Cogent Communications Holdings Inc (CCOI) 2009 Q1 法說會逐字稿

Good day, and welcome to the Cogent Communications Group first quarter 2009 earnings conference call. As a reminder, this call is being recorded. It will also be available for replay at www.cogentco.com.

For opening remarks and introductions, I would now like to turn the call over to your host, David Schaeffer, Chairman and Chief Executive Officer of Cogent Communications. Please go ahead, sir.

Thank you, and good morning to everyone, welcome to our first quarter 2009 earnings conference call. I'm Dave Schaeffer, Cogent's Chief Executive Officer. With me on this morning's call is Tad Weed, Chief Financial Officer. We're pleased with our results for this quarter in A challenging economic environment. Each of our competitors in both our Corporate and NetCentric businesses has reported sequential revenue declines while we achieved modest revenue growth for the quarter.

During the quarter, traffic on our network grew by approximately 21%. We're very encouraged by this increase in traffic, and continue to be optimistic about our outlook for the remainder of 2009. During the quarter we significantly expanded our footprint by adding 29 on-net buildings to our network, and we have begin a significant network expansion into Mexico extending all the way to Mexico City.

As we have discussed on previous earnings calls, in June of 2008, we extended a pilot program that offered volume and term discounts to all of our new Cogent NetCentric customers, and if our NetCentric customers increase their total contract value with us. During the quarter this program continued to achieve significant success.

During the quarter on the program, 730 existing NetCentric customers had a remaining contract value of approximately $12 million. These customers increased their contract value by an additional $16 million or 130%. This increase and remaining Cogent contract value increased those contracts to over $28 million. Additionally, as more and more of our customers extend their contract terms and show confidence in Cogent, we have seen our average contract length continue to expand, and in this quarter it in fact increased by about 2.9% quarter-over-quarter.

We continue to invest in the improvement and measured expansion of our sales organization. Our quarterly rev productivity was approximately 3.4 units per Rep full-time equivalent per month, a slight decline from the 3.8 FTE productivity reported in the fourth quarter. We have added over 525 net new customer connections during the quarter. Given the current economic environment, we are judiciously monitoring our customer's credit worthiness and activities. As a result of these efforts, collections from customers in March established a Cogent record.

Our days of sales outstanding at then of the quarter continues to be extremely strong at 32 days. Bad debt expense, in fact declined in the quarter to 1.9% of revenues, down from the approximately 2.1% in the previous quarter. However, these credit monitoring activities did somewhat inhibit our revenue growth as we did in some instances turn off customers more quickly than we have in the past. Throughout this discussion, we will highlight several operational statistics we believe will demonstrate our increasing market share expanding scale and operating leverage of our on-net business.

I will review in greater detail certain operational highlights and our continued expansion plans. Tad will provide additional details on our financial performance. Tad will also walk through and reaffirm our 2009 guidance. Following our prepared remarks, we will open the floor for questions and answers. Now I would like to turn it over to Tad to read our safe harbor language.

Thank you, Dave, and good morning, everyone. This first quarter 2009 earnings report and this earnings conference call discuss Cogent business outlook and contains forward-looking statements within the meaning of Section 27 A, and 21E of the Securities Act. These forward-looking statements are based upon our current intent, belief and expectations.These forward-looking statements and any other statements that may be made on this call that will are not historical facts are subject to a number of risks and uncertainties and actual results may differ materially.

Please refer to our SEC filings for more information on the factors that could cause actual results to differ. You should also be aware that Cogent's expectations do not reflect the potential impact of mergers, acquisitions, other business combinations or financing transactions that may be completed after today. Cogent undertakes no obligation to release publicly any revision to any forward-looking statements made today or otherwise update or supplement statements made on this call. Also during this call, if we use any non-GAAP financial measures, you will find these reconciles to the GAAP measurement in our earnings release on our website at www.cogentco.com.

Turn the call back over to Dave.

Thanks, Tad. Now I would like to go over some highlights of our first quarter results. Hopefully you have had a chance to review our earnings press release. As within previous quarters, our press release includes a number of historical quarterly metrics. These metrics will be added to our website. Hopefully you find this consistent presentation of metrics informative and helpful in understanding the financial results and progress of our business. Our first quarter 2009 revenue of $55.1 million was an increase of 3/10th of 1% sequentially over our fourth quarter 2008 revenue of $54.9 million. While modest, this revenue growth was an excess of all of our competitor's sequential revenue performance, and which in fact they have experienced declines ranging from anywhere from 7.6% sequentially to as modest as a 1% sequential revenue decline. This is in both in our NetCentric and Corporation businesses.

Changes in our foreign currency rates can materially impact our revenues as 30% of our revenues are from international operations. Further reductions in the year on dollar in the first quarter caused an additional quarter-over-quarter revenue reduction of about $200,000, and a year-over-year reduction Q1 2008 to Q1 2009 of approximately $2.6 million. We evaluate our revenue streams both based on product class on-net, off-net, and non-core as well as by customer type. We divide our customers in to two primary groups, NetCentric customers, these are customers who buy large amount of band width from us in data centers and/or Corporate customers who buy bandwidth from us in large multi tenant office buildings. Our corporate customers represent approximately 60% of our total customer count, corporate customers represent approximately 48% of our revenues in the Q1 2009.

This customer base revenue grew by approximately 3.4%, Revenues from our NetCentric customers, declined in the quarter by about 2.5% from the fourth quarter of 2008. The difference in growth rates can be attributed to four primary factors. Impact of foreign exchange on our European and most of our Canadian customers is concentrated in our NetCentric customer bases. Many of our NetCentric customers have elected to enter in to longer-term contracts with Cogent. These increased commitments have contributed to the expansion of average contract length of about 2.9%. However, this increase in term commitment as well as volume commitments does result in a reduction in the per-megabit price that we sell to these customers resulting in a decline sequentially in net centric revenues. In fact, our install base is at about $7.55 as of the end of the first quarter, and our new incremental sales were at about $5.28 in the quarter, showing these customers taking advantage of the lower prices. Overall, NetCentric market continues to grow at a slower than historical rate, however, Cogent continues to outpace our competitors and the general market-gaining share.

