• Legacy switch revenues continued to decline. NGN revenues not large enough to plug that hole

• Acquired Tamblin, developer of interactive TV programming and advertising for IPTV

• NGN wins during the quarter: Hits Telecom Uganda for a Mobile NGN solution; Mongolia’s Mobicom for a nation-wide NGN network, including softswitches and gateways; and Polish operator Exatel SA for an IMS platform.

• IPTV wins during the quarter: SingTel and Shanghai Telecom

• Enterprise solutions: IP usage penetration rates increased, reaching 43% year to date, versus 36% for the same period in 2006.

I have asked nearly all major vendors about the carrier RFPs that require IMS-compliant IPTV solution set. There are no such RFPs floating around … precisely because there are no set IMS specs for IPTV technology. They are still being worked at by the Tispan standards group. It is however common for carriers to seek migration path toward IMS compliance in the IPTV equipment they procure. So what you have in the market is ‘IMS ready’ rather than ‘IMS compliant’ solutions. Thomson and Ericsson seem to be ahead of the curve.

The benefits of an IMS-compliant IPTV solution are obvious. Service providers will be able to offer both voice and IPTV services over the same core infrastructure, leveraging the same CSCF, the same HSS, and the same telephony feature server for instance. There is some interest in that for sure. But if video application server is connected to IMS then any SIP enabled client (which can include mobile phones) can reach this video application server, so that with the same platform you serve multiple devices. There seems to be more service provider interest in the latter.

Just thinking what Google’s OpenSocial could lead to. If there are common APIs then I guess one voice softclient fits all. If an ASP’s VoIP softclient works with Orkut, it will work with XING, MySpace, LinkedIn … the whole lot. You are therefore likely to see a bunch of voice clients being offered as options on these sites but one or two being promoted actively. That gives social networking companies more power because they can have a bigger say in who they promote and make serious money in the process. So expect eBay to keep spending on Skype for a while.

Standardization of APIs will lead to commoditization. Apart from seeing guys like Facebook in ETSI working group meetings, features available on social networking sites will become commoditized. One of those features will be voice. This will no doubt result in more robust and off-the-shelf social networking software platform available out there that comes pre integrated with telecom applications. So will the likes of Sylantro and Broadsoft start developing social networking features or will they just opt for partnerships with some developers? Well, either one of the options will do. But they are likely to respond quick.

• Access line reduction of about 588k during the quarter. Some of that is VoIP substitution. Majority is mobile substitution. There is also some migration within AT&T’s fixed line to AT&T’s cellular services. The company added near about 2 million new mobile customers during the quarter

• Reduction in business voice lines during the quarter: 190k. Line reduction due to IPPBX and hosted VoIP, and other competitive factors

• CallVantage VoBB subscribers: not sure about the number. If they have 13 million broadband subscribers, 10% of that is around 1.3 million. So if we are talking about 10% penetration, AT&T might have over a million VoBB subscribers. That seems very unlikely though. One correction to our 2Q07 post on AT&T: the company uses Sylantro for business VoIP, and not for consumer VoIP. For consumer VoIP (CallVantage) AT&T uses its home grown solution.

• Fixed line reduction during the quarter: 16k consumer lines and 3k business lines. VoIP service launch is a recent development. Estimated subscribers: not more than 30k

• 3.8m ADSL subscribers. 378k FTTX subscribers

• 358k IPTV subscribers. Net additions during the quarter: 24k

• “3.9% Local revenue decrease and the 7.2% Domestic Long Distance revenue decrease, mainly due to mobile and VoIP substitution”

• Line reduction during the quarter: 247k lines. 21% of those lost to VoIP.

• Added 53k new VoIP subscribers during 3Q07. Total subscribers: 790k

• VoIP subscriber base represents 38% penetration into broadband accounts

• Netherlands VoIP market: KPN 790k subscribers; Cable 980k subscribers; Others 260k. Total around 2 million subscribers.

• KPN adding 10k new VoIP subscribers per week. Expecting that to go down to 8k per week over the next quarters

Is mobile VoIP part of overall mobile telephony arbitrage that includes mobile callback, FMC, and Voice over Wireless Broadband (VoWB)?

So far as termination into traditional public network is concerned, you could view it like that.

How long have you been providing mobile VoIP?

