Globe

Tutela Experience Index (TEI)

Tutela has developed a new method for estimating
Quality of Experience using crowdsourced data

Subscriber experience scores (%) An overall score is calculated based on a

are calculated for the following use weighted average of the individual
cases application types

Video
Streaming
Web
Browsing
Connected
Apps
TEI Score (%)

Content Network Video VoIP

Sharing Gaming Calling

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Quality of Experience Depends more on Consistency than Averages
Consistency is a foundation of subscriber experience

Operator B has higher mean, median and

Network Interactions

A B
peak performance…
…but subscriber experience will be better
on Operator A’s network because it has
fewer ‘bad’ interactions
Below the Netflix threshold Above the highest
for HD video - threshold for 4K video –
unsatisfactory subscriber smaller Return on
experience and potential Investment
for churn

5 Mbps 45 Mbps Download Throughput

Poor Good enough Great

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Globe Download Throughput Distribution – 12 Month Change
Improving proportion of measurements below 5Mbps

Download Throughput Distribution and Median

12 100%

Percentage of Measurements
Download Throughput (Mbps)

10
80%

8
60%

40%
4

20%
2 Proportion of measurements
below 5 Mbps has improved
by 14 percentage points

0 0%
Jan-21 Feb-21 Mar-21 Apr-21 May-21 Jun-21 Jul-21 Aug-21 Sep-21 Oct-21 Nov-21 Dec-21

< 5Mbps 5 to 45 Mbps > 45 Mpbs Median DL Linear (Median DL)

Source: Tutela, 3G/4G/5G

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Comparison of Download Throughput Distribution
Ranked by median download throughput

Download Throughput Distribution

100% 60
Percentage of measurements

50
80%

Median Download Throughput (Mbps)

40

60%

30

40%
Smart, PH 20

20% DITO, PH
10
Globe, PH

0% 0

< 5Mbps 5- 45Mbps > 45 Mbps Median D/L

Source: Tutela Experience Index, 3G/4G/5G, December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Comparison of Download Throughput Distribution
Ranked by lowest proportion of measurements below 5 Mbps

Download Throughput Distribution

100% 60
Percentage of measurements

50
80%

Median Download Throughput (Mbps)

40
60%

30

40%
Smart, PH
20

20% DITO, PH
10

Globe, PH

0% 0

< 5Mbps 5- 45Mbps > 45 Mbps Median D/L

Source: Tutela Experience Index, 3G/4G/5G, December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

KPI Comparison
Difficult to identify which network is providing the best subscriber experience

Source: Tutela Experience Index, 3G/4G/5G, December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Objective: Estimate Quality of Experience (QoE)
QoE = Subscriber Expectations vs Quality of Service (QoS)

Internet

Transport
Network
Quality Performance CORE Quality of Experience
of Service QoE includes the complete end-to-end system effects
((( ))) (client, terminal, network, services infrastructure, etc.) and
((( ))) may be influenced by user expectations and context. Hence
((( )))
in principle, QoE is measured subjectively by the end-user
RAN and may differ from one user to the other. However, it can
be estimated using objective measurements.

Device

Expectations Subscriber

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Rating: Likelihood of Meeting Expectations
Quality of Experience thresholds required to meet different levels of expectation
Network Interactions

Likelihood of
Expectations met
expectations met

Always 100%

Very Frequently 90%

Mostly 75%

Sometimes 25%

Rarely 10%

Never 0%

Quality of Experience (QoE)

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Globe: Tutela Experience Index and Use Case Scores
Globe has closed the gap to Smart over the last 12 months
Glob e , Philip p in e s Tute la Exp e rie nce In d e x
De c-2 1 m o b ile

100% 100 %
Always

80% 80% Ve ry Fre que ntly

86%

Mostly
60% 60%
53%
60%
58% So metime s
53%
40% 40% 46%
41%
Rare ly
35%

20% 20% Ne ve r

SE Score
0% 0%

TEI Vid e o We b Conne cte d VoIP Vid e o Ne twork Conte nt

Stre aming Browsin g Ap p s Calling Gaming Sharing

m a rke t p o sitio n 2nd 2nd 2nd 2nd 1 st 3rd 2nd 3rd

δ to lead er: De c-21 -0.9% -2% -0.7% -0.2% 6% -4% -0.3% -4%
De c-20 -1 5 % -3 4 % -5% -2 8 % -2 2 % -1 5 % -2 5 % -2 4 %

Source: Tutela Experience Index, December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Globe: Tutela Experience Index and Use Case Scores
Globe is steadily improving and closing the gap to Smart
70%

60%
Smart, PH

Linear (Smart,
PH )

50% Globe, PH

Linear (Globe,
PH )