And then finally, as we have done in past instances, we have extended special promotional pricing to certain NetCentric customers in order to help stimulate specific business models. In the past we have had a great deal of success by these stimulative exercises, and we believe that the contractual commitments that we have received from customers under these programs will contribute materially to our revenue growth throughout the year. Each of these factors contributed to slower NetCentric growth rates and a reduction in our on-net ARPU.

Now with regard to pricing trends. The pricing for our most widely-sold product which still remains the $1,000 per month month-to-month 100 megabit connection remains constant at $10 a megabit. We do offer discounts related to term for all of our customers, and for our NetCentric customers we give additional discounts above and beyond term for volume. One of Cogent's competitive advantages remains its ability to install orders quickly. Cogent guarantees the installation of on-net services in 17 business days or less. In the first quarter our provisioning team again exceeded these goals averaging approximately 10 day business days and provisioning on-net orders. Our net churn was approximately 3.3% in the first quarter, an increase from the 2.9% that we reported in fourth quarter of 2008. We reported our churn numbers on a gross basis, so customers who remain with Cogent but modify their contract by either extending term or increasing volume are counted in these churn numbers.

During the quarter, we sold more of these change contracts as I had mentioned previously. Our Net churn was approximately 3% in the quarter, up slightly from the 2.5% it was in Q4 of 2008. Even though we did result in substantial growth in that off-net business. Our churn rate have increased in part due to our more stringent monitoring of customer's credit history and our collection status with those customers. We will continue to monitor these customers more closely in this challenging economic environment. Our on-net ARPU declined as an impact of foreign exchange rates as well as this increase in volume and contract lag. On-net ARPU was approximately $1,087 in the fourth quarter of 2008, and $1,025 in the first quarter of 2009, a decline of about 5.7%. Our off-net ARPU continued to increase at the average size of these connections continues to increase.

Off-net ARPU increased by 7.5%, up from $1,012 in the fourth quarter of 2008 to $1,088 in the first quarter of 2009. Substantially all of our European and the majority of our Canadian revenues come from on-net services, and are more impacted by FX whereas our off-net business does not have any foreign exchange impact.

Now traffic on our network continues to increase as we increase sequentially quarter-over-quarter at approximately 21%. This increase was comparable to the increase we saw in the fourth quarter sequentially of 2009 where traffic was up 25%. In these cost-conscious times, we believe that being the low-cost provider, such as Cogent gives us a substantial competitive advantage, and we anticipate continued strong growth in network traffic, and an increase in market share.

Now Tad will cover some additional details from our first quarter results as well as reaffirm and provide details on our 2009 guidance.

Thank you, Dave, and, again, good morning to everyone. I would like to thank and congratulate our entire team on hard work and efforts, again, during the first quarter of '09. I'll begin with providing additional details on the components of revenue. Regarding on-net revenue was 44.3 million for the first quarter, a decrease of 1.1% from the fourth quarter, and an increase of 3.5% from the comparable quarter in '08. About 85% of our new sales in the first quarter were for on-net services. On-net customer connections increased by 3.7% for the quarter, and we ended the quarter with 14,674 on-net customer connections on the network, that's an increase of almost 3,000 from the same amount at the end of the first quarter of '08 or 24% increase in customer connections. Revenues from our off-net business had a healthy increase for the quarter of 7.7%. And non-core revenues increased by 8.7%. Now these revenues represent less than 2% of our total revenues.

On EBITDA and gross margin, the operating leverage of the business continues to result in healthy gross margin and EBITDA margins, our gross margin did decrease by 50 basis points from the quarter from 56.7% to 56.2. The impact of foreign exchange negatively impacted our first quarter EBITDA sequentially by about 100,000. If you have seen our quarterly metrics, you will notice there is occasional lumpiness in gross margin expansion and occasional quarter-to quarter decreases; however, if you look at the long-term trends, they are increasing gross margins, and we expect that long-term trend to continue. Our direct incremental gross margins from on-net revenues continue to be almost 100% with direct incremental on-net EBITDA margins of about 95%. EBITDA as adjusted for the quarter was 13.9 million, that was a decrease of 5.2% from the $14.7 million from the fourth quarter. As with gross margin you will see lumpiness in EBITDA expansion and occasional quarter-to-quarter decreases. This lumpiness can occur due to the timing of the performance of, for example, some of our professional services and the seasonal impact of resetting of payroll taxes which happens in the first quarter. EBITDA margin was 25.2% for the quarter.

Earnings per share, basic and diluted loss per share was $0.19. We had to adopt a new accounting pronouncement during the quarter. That added $0.025 to the loss per share for the quarter. As we mentioned in several earlier calls Cogent and companies with notes similar to ours were required to adopt an accounting change the first of this year. Pronouncement APB 14-1, which is called accounting for convertible debt instruments that may be settled in cash upon conversion. The pronouncement required that the convertible feature of our notes be separated from the notes and then separately valued. This accounting results in an increase in our debt discount, and then the other side or corresponding increase in additional payment capital at the adoption date. The increase in the debt discount, results in additional non-cash interest expense, which results from amortizing this increased discount. The accounting rules further require that the increased discount be amortized to the earliest put date of our notes which is June 2014, instead of the contractual maturity date, which is June 2027. This has no impact on cash flows, but is rather -- just a book accounting for the discount on the notes.

The adoption of APB 14-1 is expected to result in 4.2 million of additional non-cash interest expense for this year, and it did result in an additional 1 million of interest expense in the first quarter of '09. The pronouncement also required a retroactive restatement of 2007 and 2008 operating results. This restatement reduced the recorded gains from our previous debt purchases and increased the reported interest expense. Basic weighted average shares was $42.8 million for the quarter. During the quarter we purchased 130,000 common shares for $730,000 and our plan for '09 anticipates and includes the estimated impact of additional stock purchases. We have over 30 million remaining of authorized spend in our stock buyback program.