We have been around since 2001. By 2004 we had our VoIP softphone ready. We have built our technology around SIP, and not any proprietary technology. We offered the downloadable version for mobiles in 2005 and also combined that with callback solution so from the softphone people could initiate callback.

What is the advantage in building a SIP mobile client?

Nokia and other handset makers are coming out with WiFi-GSM dual mode phones supporting SIP. This will drive the mobile VoIP market and serves as just one of several reasons for building a SIP mobile client.

When exactly did you decide to focus more towards white labeled mobile VoIP?

We launched our mobile VoIP retail service called Challenger Mobile in December 2006. During 2007 however we changed the focus towards white labeling our offering.

Who are your potential customers for white labeled offering?

We believe that players like MVNOs and wireless ISPs will be interested in providing this service to their customer base. MVNOs will be the first ones to move in this direction. As soon as we see one MVNO success, the others will follow.

I think I might have found the reason for the lack of killer applications in the communications industry. It is the lack of female staff in the industry. There you go. Bingo :)

I have never had problems with women talking Internet Protocol. In fact I have hardly ever felt the difference when being briefed by a female on technology aspects. And I am sure you boys and girls have no issues either. However the fact is that there is only about 10 to 15% female staff working at tech companies. There is a serious dearth of female talent in this industry.

What that means is that technology itself is lacking a critical input. You cannot hope to build the solutions and applications for all of us without having active input from nearly half of us.

If content providers cover more and more stuff that women do in communications industry, that might attract more female talent towards this important sector that is changing the way we live. We at iLocus are making a start today. We start the crusade with interviews with two ladies, living continents apart. Both CEOs, both founders and both enabling a similar revolutionary service: mobile VoIP. Christine is CEO and Co-Founder of Challenger Mobile and Rapelang is CEO of Yeigo. Rapelang’s interview will appear on the site tomorrow.

I am a great fan of the Webinar concept. Those who are required to travel a lot know the value proposition of this wonderful alternative. Large business transactions in telecom may still require face to face meetings. But attending conferences and interactive discussions can easily be substituted by Webinars. And while you are at it, you help reduce those CO2 emissions related to business travel.

I am however intrigued by the lack of Webinar offers and awareness in Asia Pacific. The region is decidedly underserved in this regard. I get a Webinar invitation almost every day from one source or the other. But none of these are geared towards the Asia Pacific audience. We are therefore inclined to take up the challenge. Watch out for our first live Webinar focused on Asia Pacific service provider community. The topic is “Pre-Requisites for VoIP Peering”. Date: 19th December. That will be the first in our series of Webinars focused on Asia Pacific.

The commercial launch of Sprint’s mobile WiMAX is expected in 1Q08. Most other service providers are expected to formulate their WiMAX strategies accordingly. There are several small commercial WiMAX services around. However they are mostly fixed in nature. Mobile WiMAX is largely confined to trial networks. There are many skeptics who believe that mobile and WiMAX cannot go hand in hand.

A more important question is whether Sprint will be able to drive new subscriptions or are we mostly talking about migration from one service plan to the other. Nevertheless, we are looking at a major test bed. Sprint made investments of $73 million during 3Q07 into its WiMAX project. One of the tests Sprint will carry out for us is whether mobile WiMAX can be something beyond data centric. Once it is established how mobile WiMAX fits in the existing telecom ecosystem in terms of user experiences it will be touch clearer for others to have a go at the service.

Sprint expects revenues of over $2 billion per year by 2010 from mobile WiMAX and majority of that is expected to be new business.

Are the wireless networks going to migrate over to VoIP in much the similar way that fixed networks did: core first and then the edge?
That is one logical path of pushing packetization to the edge. But in the wireless market, the end user VoIP is taking off in parallel. That is something fixed networks did not see in late nineties.

There are various flavors of end user mobile VoIP. How would you categorize them?

Well, apart from mobile callback, which is not necessarily mobile VoIP, you have the likes of Skype and Fring that are targeting PC based VoIP users more and trying to transfer or migrate them over to mobile VoIP. They are looking to leverage the existing PC VoIP user market to build their user base. We view mobile VoIP as a new standalone market where PC is cut out of the equation. Truphone type companies leverage SIP client capability of existing smartphones. They are not able to control user experience. Those three approaches determine the various flavors of mobile VoIP you see out there.