40% DITO, PH

30%

20%

10%

0%
Aug-20 Sep-20 Oct-20 N ov-20 Dec-20 Jan-21 Feb-21 Mar-21 Apr-21 May-21 Jun-21 Jul-21 Aug-21 Sep-21 Oct-21 N ov-21 Dec-21

Source: Tutela Experience Index

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 5G NSA: Early Indications Are Positive
Subscriber Experience Expectations Met Subscriber Experience Expectations Met

100% 100%

80% Always Very Frequently 80% Always Very Frequently

76% 76%

67%
60% Mostly Sometimes 60% Mostly Sometimes

54%

40% 40%
Rarely Never Rarely Never

20% 20%
TEI TEI

10%

0% 0%
Globe, 3G Globe, 4G Globe, 5G NSA Globe, 5G NSA Smart, 5G NSA

Source: Tutela Experience Index, December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 Experience Score: Video Streaming (mobile)
Video Streaming Expectations Met Download Throughput Distribution

100% 100% 18

16
> 25Mbps
15.6
80% Always Very Frequently 80%
14
<=25Mbps
12

60% Mostly Sometimes 60%

11.2 <= 5Mbps
10
55%
53% 9.6
8
46% <= 3Mbps
40% 40%
Rarely Never
6
<= 1.5Mbps
4
20% 20%
TEI
median (Mbps)
2

0% 0% 0
Globe, mobile Smart, mobile DITO, mobile Globe, mobile Smart, mobile DITO, mobile

Source: Tutela Experience Index, December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 Globe Video Streaming (mobile)
Video experience score and related QoS KPIs
Glo b e , Ph ilip p in e s TEI Vid e o KPI Distib u tio n s
De c-2 1 m o b ile

1 00 % 1 2.0 1 00 % 5 .0

Always
> = 4 5 Mb p s 4 .5 Bu ffe r Fill
> 1 .5 Mb p s
1 0.0
0% <2s
8 0% 8 0% 4 .0
Ve ry
Fre q u e n tly < 4 5 Mb p s
Packe t
3 .5 Lo ss/ Discard
8 .0 < = 1 .5 Mb p s
Mo stly <= 2% <4s
6 0% < 1 1.6 Mb p s 6 0% 3 .0
53%
Up lo a d
6 .0 2 .5
So me time s
< 4 .5 Mb p s
< = 0 .7 5 Mb p s <= 4% <6s
4 0% 4 0% 2 .0
Do wn lo a d
Rare ly 4 .0
< 1 .5 Mb p s
1 .5

> 4% > 6s
2 0% 2 0% 1<.0= 0 .5 Mb p s Ne two rk
Ne ve r
2 .0< 0 .7 Mb p s

0 .5

0% 0 .0 0% 0 .0

Vid e o Dow nload Up load Packe t Loss/ Discard Buffe r Fill ' Ne ve r'

Quality of
Quality of Service
Experience
Source: Tutela Experience Index, December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Drill-down Analysis
 Variable Subscriber Experience in Zamboanga City
Congestion analysis confirms performance issues

• Globe’s Tutela
Experience Index in
Zamboanga City is 37%,
well below the national
figure of 53%

• The map view highlights

many areas where
performance is below the
minimum level for many
mobile services

• Time of day analysis

highlights congestion
from 6am to 1am

• Packet Loss/Discard is
particularly high, with
33% of all
measurements having
some degree of loss or
discard
Globe: 33% of measurements with packet loss
Source: Tutela, October – December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 Area Near Zamboanga City Medical Center and Shopping District
Globe network has high volume of packet loss/discard and high latency, impacting subscriber experience

Tutela Experience Index:

Globe:
44%
Hospital Smart:
51%

Avg. Packet Loss/Discard:

Globe:
6.8%
Shopping Mall
Smart:
7.8%

Avg. Latency:

Globe:
98ms

Smart:
69ms

Avg. Download Throughput:

Globe:
40% of Globe measurements have packet loss/discard (Smart: 33%) 11.5 Mbps

Smart:
Source: Tutela, November/December 2021 15.4 Mbps
Copyright Tutela Technologies Ltd. Confidential and under NDA.
Avg. Upload Throughput:
 Difference in eNodeB performance suggests RAN performance issues
Two eNodeBs in same location with substantially different performance

Tutela Experience Index 48% Tutela Experience Index 32%

13% of measurements have 47% of measurements have packet

packet loss / discard loss / discard

Average download throughput: Average download throughput: 7.6

7.0 Mbps Mbps

Average upload throughput: 5.6 Average upload throughput: 3.7

Mbps Mbps

Average latency: 63 ms Average latency: 115 ms

Average packet loss/discard: Average packet loss/discard: 11.5%

2.8%
Average signal strength: -99 dBm
Average signal strength: -98 dBm

eNodeB ID: 342003 (FDD) eNodeB ID: 244003 (TDD)