Dave touched a little bit on foreign exchange. I'll provide some additional details. About 30% of our business is outside of the United States. For the first quarter 22% of the revenues were in Europe, and 7% in Canada. That's very comparable to the percentages for the last quarter of last year. Continued changes in the Euro and Canadian dollar, US dollar conversion negatively impact the comparable quarterly revenues by about 200,000 but if you look at the first quarter of '08 the difference about 2.6 million. The rates were Euro to US dollar average was $1.32 in the fourth quarter, and reduced to $1.31 in the first quarter. The comparable rate in the first quarter of '08 was $1.50.

The Canadian dollar average for the fourth quarter was $0.83 and dropped to $0.80 for the first quarter of '09, the same rate for the first quarter of '08 was parity at $1 -- Canadian dollar to US dollar. Capital expenditures were $11.7 million for the quarter. And $5.2 million for the fourth quarter. You really should look at those two together and average them out because on a quarterly basis we do have seasonal variations. Typically lowest level of CapEx in the fourth quarter and then a pickup of projects after the holidays and activity picks up in the first quarter.

We added 29 buildings to the network in the quarter. We continue to expect to add another 100-plus building to the network in '09. In the end, we expect our 2009 CapEx rate to be very similar to the rate we had in 2008.

On the balance sheet at the end of the quarter we had cash of 67.2 million, we have about 92 million remaining of our original 200 million face value of convertible notes, reduced from the purchases we made in the fourth quarter of last year. The notes are due in June 2027. They are reported on our balance sheet now net of this large discount at 62.8 million from the adoption of APB14-1.

Capital lease obligations on IRUs were 106 million at the end of March, and 10 million-plus of this is a current liability. As a reminder, these are obligations for dark fiber leases, and are being paid over remaining life on average more than 10 years. Dave has touched on this, but our days sales outstanding for worldwide accounts receivable was excellent at 32 days at the end of the quarter, much better than our target of 40, and our worldwide billing and collection team continues to do a great job on collections and credit monitoring. In this economic environment we'll continue to closely monitor credit standards and customer collections.

Operating cash flow was $12.8 million for the quarter, up from $10.8 million for the fourth quarter, increase related to an improvement in working capital. On guidance for '09, our guidance for '09 is based upon current and expected run rates of our business, and includes the estimated impact of our share program, increases in sales and marketing programs, and current and anticipated network expansion, and current trends that we're experiencing in rev productivity, pricing and network traffic. The estimates are also based on recent foreign exchange rates so any material changes in these rates or the other assumptions will materially impact these forecasts. But we are affirming the guidance we provided last call for '09 which, as a reminder, is revenue expected to be over 250 million, EBITDA is adjusted to grow to between 75 and $80 million, and a loss per share between $0.15 and $0.20. That includes obviously this new accounting pronouncement, which adds $0.10 per share to our loss for '09. And the loss per share assumes about 43 million weighted average shares outstanding.

With that, I will now turn the call back over to Dave.

Thanks, Tad. Take a moment and talk about sales force activity and productivity. We began the first quarter of 2009 with 244 sales reps and ended the quarter with 245. We hired 67 reps in the quarter the same number that we hired in the fourth quarter. We did ask 66 reps to leave the Company in the quarter, an increase from the 46 that left the Company in fourth quarter of 2008. The turnover rate of the sales force was approximately 9% in the first quarter of 2009, slightly down from the 9.3% in the first quarter of 2008.

We do expect some seasonality in our rep churn, and we typically see lower churn numbers in the fourth quarter and higher rep churn in the first quarter due to the seasonal factors and our holiday observances. We began the first quarter of 2009 with 219 full-time equivalents, and ended the quarter with 221 full-time equivalent sales reps. Productivity on an FTE basis for the quarter was 3.4 units per month per rep, a decline, slightly from the 3.8 that we experienced in the fourth quarter of 2008. As a reminder, these productivity rates are derived based on customer installations not upon sold or booked business but rather only upon the actual installation of that business.

We added 29 buildings to our network in the first quarter of 2009, having over 1355 buildings directly connected to the network at the end of the quarter. We expect to add slightly over 100 buildings to our network for full year 2009. Our network scale and size continues to grow. We have over 12,200 miles of metro fiber. We have over 33,700 miles of inter city miles of fiber. We continually evaluate additional fiber routes both in Europe and North America for potential expansion. Today we remain one of the most interconnected networks in the world. Today we are directly connected to over 2,680 other internet networks, approximately 240 of these are settlement-free partners, the remaining 2,440 are in fact Cogent customers who buy connectivity from us. We believe our network has substantial capacity to accommodate our future growth. We are currently utilizing approximately 22% of the lift capacity in our network.

So in summary, the network traffic continues to grow on our network at substantially faster-than-market rates, and faster than the rates of our competitors. We believe that our strategy of being the low-price provider in the market, and the pricing discipline as well as changes that we have implemented continue to reemphasize this position. This program has attracted new customers, increased our average contract term, increased our contractual commitments from customers, including ramps, including volume commitments for customers for future purchases. Our business remains completely internet focused. We do not sell other telecommunications services. Our business is a necessary utility and in these challenging economic times, more necessary than ever. We have a strong balance sheet, especially when compared to others in our industry and continue to generate cash. We are encouraged by the results of our sales initiatives and the increase in traffic that we are experiencing.

