Executive, HSPA Femtocell

4.2.7

The Communications Enabled Firm

In this last scenario we take a look at the enterprise femtocell. Here the starting valueof the firm is extremely high (€28,205.51). Traffic offload benefits are also large

(€3101.66). The network savings bar looks small simply because of the scale of itssurroundings.

Interestingly, some of the greatest upside in this scenario is in the opportunity torecruit a few of the other 30 employees not using the femtocell to the operator’snetwork. The value this bar (€ 9,826.93) is the number of potential new subscribers(employees not yet connected) times the value of a new subscriber, discounted by theprobability (20%) that the person would “convert.” In a family this usually results in asmall impact. In a corporation, where there are lots of employees and most are notusing the femtocell, but see their colleagues using it, the potential impact is much

greater. Finally, applications like PBX integration, presence (who is in the office andwho is not) and free calling within the office should be relatively easy to sell toenterprise customers.


36/88


Version: 013.06.02 24

Figure 4-7 The Mobile Executive, LTE/HSPA Femtocell

If there is a segment most likely to desire LTE-capable femtocells, it is the enterprisecustomer. This segment is already paying a premium for an enterprise femtocell (€2000). The incremental cost of an LTE capable femtocell one (€500) is small. Moreimportantly, the enterprise is likely to have a lot of demand for wireless data, is likely

to have a lot traffic originating and terminating on internal servers, and is likely tohave a fat backhaul connection. Collectively these factors make an LTE capablefemtocell a good value for the enterprise.

Figure 4-8 The Communications Enabled Firm, LTE/HSPA Femtocell


37/88


Version: 013.06.02 25

5. Operator View: Multiple Segments

In this chapter we look at the economic proposition of an operator targeting multiplecustomer segments, such as those that we identified in Chapter 4.

5.1 Rationale / Economies of Scale across Segments

We have up to this point discussed the business case for promoting femtocells within a

single customer segment. While an operator might initially target a single segment,once they gain confidence in the femtocell technology and in the correctness of thebusiness case it is likely that they would seek to exploit femtocell technology acrossmultiple segments. This strategy could be as simple as pursuing a consumer segment

in parallel with an enterprise segment. Many of the costs associated with femtocelldeployment are fixed (e.g. billing and provisioning integration costs and initialplatform investments). The more femtocells that are deployed the less significantthese upfront costs are to the business case.

In this study we look at an operator deployment in which the operator addresses each

of the four segments described in Chapter 4. This exercise will provide significantinsight into the economics of a multi-segment deployment. It will also showcase someof the analytical “views” available in the latest version of the femtocell businessplanning model, which is available to full members of the Femto Forum and developedby SRG.

5.2 Key Assumptions

In framing the problem we must ask two questions:

1.

How large is the total femtocell market?

2.

What portion of market will belong to LTE/HSPA femtocells and what portionwill belong to HSPA femtocells?

We assume for purposes of the scenario (not as a result of an analytical exercise) thatat the end of the planning period one in ten subscribers will have access to a femtocelland 25% of those subscribers with femtocell access will be using dual-mode LTE/HSPAfemtocells as opposed to HSPA-only femtocells. The number of subscriber then withLTE/HSPA femtocells will be 10% x 25% = 2.5% of the total subscriber base.

We then ask, “How are those femtocells distributed across the various segments?” Ifwe know the number of devices served per femtocell within each segment (where a

device is a phone, USB modem, or MID) then we can construct the information

contained in Table 5-1.

Market Segment Femtocells Devices/Femto Subscribers

Enterprise 5.0% 25 30.9%

Smart Phone Family 15.0% 4 14.8%

Smart Phone / Laptop / MID Family 30.0% 4 29.6%

Broadband-Connected Individual 50.0% 2 24.7%

Table 5-1 Distribution of Femtocells and Devices in a Multi-Segment Market

Interestingly, we see two very different ratios. One ratio is the distribution offemtocells across the various segments. The other ratio is the distribution ofsubscribers across the different segments. Even though the enterprise segment

represents only 5% of the femtocells it serves 30.9% of the devices with femtocell


38/88


Version: 013.06.02 26

access. In contrast, the broadband-connected individual represents 50% of thefemtocells but serves only 24.7% of the subscribers.

In this discussion we are considering “devices” and “subscribers” to be analogous. Inthe context of a European operator each would represent a SIM card. We could,

therefore, talk about SIM penetration, even through the devices and the ARPUs arequite different and in many cases the same physical person might use both devices(e.g. a phone and a laptop).