Source: Tutela, 1st November – 31st December

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 Difference in eNodeB performance suggests RAN performance issues
Two eNodeBs in same location with substantially different performance excluding typhoon dates

Tutela Experience Index 52% Tutela Experience Index 38%

10% of measurements have 40% of measurements have packet

packet loss / discard loss / discard

Average download throughput: Average download throughput: 8.9

6.9 Mbps Mbps

Average upload throughput: 5.9 Average upload throughput: 4.0

Mbps Mbps

Average latency: 65 ms Average latency: 124 ms

Average packet loss/discard: Average packet loss/discard: 12.5%

2.5%
Average signal strength: -99 dBm
Average signal strength: -98 dBm

eNodeB ID: 342003 (FDD) eNodeB ID: 244003 (TDD)

Source: Tutela, 1st November – 7th December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 Difference in Cell performance suggests RAN performance issues
Five cells with significantly lower performance than the others on the same eNodeB

eNodeB ID: 244003

Connection CID Download Upload Throughput Packet Loss/Discard Latency (ms) Signal Strength (dBm)
Throughput (Mbps) (Mbps)

62464911 10.4 2.9 4.3% 59 -94

62464910 2.5 0.6 13.3% 148 -96
62464909 5.4 1.4 10.7% 111 -96
62464901 6.7 1.7 3.9% 64 -91
62464900 5.9 0.4 23.9% 194 -101
62464899 8.5 0.7 17.3% 160 -95
62464811 9.4 2.3 6.0% 68 -99
62464810 5.0 0.9 31.0% 218 -97
62464809 12.2 3.4 8.4% 105 -90
62464801 13.6 6.5 6.3% 80 -97
62464800 8.9 5.6 2.9% 65 -104
62464799 4.8 5.4 6.8% 78 -100

KPI figures are averages

Source: Tutela, November/December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 Second Example of FDD outperforming TDD eNodeB
Two eNodeBs in same location with substantially different performance

25% of measurements have 38% of measurements have

packet loss / discard packet loss / discard

Average download Average download

throughput: 8.6 Mbps throughput: 11.3 Mbps

Average upload throughput: Average upload throughput:

7.0 Mbps 3.9 Mbps

Average latency: 80 ms Average latency: 78 ms

Average packet loss/discard: Average packet loss/discard:

3.1% 9.6%

Average signal strength: -95 Average signal strength: -99

dBm dBm

Average Tutela Experience Average Tutela Experience

Index 52% Index 30%

eNodeB ID: 343391 (FDD) eNodeB ID: 243391 (TDD)

Source: Tutela, November/December 2021

Copyright Tutela Technologies Ltd. Confidential and under NDA.

Follow-up Analysis (February
2022)
 Additional Analysis for eNodeB 244003
Indoor performance and fibre outage do not explain difference between TDD and FDD performance
1800MHz single beam
All results Indoor** Outdoor** No fibre outage++

Tutela Experience
38% 36% 35% 41%
Index

measurements with
40% 45% 34% 38%
packet loss/discard

Average packet
12.5% 15.3% 11.0% 12.2%
loss/discard

Average download
8.9 8.5 9.7 8.9
throughput (Mbps)

Average upload
4.0 4.3 3.9 3.6
throughput (Mbps)

Average latency (ms) 124 139 116 122

Average signal strength

-99 -100 -96 -100
(dBm)

eNodeB ID: 244003 (TDD)

**some results are unclassified for indoor/outdoor, i.e. where it is not possible to determine
Source: Tutela, 1st November – 31st December
++excludes measurements from days with a known fibre outage
Copyright Tutela Technologies Ltd. Confidential and under NDA.
 FDD/TDD Indoor/Outdoor Comparison
…

Indoor** Outdoor** Indoor** Outdoor**

Tutela Experience Tutela Experience

51% 45% 36% 35%
Index Index

measurements measurements
with packet 12% 14% with packet 45% 34%
loss/discard loss/discard

Average packet Average packet

2.4% 3.8% 15.3% 11.0%
loss/discard loss/discard

Average packet Average packet

1.9% 2.7% 9.4% 5.7%
loss loss

Average download Average download

6.7 6.8 8.5 9.7
throughput (Mbps) throughput (Mbps)

Average upload Average upload

5.6 5.6 4.3 3.9
throughput (Mbps) throughput (Mbps)

Average latency Average latency

62 66 139 116
(ms) (ms)

Average signal Average signal

-98 -95 -100 -96
strength (dBm) strength (dBm)
eNodeB ID: 342003 (FDD) eNodeB ID: 244003 (TDD)
**some results are unclassified for indoor/outdoor, i.e. where it is not possible to determine
Source: Tutela, 1st November – 31st December