Our revenue growth was modest for the quarter, however, it was significantly better than the revenue performance of virtually all of our peer group. Unlike many of the negative headlines that many other companies have been issuing, Cogent has no intention of reducing its workforce, in fact we are continuing to hire sales reps, and grow our sales force and expect that number to continue to grow throughout the year. We do not need additional capital. We in fact are generating free cash and intend to put that cash back to work for investors as well as return it to investors. We are expanding our sales force. We are expanding our on-net footprint. We are expanding our cross-sectional capacity, and we are expanding the number of markets we serve. Our North American expansion into Mexico is significant for Cogent taking us in to cities such as Monterey, Guadalajara, as well as Mexico City. We are not suspending our buyback program, but rather continuing to return value to our shareholders with the excess cash that we continue to generate. With that, I would like to open floor now for questions

Thank you.

The question-and-answer session will be conducted electronically. (Operator Instructions). And we'll take a moment to assemble our queue. We'll take our first question from Jonathan Schildkraut with Jefferies.

Wonderful. Thank you for taking the questions. First I just want to start with some housekeeping items. I know I ask this every quarter, Dave, but can we get one more significant digit on the breakdown between the enterprise and the NetCentric revenues, and also what was the percent of traffic that came out of the enterprise group of your total traffic?

I'll take the traffic one, and I'll give Tad the exact breakdown of the mix of revenue and customers. Traffic is approximately 4% from our corporate base, and about 96% of traffic comes from the NetCentric base.

For the first quarter, revenues corporate was 47.8%, and NetCentric 52.5, and if you base this on connections, corporate was 58.7%, NetCentric 41.3.

Great. So, Dave,one of the things you mentioned was that Cogent's traffic was growing faster than the market, and if we take a look at your year-over-year traffic growth, it looks like, call it 56%%, based on the 21% quarter-over-quarter gain. I was wondering how fast do you think the overall market is growing right now, and additionally what do you think the drivers are of traffic today? Have they changed much over the last six months and maybe what your view is kind of going forward, and then finally, in the past you have given us some traffic trends, and wondering if you could do the same for us today.

Sure, John. First of all, we have seen a level of maturity in the business models that have for the past three to four years driven traffic growth, these being casual video, as well as social networking sites. We are seeing a significant uptick in professional video migrating to the internet, displacing broadcast, cable, and satellite. In fact, some of the promotional pricing that we have put in place, is specifically designed to accelerate this trend, and really help incubate some of these business models, and there are numerous models, and numerous providers. We are not so pre assumptions that we can pick the provider , but we think by stimulating them through lower pricing, we are incenting them to put more of than consent on think internet.

I believe underlying market growth, and this is supported by data measured out of either public exchanges or compiled from competitors who do report data as well as independent third-parties, indicates that the underlying market is growing at an annual rate of about 25 to 30% annually, and we are growing at about twice that pace -- continuing to gain share. I think though drivers of growth are really these video business models that are putting large files on and consumers are using those files for a longer period of time. We are continuing to see strong traffic growth in this quarter in April and in early May. We're going to try to get away from getting monthly data, and try to stick to quarterly. It just seems to be kind of a non-stop question I get almost -- people are starting to push me now for weekly data. So I think we'll continue to report on a month -- on a quarterly basis, but you should expect to see the trends continue that we have demonstrated in the first quarter.

Great. If I may ask just one more question. Dave, you talked about average price per bit and I think last quarter you said it was about $8, this quarter about $7.55. You also gave us a indication as to where you are selling. Do you have any sense as to what percent of your NetCentric base has taken advantage of the newer pricing card? We -- kind of give us a sense as to what is left in the base that might come down? Thank you?

Sure, Tad will give you those exact numbers, but what I can tell you is included in that price is some pricing that is actually below our rate card that was not implemented for competitive reasons. We did not have a single instance in the quarter where we had to match a competitor's price point, but we did offer some pricing below our standard grid of pricing for these new business models, and that is reflected in that pricing that I indicated, and in -- in exchange for those promotional pricing, we got contractual commitments from customers for significant ramp volume, and a reversion back to our standard pricing and a reversion back to our standard pricing. With that I'll turn it over to Tad, and let him give you some of the specifics on contract duration.

Yes, as we mentioned on the call, the weighted average increase in contract term for the entire base went up about another 3% for the quarter. If you look in to some of the details, we had a substantial increase in those customers elected to take longer term, couple of ways you can look at it, but if you look at just contracts, two years plus for install, then the first quarter of '09 was over -- over 25%, and if you look at kind of the install base prior to that, it was about 10%, so you are seeing a substantial increase of customers taking longer-term, so we are getting longer commitments, and as we pointed out that lowers the average price for those customers.

And to the volume piece, yes, which only applies to the NetCentric base, that is a bit of a moving target as, you know, different customers business models change, what we're trying to do is incent them to put a greater percentage of their traffic on Cogent than any other supplier. Virtually all of our NetCentric base does multi-home, meaning they have more than one supplier, but I think our growth rates show that they continue to shift share to us, so we do anticipate that trend of lower price per megabit to continue.

Another way to look at that and to state it is for the first quarter of '09, only about 15% of orders were month-to-month, and if you look at the install base kind of prior to that, buzz about 33%. So new business is essentially all signing up for term and extended term.

We'll take our next question from Frank Louthan with Raymond James.

Thank you. Just had a question on the guidance. Just wanted to see what gives you the confidence there in reaffirming the guidance. It looks like you have to have a pretty significant maybe 20% sequential growth in the top line, and 500 basis point sequential improvement in margins to get to the run rate, sort of the lower of guidance, which implies 16% growth on the top line as compared to your 6% run rate here in the first quarter. Where is the new business coming from that gives you that kind of confidence, you can hit that -- the top line are -- the top line ramp there at least. Thanks.

Sure, Frank, it's Dave. Couple of things, one, we have contractual ramp commitments from customers, these customers have taken a certain amount of bandwidth, and have also contractually obligated themselves to taken creasing amounts over the course of the year. Secondly, we do believe that some of the promotional pricing that we put in place will revert back to more of our standard pricing later in the year. As these were introductory prices to get customers to put more on our network or grow their business models. With respect to the sales force, we do expect to continue to hire people, and we do expect to see continued improvement in the rep productivity. And in terms of margin, we have demonstrated over the past four years, a consistent long-term trend line of that, roughly 100 basis point margin expansion quarter-over-quarter.