Would you consider FMC providers as part of the mobile VoIP thing?

There is definitely a space for them. But the FMC services rely on hardware based solutions which make it difficult to upgrade in terms of functionality and dynamically change according to the market needs. A software application on the other hand can be upgraded more frequently and allows you to brand and give you so much more control. It is much easier to add value added applications to software based mobile VoIP service.

Do you see companies like yourself adding handover capability in the future?

We can do so. It is not beyond our capability. However the users are still trying to understand difference between GSM call and a VoIP call. It becomes very difficult to communicate with the FMC user why the cost of the call changes, when it changes, and why it changes. It just complicates things a bit more for the service provider.

Of the $391.4 million total revenues in 3Q 2007, communications chip manufacturing accounted for about 50%. Sequentially SMIC had a growth of 9.3% on q-o-q basis. Most of the communications chip revenues were generated from handset chips that it manufactures for chip vendors. Gross margins for the quarter stood at 10.8%. The company posted a net loss of $25.6 million during the period. However the loss was mainly due to severe price decline in the DRAM market which is the other focus area of SMIC.

During the quarter major customers continued to be Broadcom, TI and Qualcomm. SMIC entered into an agreement in 3Q07 with memory silicon manufacturer Qimonda for manufacturing DRAMs exclusively for them. Qimonda is a major supplier of memory chips for communication devices like cellular handsets. The quarter also saw collaboration with Cadence for RF Design Kit for Wireless IC.

Added 30 new service provider customers during the quarter. Acme Packet is now deployed at 450 service provider customers worldwide. That includes 79 (up from 76 end of 2Q07) of the top 100 largest tier ones in the world. Deployed in 85 countries now.

Revenues $29.6 million, increase of 9% sequentially. Net income $5.5 million. 81% gross margins. $127.8 million cash balance as of 3Q07. Raising business outlook for 2007. Revenue forecast for 2007 now $112 million to $113 million. Q4 revenues expected to be between $30.5 to 32m.

Increase in 3Q07 revenues due to higher sales in Europe and increased indirect sales channels. Indirect sales accounted for 64% of the revenue reflecting Acme Packet’s growing strength in Europe. Customers accounting for 10% revenues or more: Nokia Siemens Networks, Alcatel-Lucent, and Ericsson. These three generated 46% of 3Q07 revenues. 40% of revenues came from North America, 60% international. The company sees softness in North American market.

• Estimated number of OneFlex VoBB subscribers: 150k to 190k

• VoIP minutes handled during 3Q07: between 9.5 billion to 10 billion minutes (nearly all internal long distance and carrier wholesale voice traffic). Our estimates are Qwest handled about 28 billion minutes of long distance voice traffic in 3Q07, about 35% of which is VoIP now.

• Access lines lost: 682k consumer, 97k business, and 226k wholesale local voice lines lost from 3Q06 till 3Q07. If a third of those is to VoIP (other two-thirds being shared by mobile and cable), then it seems like Qwest is losing around 80k access lines to VoIP each quarter.

Revenues in the 3Q07 were $4.3 million. In terms of units of chips sold, this quarter saw the highest number in 2007 with estimated shipments of around 160,000. Of these, around 16,000 chips were for base stations for Fixed and Mobile WiMAX while the remaining 144,000 chips went into subscriber stations. The total number of chips shipped for Fixed WiMAX were estimated at 135,000 while around 25,000 chips were for Mobile WiMAX during the quarter.

The total number of customers at the end of the quarter is 26 but the company did sign up many existing customers for other solutions mainly for the Mobile WiMAX chips. During the quarter WiNetworks launched WiN7000 compact Base Station using SQN2130 Mobile WiMAX Base Station ASIC. Telsima India also chose Sequans chips for StarMAX product line. It ordered around 100,000 chips for subscriber station and another 10,000 for base stations to deploy WiMAX with Tier I operators in India. Other selections were by Alvarion for mobile WiMAX chips for Pico Base Stations and SOMA Networks selecting SQN2130 base chip for FlexMAX mobile WiMAX system.

3Q07 VoIP revenues $19.4m, a decline from $21.4m in 2Q07. Veraz stock suffered a massive 44% decline yesterday, the biggest decline among all stocks on Nasdaq on Wednesday. The marginal decline in VoIP revenues was due to volatility in expansion orders. VoIP revenues from existing customers decreased from approx $16m in 2Q07 to $8m in 3Q07.