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 342003 – Indoor/Outdoor Comparison
Indoor KPIs stronger than Outdoor may be due to closer location to base station as shown by Timing Advance figures

Indoor** Outdoor**

Tutela Experience Index 51% 45%

measurements with packet

12% 14%
loss/discard

Average packet loss/discard 2.4% 3.8%

Average download
6.7 6.8
throughput (Mbps)

Average upload throughput

5.6 5.6
(Mbps)

Average latency (ms) 62 66

Average signal strength

-98 -95
(dBm)

Average Timing Advance 0.6 3.0

342003 (FDD) - Indoor 342003 (FDD) - Outdoor
**some results are unclassified for indoor/outdoor, i.e. where it is not possible to determine
Source: Tutela, 1st November – 31st December

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 244003 – Indoor/Outdoor Comparison
Again, Indoor measurements have significantly lower Timing Advance due to proximity of the Medical Centre

Indoor** Outdoor**

Tutela Experience Index 36% 35%

measurements with packet

loss/discard
45% 34%

Average packet loss/discard 15.3% 11.0%

Average download
throughput (Mbps)
8.5 9.7

Average upload throughput

(Mbps)
4.3 3.9

Average latency (ms) 139 116

Average signal strength

(dBm)
-100 -96

Average Timing Advance 2.3 4.6

244003 (TDD) - Indoor 244003 (TDD) - Outdoor
**some results are unclassified for indoor/outdoor, i.e. where it is not possible to determine
Source: Tutela, 1st November – 31st December

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 Area Near Zamboanga City Medical Center and Shopping District
Busy Hour analysis shows significant difference in packet loss/discard between quiet and busy hours

13% peak, compared to

6.5% across Zamboanga

Surge in packet loss/discard is double the peak seen

across Zamboanga indicating congestion may be
localized (RAN/fronthaul/backhaul?)

Source: Tutela, 1st November – 7th December

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 Download Throughput Distribution in Zamboanga City
Over a third of download throughput measurements are below 5Mbps

Source: Tutela, 1st November – 7th December

Copyright Tutela Technologies Ltd. Confidential and under NDA.

 KPI Analysis for eNodeB 243391
TDD base station with better performance – more 1800 MHz cells than 342003?

All results

Tutela Experience
48%
Index 1800MHz dual beam
measurements with
36%
packet loss/discard

Average packet
2.8%
loss/discard

Average download
14.2
throughput (Mbps)

Average upload
3.6
throughput (Mbps)

Average latency (ms) 80

Average signal strength

-100
(dBm)
eNodeB ID: 243391

Source: Tutela, 1st November – 7th December

Copyright Tutela Technologies Ltd. Confidential and under NDA.

UDP Latency, Jitter and Packet Loss Test Methodology
Closest test servers to Globe are located in Seoul and Singapore
A server response (SR) test is defined as a test which determines the latency, jitter and packet loss of the current connection using UDP. When a server response test is performed, the Tutela
SDK determines (based on the current configuration) how to set up the server response test. Server response tests are performed against one of twenty test servers located around the world.
Multiple servers are used to provide low latency based on geography .
To accurately measure these KPIs, a predetermined number of packets are sent to the test servers, usually 100 packets. Each of these packets contains a 32-byte payload which holds a number
indicating the ordering of that packet in the sequence of packets sent out.
These tests are then performed as follows:
1. The SDK chooses the closest test server based on the geographical distance between the device and the server locations.
2. Each packet is then sent to the test server via UDP which is connectionless. This means that, unlike TCP connections, there is minimal handshaking involved which could produce
measurement results that indicate slower-than-actual speeds.
3. UDP protocol is also used to simulate latency-sensitive applications, such as video and VoIP calling.
4. The precise time that each packet is sent out is recorded to nanosecond precision and stored for later calculations.
5. The packet spacing is 20ms between packets
6. The test server receives the packets sent from the SDK and then immediately echoes them back to the listening device. The precise time that each packet is received back on the device is
also recorded along with the specific payload of each packet. As this payload indicates the ordering of the packets returned to the device.
7. The Metro Ethernet Forum (MEF) 10 specification is used to calculate the latency, jitter, and packet loss. The MEF 10 specification is known to accurately measure jitter in real networks
where packets can be sent with any uniform or bursty traffic patterns.
1. Latency is determined as half of the round-trip travel time of each packet from the time the packet is sent from the SDK to when the same packet is received.

2. Jitter is calculated by measuring the change in latency from packet to packet received by the SDK.

3. Packets that are never received on the device are classified as lost, while we categorize packets that were either duplicate or out-of-sequence as discarded as they can no longer be used for latency or jitter measurement calculations.

Copyright Tutela Technologies Ltd. Confidential and under NDA.

More insights and data
available at tutela.com
John Davies
[email protected]
London: +44 7826 871629
31