As Tad mentioned, there is some seasonality. There is some lumpiness, and while our margins step cliented this quarter, and that has occurred in the past, we usually rebound with substantial margin expansion, and we can achieve those types of margin expansions because of the high incremental leverage in our on-net business, where we do have literally 100% gross margin and 95% incremental EBITDA margins, and these are results that we have demonstrated in the past. Most of our fixed-cost price increases have already occurred through CPI resets in the first quarter, and then generally have a fairly stable cost base for the remainder of the year.

So are these take or pay contracts that you signed and that $5 or -- and change or so, increase mental price per meg, you are saying some of that is introductory. Have you done that in the past, where you have given that sort of pricing for a new months, a this then where does it revert back to closer to $10 a meg or do you average the 7.55?

So we have done that in the past. We have done that for a very prominent casual video site, that has achieved a great deal of success. We have also done it to some business models that don't do as well, but these are take-or-pay contracts, they are contracts where customers are obligated to increase their monthly spend, and then revert back to our standard price grid, which is based on contract term as well as volume, and that grid can take prices from -- anywhere from $10 a meg to as low as 4. And some of the pricing that we did put in place is actually below that $4. Again, not in response to competitors, but rather an attempt to help these businesses either raise capital or stimulate their business model to the point where they could become self-sufficient.

What -- what else could change that -- that might make it more difficult for you to hit that guidance? And what are some of the other areas that we should keep our eyes on? I hear you with the take-or-pay contracts and new business, but what are things that you are going to need to keep your eye on?

Sure. FX remains a concern. Our guidance is based around $1.30 for the Euro and about $0.80 for the Canadian dollar. That remains, you know, something that we have little or no control over. or no control over. Two, we have implemented this tighter credit monitoring policy, where we are actually proactively disconnecting customers more rapidly than we have in the past based on their payment histories, just because we are concerned about the general economy. We do not anticipate any material degradation in general credit quality, but that could have a impact. We look at our sales force. We're comfortable that we are continuing to have an ample supply of very strong candidates. We are trying to manage out those that are less productivity, but we do carefully monitor our rep productivity, and I think all of the initiatives we have put in place over the last year are paying dividends, and are showing, results and our growth in a very difficult environment compared to others.

We also look at ARPU and we look at ARPU, really, two ways. We report on a per-connection basis. We have obviously seen that come down, but we have also seen the connection volumes increase so we have seen that trend on our off-net business where now we have over a year of sequential quarter-over-quarter ARPU expansion, not because the price per megabit is going up -- in fact, it is probably coming down -- but the average connection size is rising. We expect to see those types of trends in our on-net business as we continue to see some acceleration in our NetCentric business. Our corporate business has actually continued to perform quite well in a challenging environment. In fact, our on-net corporate business grew at 3.4% sequentially quarter-over-quarter, and our off-net corporate business actually grew 7.7% quarter-over-quarter. It was really the NetCentric business that underperformed. Part of that was FX. It is heavily concentrated out of the US Part of it was these promotional pricing. A part of it was the customer ramps. We monitor all of these things, and look at them literally on a daily basis, and that's why we feel comfortable with the guidance we're giving.

Okay. Great. Thank you.

Thanks, Frank.

We'll take our next question from David Dixon with FBR.

Thanks. Question on the balance between the on-net and off-net businesses going forward. I was thinking about that in terms of OpEx, and I'm picking up from you that there's a tremendous amount of confidence in being able to reset customers on to higher price points. Could you talk about that confidence in the context of the competitive requirement? Because we are hearing different drivers out there, different levels of competition throughout to what I guess is embedded in that confidence. That would be helpful. And then just a question, (Inaudible) -- off-net and moving in to enter national areas on CapEx.

Sure, Dave. Let me try to take them in reverse order. We are in fact not expanding internationally off-net but doing them on-net, so this is our own fiber we have secured, placing network on that fiber. (Inaudible) So our expansion in to Mexico, for example is based on our securing fiber strands, down through Monterrey, a spur over to Guadalajara, through a couple of cities that quite honestly I would butcher their names if I tried to pronounce, and then terminating in Mexico City. Our expansions in to the Ukraine, Bulgaria, (Inaudible) all are based on on-net, on our own facilities. We do not lease capacity to expand our network, but rather do it in on our own footprint. That is baked in to our capital guidance, which is effectively equivalent to where it was last year. As Tad said, you really do need to look at it on a kind of normalized basis of about 32, $33 million. We did 5.4 in the fourth quarter. About 11 in the first quarter. We generally do a lot less construction in the dead of winter, but, you know, that kind of 8 to $9 million run rate is about the right way to think about our CapEx, and that includes these expansions internationally as well as augments to existing network segments, and finally, the additional buildings in both existing and new markets.

Now to the question about our competitive dynamic, and our ability to reset pricing. I do not believe we will be raising pricing off of our standard grid? Fact, my long term belief is pricing will continue to decline perpetually, and I in fact expect Cogent's price per megabit to continue to decline as customers take longer-term contracts and greater volumes. What I did say, however, is there are some customers who got abnormally low prices below our price grid. That pricing was short-term in nature to allow them to scale their businesses. They then revert back to our standard price grid, and have contractual obligations to then take greater volumes, and even longer terms, so they will then be priced off of our standard grid. As I look at the competitive marketplace on the NetCentric side, we have seen actually less competitive pressure in the past six to eight months than we have seen over the past three or four years as a number of competitive exit the market, a number of those competitors are capital constrained, network constrained, and are repositioning their businesses to sell other products and services. We remain solely optimized and focused on internet, and are absolutely committed to being the lowest-price provider on both the NetCentric and corporate side. But on the NetCentric side, we have not seen the broad set of competitors in the past quarter that we saw maybe two or three quarters ago. On the corporate side we typically complete with incumbent providers who are using copper-based facilities. We have a substantially soup your product at comparable pricing. We have not seen any material price degradation or price increases in that environment, but rather stable pricing for the past couple of years, and we are able to differentiate our product and gain market share in our corporate on-net business because of the superiority of our product.