In 3Q07 more than half of the VoIP revenues came from early stage NGN projects from five new service providers. Early stage NGN projects typically carry lower gross margins than expansion sales. There is usually 2X to 3X sales related to follow on orders within 18 to 24 months for an NGN deal according to Veraz.

70% VoIP ports shipped to wireless providers in 2Q07. “Fairly similar” breakdown for 3Q07. Contract signed with wireless carrier in Brazil which is yet to be announced. Veraz now has 22 wireless customers for packetized wireless trunking, four of whom have been announced. The type of traffic these wireless carriers are shipping over Veraz platform includes switched and unswitched packetized voice, and SS7 backhaul.

China has a huge number of ODMs/OEMs. But none are able to beat the Taiwanese vendors at the game. One major factor in selecting location for outsourcing of chip manufacturing is the proximity to the country/region of consumption i.e. ODMs/OEMs which will ultimately integrate these chips inside various devices. That logic tells us that China should have attracted more chip foundry business and it should have been the likes of SMIC rather than TSMC and UMC leading the market.

TSMC’s 3Q07 revenues touched $2.75 billion while SMIC managed $391 million during the same period. Both chip foundries are among the top 5 in the world.

One explanation that can be offered is the fact that Taiwanese chip foundry business has been in place for over two decades now. In comparison major Chinese chip foundries like SMIC are only about seven years old. Chinese foundries are not fully equipped or don’t have the requisite capacity to serve growing chip markets like Nano Chips which are widely used in communications. If the foundries of the two countries are compared, technology wise there is a gap of around three years, according to the Taiwanese vendors, which is quite substantial.

However I would conclude this post by saying that chip foundries in China are strategically better positioned to serve the large number of domestic ODMs/OEMs there. They are better positioned to buffer and offset the pressures of price declines in chip market.

The search continues .... The prospect we discuss today is Bangalore based Tejas Networks. Since 2000, when Tejas Networks was established, the vendor has generated accumulated business worth $54 million till the end of year 2006. By the end of 2007, that figure is expected to reach around $100 million. Tejas provides next generation SDH as well as switched SDH solutions and has been witnessing YoY growth in the range of 70-90%.

Tejas has so far raised about $59 million in venture funding.

The solution set spans all three areas of optical networking – core, metro and access. However, the primary focus of the company remains the transmission side. Optical networking equipment market in India has two main contenders: ECI and Tejas. Together these two command around 50% of the market. There are some 7 or 8 other vendors which share the other half of this equipment market. Tejas boasts of nearly all the leading Tier 1 operator customers in India, which include MTNL, BSNL and Bharti Airtel.

Sonus’ patience with BT is finally bearing fruition. Sonus was clearly disappointed two years back when its partner Marconi failed to get a piece of the most hyped up NGN tender in history. BT instead went with Ericsson for the softswitching requirements. Now BT is not sure about the softswitch choice any more. It has awarded some of the VoIP access control business to Sonus. Obviously, that is what happens when you don’t buy iLocus reports.

Anyway, most of the control elements, the softswitching, the CSCFs …. that part of the BT 21CN tender was won by Ericsson. BT calls this the i-node. It seems that the i-node functionality has been chopped into pieces and Sonus will be taking care of what it calls the AGCF (Access Gateway Control Function). Bear in mind that access and core VoIP control functions were not split along IMS lines in the original tender. So Ericsson was pretty much the only call agent vendor.

However the bottom line is that BT 21CN project has not made any substantial progress yet. In fact some of the top management personnel looking after BT 21CN project have left BT recently, no doubt a result of slow progress and wrong vendor choices two years back. Marconi’s exclusion in particular must have raised a few eyebrows in the UK political circles back then.

Voice blogging continues to evolve. Still very early days. I am bullish but I think certain things need to be done in order to get this phenomenon going. First of all, we need a completely different voice blog publishing platform (I will revisit this some time later in the year). Movable Type and Wordpress are optimized for textual stuff. Some of the add on applications that could be added to that voice blog publishing platform could be things like transcription application i.e. when you click on the voice file, embedded in the player will be some transcription buttons (controlled by hot keys or footpedal) that let the listener pause, play, rewind etc while making notes. Manual transcription of premium voice content will be inevitable.