The distribution of devices – times an adoption growth curve – is shown in Figure 5-1.

Figure 5-1 Number of Devices with Access to an LTE/HSPA Femtocell

The distribution of femtocells – times an adoption growth curve – is shown in Figure 5-

2.


39/88


Version: 013.06.02 27

Figure 5-2 Number of LTE/HSPA Femtocells by Year

Finally, it is important to remember that in the context of the total subscriber base weare talking about relatively small numbers (see Figure 5-3). It is possible that themarket could be many times larger.


40/88


Version: 013.06.02 28

Figure 5-3 Number of Devices with Access to an LTE/HSPA Femtocell

5.3 Summary Results

The following graphs show some of the important composite results. If an operator

were serving only a single segment we would present a similar set of views. Figure 5-4shows the present value of several important cash flows: (1) the present value offemtocell investments (show by category), the present value of the femtocell salesrevenue (the price of the femtocell to the subscriber times unit volumes, discountedby time), the present value of effective subsidies (the difference between the first twostacked columns), and finally the present value of profit contributions.


41/88


Version: 013.06.02 29

Figure 5-4 Present Value of Femtocell-Related Expenditures, Retail Sales, Subsidies, and

Profit Contributions

Figure 5-5 shows the femtocell-related investment in greater detail. We see thatduring the first year the operator is putting in billing and provisioning systemintegration and it is making the required up-front platform investments, a totalinvestment of $2,250,000.

Each subsequent year is dominated by femtocell purchases with small allocations forplatform purchases. We assume that the operator offers each subscriber a femtocellupgrade every three years. An upgrade expense, therefore, appears every three yearsfollowing the original purchase.


42/88


Version: 013.06.02 30

Figure 5-5 Femtocell-Related Expenditures by Category by Year

How long does it take for the business to become profitable? This analysis, as shownin Figure 5-6, indicates that the business is marginally profitable in the first year ofactual deployment. We assume, for the sake of this illustration, that the market growsrapidly for five years then remains static. No real market would behave this way (theywould normally keep growing) but modelling such an adoption curve is extremely

helpful in understanding profitability during and after a dramatic growth spurt. We seethat once the market plateaus and the operator is no longer investing heavily infemtocell deployments that profits soar. We have seen these patterns before in thecellular industry as its growth phase gave way to a more mature market. The yearlyprofit bars have a “wave” due to the three year “echo” of femtocell upgrades.


43/88


Version: 013.06.02 31

Figure 5-6 Profit Contribution by Year

An equally instructive view is presented in Figure 5-7, which shows cumulative profits.Here you can see that it takes until 2015 for the market to break even. Theinvestment in integration and in platforms in the first year takes several years ofoperating profits to amortize.

Figure 5-7 Cumulative Profit Contributions / Breakeven Analysis


44/88


Version: 013.06.02 32

6. Operator View: Large and Small Operators / Economies ofScale

The initial business cases we developed for femtocells assumed relatively large

operators. Increasingly, small operators have asked, “What are the economics offemtocells in my environment?” Other small operators have committed to full-scaledeployments. We will have much more definitive answers a year or two from nowwhen more small operators have made the decision to deploy. Even so, it is possibleto discuss the implications of scale and highlight some of the trade-offs that arepossible.

In the study we will imagine that an operator provides femtocells to the Technology

Savvy Family. We further imagine that 10% of their total subscriber base – after fiveyears of marketing – has access to a femtocell. Since each femtocell is serving afamily this translates into 2.5 femtocells per 100 mobile devices, where a device is aphone, a laptop, or a MID. We then look at the magnitude of the operator investment

as a function of the size of the operator. We consider very large (50 millionsubscribers), large (10 million subscribers), small (1 million subscribers) and verysmall (200,000 subscribers) operators. The focus of this study is to understand how

fixed and variable costs affect large and small operators differently. The objective is tohelp small operators understand the choices available to them to manage the up-frontcosts associated with femtocells.

6.1 Important Up-Front Costs

Two “line items” in the list of operator expenditures stand out as extremely scale-dependent. These are (1) billing and provisioning integration costs, and (2) theminimum investment in femtocell-related core network infrastructure.

Billing and provisioning integration costs are up-front expenditures. They typicallyoccur during a “pilot” stage, before the operator has launched a commercial offer. Ifan operator has not fully integrated their essential systems the result could be a poo

013 Business Case Femtocells Mobile BB Era

Documents