To your final question of off-net versus on-net and its implications for margin, our corporate business, which is driven both by on-net and off-net, we have seen that business continue to perform very well in a challenging environment because of the value proposition that we offer. We typically sell on-net services first, but many of our corporate customers have ancillary locations that require off-net connectivity. In fact we saw in the quarter off -net revenue, which was entirely corporate grow 7.7%, while on-net grew corporate at 3.2%. So what we were doing is selling someone who had maybe one on-net location, and two off-net locations. Also these off-net services tend to be larger capacity purchases than they were in the past, moving from (Inaudible) to mostly internet based. The imply indications for margin are that for our on-net business we have this high operating leverage on-net business we have this high operating leverage of 100-gross margin, 95 incremental EBITDA margin. Whereas for off-net we do have a cost of goods sold, we have a tail circuit to buy, and that represents approximately a 50% gross margin product. If off-net increased as a percentage of revenue that would not be as a accretive to margins as if our on-net business continues to grow as a faster base. We believe our margin expansion will be driven mostly by the on-net business, and only modestly by the expansion in the off-net business.

That's very helpful thank, thanks, Dave.

Thanks, Dave.

We'll take our next question from James Breen can Thomas Weisel Partners.

Thanks. Could you just talk a little bit where the new business is coming from? Whether it's new customers coming on line relative to your existing customer base? Sort of a year-over-year comps, same-store sales?

Sure, Jim, on the corporate side, most of our business comes from new customers -- about 30% of business comes from existing customers taking existing -- or taking additional services at new locations, much like the example I just gave of needing off-net locations as well as on-net. On the NetCentric side, we have actually seen a greater percentage of our business come from brand new customers or customers that had a very small percentage of their traffic with Cogent increasing their traffic. In this environment, we have seen less venture-backed companies being formed, and less brand-new customers coming in, starting up new business models. We have seen some more established companies that typically had a minimal web presence, making their web presence much more seminal to their business model, you see that from a lot of the traditional media and broadcast companies, and we're seeing significant uptick in bandwidth from our customers.

So I would expect over time, we're going to get a much greater percentage of growth from NetCentric customers adding on as opposed to brand new customers, but today it still tends to be a lot of new customers.

Great, thanks.

We'll take our next question from Colby Synesael with Kaufman Brothers.

Thank you for making my questions. As it relates to the credit policy you had, are you actually going and pushing current customers out of your base, and if so, how far are you in to that process? And as it relates to tightening up your credit policies for new customers, have you finalized that, and do we see a full quarter's reflection of that, or should we see a bigger impact in the second quarter? Thanks.

First of all with regard to new customers, there has really been no change in our policy. For corporate on-net, we do not require credit checks because we rely on the credit due diligence that the building owners due, and are comfortable with that, and have not seen significant churn in that base. So we have made no change. And or off-net customers where we do have an exposure for tail- circuit liability, we have kept our policy the same. We do a full credit check on those customers, and we do a full credit check on our NetCentric customers. So on the in-bound new customers, there has actually been no change in policy. The change in policy that we implemented is the number of days of delinquency that we were willing to allow a customer to go before they were done, and before they were terminated. We have short ended that from what we have historically done to a shorter date. I would prefer not to give the exact date, because we use that as a bit of leave rage tool when we negotiate with customers trying to force their payment. I do not anticipate a further contraction in the number of days of delinquency that we allow. I think we are through the worst of the kind of turmoil in the market, and we think we're at the kind of bottom of credit quality. This was really intended to make sure that our DSOs remain as good as they are. I think virtually every other company in our space has seen their DSOs expand, we continue to operate with 32 days of sales outstanding.

As I mentioned earlier, we saw the percentage of revenue that is attributable to bad debt and a deteriorating macro environment go from 2.1% to 1.9%. Again, everyone else is reporting worse trends, and I do believe the customers are more stressed, but what we're doing is making sure that we're the first ones to get paid, putting them on a shorter leash, and willing to terminate that utility more quickly than we have in the past, but I do not anticipate any further tightening or any changes in subsequent quarters.

And if I could just get one more question in. As it relates to your salespeople. It seems like sales productivity is very much dictated by to market as opposed to sales guys ramping and becoming more productive, and that's because you are turning out a certain portion of them. Are you assuming then, that -- I guess what I'm getting at is what is going to improve the productivity of each sales reps, other than the market itself improving?

We see a continued improvement in productivity based on reps longevity that does plateau at some point in time. We also have a program where reps are internally promoted based on productivity and at some point can migrate in to a manager role, and therefore no longer be quota productive and purely managed, there's some of that transition going on. Some is market driven, and the market is more receptive, the rep does better, and particularly that's true on the NetCentric side, because if a rep sells a customer the easiest sale that they can make subsequently is when that customer calls up and just takes more bandwidth. That's, a very easy sale as opposed to going out and prospecting for new customers. We have seen our rep productivity bounce between 2.7 and as high as 4.5 -- well actually we have been as high as 4.7 with an average of about 4.4 over the past three years. We are slightly below that. We think that we will continue to see improvement, both based on the market and just reps activity levels, but we think that the productivity that we're achieving and our cost of revenue acquisition remains much better than anyone else in the industry.

We'll take our next question from Timothy Horan with Oppenheimer.

You said the volumes are growing in the 30% range. Where do you think that was a year or two ago?