By the way, I have actually not seen a good transcription add-on to Windows Media Player and Real Player. I need to make notes when listening to conference calls. But I can't do it properly. Anyway, the idea I wanted to float today is somewhat different and a simple one. If companies like Snapvine could add on a Karaoke client, let me record my song, and then let me upload my song on my web, that would be great. I mean I really would love to try something like that. There are a number of singing talent shows wherever you switch on the TV these days. US, UK, Europe, India, …., not to mention East Asia where people are absolutely crazy about this stuff. Karaoke blogs could help promote new talent. And participants don’t have to travel to TV studios and auditions. You can do it right from your bedroom. And while you are at it, you could have a separate site that aggregates the Karaoke blogs, kind of Yahoo for online Karaoke.

Search engine privacy issues usually surround consumer protection from online ads. I think there are other equally important issues. To start with, as a business user of Internet I don’t want anyone (least my competitors) to know what I have been searching for. If you have a web business idea and you search for the feasibility of the idea, whether there are other companies already working on the idea, whether they have built feature X or not, and so on, ……. you might hesitate putting in that search string because the Big Brother search engine company will get access to your idea and guess the specs for free.

It does not take a PHD to figure out that search engines like Google will have to face some regulatory rulings in this regard some time in the future. Regulators could come up with some encryption algorithm that encodes search strings. These encryption gateways could be installed by the ISPs. Search info collected by Google robots on the other hand will also be encrypted by the same algorithm (with Google or the ISP having no access to the algorithm). With this solution, Google will never come to know what you searched for and yet give you the results of your search. The problem is that we cannot monetize the encryption gateway so much if the algorithm is written by the regulators themselves. But at least we stop the possibility of a Big Brother emerging.

Congratulations with your win at BT 21CN. You were originally competing for business there two years back along with your partner Marconi.

Thanks. This is certainly an important win for the company, and we’re very proud to be an official 21CN vendor. Our strategy since day one has been to associate with all major network operators that have embraced IP as the future of communications networks, and BT is certainly a pioneer on that front.

I have couple of questions about your strategy for competing with the legacy vendors. But let me touch upon another big opportunity, the wireless space. You have had great success with AT&T in wireless. If this part accounts for 10-15% of your business right now, how far do you think that opportunity is going to develop in the future?

It is good to think in terms of a timeline. Major changes in the telecom world typically evolve over 20 years. Today we are eight years into the transition from digital switching to IP. The wireline industry is moving from early adopter phase to larger scale deployments, but even here, there’s still a major opportunity for growth. Wireless is following wireline in the sense that adoption is happening in the core of the network first and then the edge. But keep in mind, the wireless market started about six years later. While I cannot point out a particular year, overtime our wireless revenues will exceed our wireline revenues because wireless is fundamentally larger in terms of the subscriber base.

What drives IP adoption in the core of the wireless networks?

Traffic and subscribers are growing at a rapid pace in the wireless world. Much of the traffic – such as roaming and long distance traffic - for which wireless operators are able to charge is actually becoming free. As a result the traffic is growing. As a wireless operator you should be looking to reduce the costs at the core.

With regard to wireless, are you limiting yourself to core packetization for a while?

Optical Networking chips generated an estimated $264.5 million during 3Q 2007. That represents growth of about 6% q-o-q. PMC-Sierra led with market share of just under 32% which is roughly the same share as in last quarter.

Top 5 vendors in optical networking chip market - PMC-Sierra, Teknovus, Broadlight, Ikanos and Conexant accounted for an estimated 53% of the total revenues during the quarter, which again is somewhat similar to their combined share in 2Q07.

Unit wise the top 5 vendors supplied 57% of the total consumption during the quarter.

Since optical networking deployments of carriers have not yet translated into a mass end-user service, the chipset contribution on CPE side is just around 20%. The rest comes from the chips consumed in core optical networking devices.

• 3Q07 revenue $153.6 million, down from $155.6 million in 2Q07. Wholesale revenue $121.6m (down from $127.8m in 2Q07); Retail revenue $32m (up from $27.8m in 2Q07). iBasis net income for the quarter: $0.2 million.

• Ended 3Q07 with 637 wholesale customers compared to 636 customers at the end of 2Q07. Out of the wholesale customers, 75 are consumer VoIP service provider customers.