I think last year it was probably comparable. I think two years ago, it was probably more like 70 or 75%, as we were in the inflection of many of these social networking and casual video site business models, but as those models have reached a level of maturity, we have seen a plateauing of growth, and we're starting to see some glimmer of acceleration in market demand because of this transition of more broadcast quality video migrating to the internet, and part of what we're trying to do is identify business models that are able to capture that trend, and help migrate a greater inventory of content to the internet, and price as a tool -- and because we have so idle capacity having to use that. We think it is going to work again, but, again, it's portfolio-model where some are going to make it and some aren't. But I do think we are starting to see some inflection as we have seen our growth rate accelerate, and unfortunately, there's not as great of transparency to full market data as we would like, so we look at things such as public exchange data, we look at publicly reported data by our competitors, and finally we look at some research search institutes to get their data.

I understand you trying to stimulate the market, but maybe you can talk about what percentage of the expenses, IP transport is for these companies, and how much you can really, on your own stimulate it? And the reason I ask, maybe the last two quarters you have almost 50% volume growth, yet your revenues are kind of flattish, just broadly speaking. It doesn't seem like the strategy of trying to incremental revenue with the lower pricing really working out at this point, does it make more sense, to get more stability on pricing, particularly if your peers are not coming down to your price level. And then in that regard it sounds like you are doing some of that, but how much of a dollar impact can that have on the second half of the year, or the next three quarters? Basically I'm just real curious on your outlook here on pricing on the strategy there?

Sure. So -- so there were two -- two kind of different questions. The first one is what percentage of the customers expenses are bandwidth? And it varies greatly, but actually for a number of these broadcast video models, it is by far and away their largest expense. Their largest expense is brand width, generally two to three times as much as then their next largest expense, which is their server infrastructure to support that bandwidth, and that is then bifer indicated in to the data center space they need as well -- the bandwidth generally tends to be two to maybe as three times as much as those other two components. So these business models are very sensitive to pricing, and that's from, you know, talking to a number of different customers with a variety of different business models.

To the question about pricing, since we have introduced our price grid, we have actually held on that grid and initially we had a couple of instances where we had to get promotional pricing to match competition, for but for the past couple of quarters, we haven't seen that. We have idle capacity in the network. Every day that capacity is not sold, it's wasted. It goes away. So we want to try to stimulate these customers, and what I care about is total dollars spent, and what we get from the customer is a commitment to increase volumes substantially, and in some cases, their price per bit will go up, but more importantly, their aggregate commitment is going up, their monthly spend is going up, and as their business models get traction, and I think that's really the matrix to focus on. The price per megabit is -- is a relevant metric in the NetCentric business. It has no relevance in our corporate business where we have seen pricing pressure. But in the NetCentric business, that price will I believe actually fall perpetually it's not a short-term thing, and eventually at some point we'll lower our grid based on technology advances. It would be like trying to ask IBM when they were renting out mainframe space saying what do you expect the price per MIP of computer power to be? It's no longer the right matrix to measure by. And you are right, Tim, we have not seen the type of revenue growth commerce rate with volume growth. Part of that is driven by by the FX, part of it is driven by the shift in contract term, but we do anticipate continuing to gain share and therefore grow our total aggregate NetCentric revenue, but I do not believe as I have said many times before, the aggregate NetCentric market is necessarily growing in dollar terms.

And the repricing of those contracts, how much dollar impact did that have this year? The ones that are going up?

Well, they are going up in aggregate value. Some are going up in price per megabit, some are not. Some are just taking more -- they may go up on a nominal basis based on their volumes today, but because their volume commitment is greater, it is increasing their aggregate monthly spend, and all I would feel comfortable with, rather than disclosing specific contracts is, is it's what gives us confidence in our ability to hit our full-year numbers.

Dave, I mean, you need another $30 million of incremental revenue to hit the lower of the guidance, and $20 million of EBITDA. It seems like you need an 8% sequential revenue growth per quarters. Numbers we haven't seen out of anybody, and numbers we haven't seen out of you guys in a couple of years. Is the confidence because of the vetting of the confidence. Of that $30 million, what -- we just really have no basis to base that on history or on this quarter other than you are saying you going to, see some contracts kind of -- kind of -- that are going to reset. It just seems like a huge stretch.

Well, remember, as you said Cogent has demonstrated sequential revenue growth as high as 11% quarter-over-quarter. It's true we have not done that recently, and we are not guiding to specific quarters going forward, but we do believe, based on the continued strong performance of our corporate business, as well as the growth in our NetCentric business that we have both been trending towards as well as have contractual obligated that we feel comfortable with our ability to hit the $250 million in aggregate top-line revenue, and again, to the EBITDA number, we have demonstrated pretty consistently, you know, incremental blended margins, inclusive of building and footprint expansion of, you know, 68, 69%, which, if the revenue is there, the EBITDA is there, and therefore the cash flow.

Sure. Okay. Thanks, guys.

Okay.

We'll take our next question from Nick Netchvolodoff with Barclays Capital.

Thank for taking the questions.

Hi, Nick.

I just wanted to follow up again on sort of the guidance and the achieve of it. Do you expect some of this uplift in the second and third quarters or will it lean more towards the back of the year? Because you see these -- you see the ramp-up, I guess, coming up, just when do you see it happening specifically during the year?

We will see a modest ramp in the second quarter. We will see a more profound ramp in the third and fourth quarter. Based on what we have implemented. And what our funnel is today, but there is, I think pretty good visibility based on what we're seeing in terms of contract values that have been signed, and commitments.

And you mentioned that you sort of indirectly mentioned the type line just now. How does the pipeline look in terms of additional customers in the second and third quarter?

We anticipate an uptick in sales force productivity in the remainder of the year, reverting back more to the mean. We look at the sales funnel, and see a strong number over 30,000 active prospects both on and off-net being worked by the sales force. That's both corporate and NetCentric. We feel comfortable that both the number of reps as well as their productivity based on what they have in their funnel will be sufficient for us to add new customers and the contractual commitments we have from existing customers, the combinations of the two factors will help us achieve our revenue targets, but as I said, the ramp will be more modest in the second quarter, and it will be steeper in the third and fourth quarters.