• Traffic carried: 3.7 billion minutes, up from 3.5 billion minutes in 2Q07.

• Average revenue per minute 4.20 cents, down from 4.41 cents in 2Q07. Average cost per minute 3.71 cents.

• Cash balance as of September 30, 2007: $58.7 million

• Merger with KPN GCS: Here is the comparison of ‘before’ and ‘after’. iBasis is the legal acquirer, KPN GCS is the ‘accounting acquirer’ since KPN owns majority stake in iBasis now. KPN paid $55 million for 51% in the combined entity.

Revenues for the 3Q 2007 touched $2.7 billion up by around $400 million compared to 2Q 2007. Total shipments for the period were 2.2 million units (all 8” wafers now). Shipments for the quarter were up by around 20%. 66% of the quarter’s production was consumed by fabless/system customers while IDMs consumed the remaining 34%.

Geographically North America accounted for majority of the sales: 74%. Asia followed in sales with 13% of total. Europe accounted for 10% of the sales. Japan contributed 3% of the sales.

Communications segment continues to be major sales contributor segment for TSMC. In 3Q 2007, around 42% of the sales came from this segment. Other segments contributing to the quarter revenues were Computer 32%, Consumer 17%, Memory 5% and others 4%.

PMC-Sierra’s revenues for the quarter were $117.5 up by over 12% compared to Q2. Of the total revenue Optical chips contributed the main chunk with estimated revenues of $85 million which it generated by shipping over 3.62 million such chips in the quarter. Optical chip constituted around 72% of the revenues.

Revenues for the period saw improvement due to activities in wireline business. FTTH and MIPS related micro processor business showed a slight decline this quarter. The other business, enterprise storage, also improved during the quarter.

By region, Asia contributed 67% of the business followed by North America with 20% of the business. Europe brought in 10% of the business.

Metalink had revenues of $3.9 million for 3Q 2007. This was sequentially up by 56% compared to 2Q 2007. All the revenues came from xDSL chips except the $200,000 generated from production sales of WLANPlus chipsets. This was the first quarter which saw the first WLANPlus production shipment. These were shipped to an ODM. The net loss during the period registered was $5.2 million.

Europe and Israel contributed around 57% of the entire quarter revenues while North America brought in around 35%. Asia contributed 8% of the revenues. The 10% customers for the quarter were ECI and ADC.

The Video over 802.11n market started getting traction with some manufacturers in Europe, NAM and Asia starting off volume production of STBs. Another achievement on this side is that some of the TV makers in Japan have started working with Metalink to develop integrated 802.11n Digital TVs. These two items are expected to generate significant volume deployments for Metalink’s 802.11n chipset.

Worldwide cellular chip market revenues for 3Q 2007 touched an estimated $2,063 million. An estimated 141 million chips were shipped during the quarter.

The market was led by Texas Instruments which took market share of around 24%. That lead was followed by Qualcomm, Freescale, Broadcom and STMicrosystems in that order. Together these five generated around 69% of the total revenues generated by cellular chip sales.

Compared to 2Q 2007, revenues and shipments for the cellular chip market were up by 8% q-o-q. Qualcomm was the only chip manufacturer which saw some increase in the revenue shares among the top 5 cellular chip vendors. Perhaps the reason behind this could be the decreasing average margins per chip in the GSM space, and that would not have affected Qualcomm with the vendor being into CDMA.

Cellular Chips 3Q07 section of the upcoming quarterly chip update tracks the shipment of chips that are mainly used in handsets and core network elements in cellular technologies that include GSM, CDMA, and 3G.

• 3Q07 revenues $211 million, up 2% q-o-q. Over 2.5 million lines in service at end 3Q07. Added 78k new lines during 3Q07. ARPU: $28.24 down from $28.38 in 2Q07.

• Took $145 million charges for patent litigation settlements: $11 million 3Q07 royalty to Verizon, the rest $134 million splits into $70 million for Sprint, $33 million for Verizon, $29 million to AT&T, $2 million for other minor settlements.

• Marketing costs: $62m down from $68m in 2Q07. Subscriber acquisition cost reduced to $206 per subscriber. Customer acquisition cost has been around $280 average over last 4 quarters. Vonage is shutting down poor performing marketing channels. TV and online marketing will be the main areas of spending going forward. New TV ads are generating better results according to the company.