And the number of sales reps, that was relatively stable this quarter, do you see yourself getting to close to 3,000 or 290 by the end of the year? I think that may have been a goal

Yes I think --

-- but it is possible or?

It is absolutely possible. We have no shortage of applicants. We have been actively hiring. We always have our most significant turnover in the first quarter -- What we indicated was in 2008 we add about 50 reps for the year. We anticipate our net additions in 2009 to be equivalent. We didn't give an exact target, but that's --

Right.

Between 290 and 300. We feel comfortable we'll get to that number of reps. I didn't see today's unemployment numbers, but my guess is there are still plenty of ant contacts throughout.

And without jeopardizing productivity, I guess.

Well, it's a question of having the management in place. The systems, which we have done a lot of work on, and I think you have seen the results of that, as well as the training that we have put in place, and, again, we tend to be very disciplined about managing out under performers, and we have been criticized by some for being too harsh on our sales force, but it has allowed us to achieve cost of revenue acquisition it has allowed us to achieve cost of revenue acquisition that are substantially -- I mean -- 1/5th, 1/6ths of what others in the industry are spending.

Right.

It may sound cold and callus, but we do need to keep a high degree of operating leverage out of our keep a high degree of operating leverage out of our sales force.

And you talked about getting broadcast quality video on the internet. Is (Inaudible) -- who is is -- who -- what does Loo look for in terms of a bandwidth provide, and are you sort of in the mix for a company like them?

Sure, I can't comment on a specific customer because of NDA. I can tell you that many of the partners in that specific customer have allowed us to mention their names. Matter of fact, I think all of the partners are today Cogent customers, and I think what they are looking for at the end of the day -- back to Tim's question. I hate to say it is price. They have SLA guarantees that are almost equivalent between all of the different potential providers, and they ultimately bake us off based on our ability to serve locations, which Cogent has probably the greatest on-net foot print of any of our competitors, and then finally price. And that's where we tend to win the lion's share of that business. And once customers use that network, they are very comfortable with the quality and continue to be comfortable in giving us more and more of their traffic, and I think we -- I was with a customer recently who was introducing some, MPEG-4, HS 264 high-def quality on their broadcast site, and while that's more of a beta-type rollout, I think we're going to see that become more common on the internet for longer session times.

But who is the kind of customer you are looking for, I guess.

That is a name of a type of customer, but I will not necessarily say they have the winning business model. I hope they do. I think many of our other customers that have competing models do as well.

I really appreciate you answering all of these questions. I just had one more. (Inaudible) What do you think the impact will be on bad debt of that transition or that process?

We have been pleasantly surprised, Nick, with the bad debt experience. I will tell you that for the past three quarters we have done better than what I have had in our plan. While we look at resent experience and use that rate for the plan, I guess continue to be somewhat pessimistic out of my nature and using a higher percentage of revenue for the plan. I would say going forward, I do not expect any material change from the percentage that we have experienced over the past couple of quarters. That would make one exception to that, which I do forecast higher rates for new markets just because of unknowns, but on a blended basis overall on the company I don't expect much material change from, if you look at the past two quarters, you are looking at about 2%.

We'll take a follow-up question from Jonathan Schildkraut with Jefferies.

Great. Thank you. Just a couple of quick ones here. Dave, you gave the average bit price at 755 in the quarter. Do you have a number for the year-ago quarter? And then I was wondering if you might give us a sense as to how short the promotional period is on some of these pricing? Does it variety customer or is it set at three or six months? And then finally, just in terms of the comments through the Q&A session here, I'm wondering if you haven't changed the methodology for providing guidance. In the past you had been very consistent with the way you approached guidance. This time around it sounds like you are take some more factors in to account. Thank you.

Sure, Jonathan. A couple of things. I don't think we have specifically disclosed that number before Q1 of 2008. Sew I'll give you that number, which was $8.63. So I think that was the first time we gave the installed base number.

And that was the first quarter of 2008?

That is correct.

Thank you.

To the guidance question, I think we are trying to use the exact same methodology we have used in the past taking all of the key drivers in the business from rep productivity to pricing. We do look at outlier, either plus or minus and kind of try to throw them out if we think there are special circumstances associated with them, a particular customer that's under financial stress or one that appears to be, you know, getting a lot of visibility and support, so we may factor in to our analysis, but at the end of the day there is not a lot of customer concentration at Cogent. So we really do look at these big broad trends. And so your final questions about the duration of these promotional ramps, they can be anywhere from as short as one month to as long as one year. They are all negotiated on individual-case basis, and they all require a customer to guarantee to increase their aggregate spend with us over the life of that ramp. I mean, that's probably as much case say in general, and then each one is very specific, which I'm with the specific.

Thanks for taking all of the questions.

We'll take our next question from Robert Dezego with Sun Trust Robinson.

Good morning. Two quick questions. One is maybe can you maybe give us a dollar amount on the expected ramp from these contract price ups that you are talking about for some of these promo discounts in the year to get a feel for how much revenue that can add in the back half of the year. And can you walk through your top five customers generically by business models and what these companies do.

Sure, we're very fortunate that we don't have a lot of customer concentration, in fact our concentration is even going down where today our top 25 customers represent 14.3% of revenue, as opposed to last year, that was about 15.1, so we're almost less concentrated. As I look at our top customers, I feel a large well-known video site, casual video. I see a professional video site. I see a large hosting company, and a couple of very large cable operators. That would kind of encompass our top five.

Okay.

And in terms of the ramp, I don't want to be too specific, but I can tell you that the customers that have gotten promotional deals, and have commitments to us represent multiple millions of dollars of monthly reoccurring revenue. Some of that is additional, some of it that is already being billed, but it is single-digit millions, but more than one or two in monthly revenue.

Okay. So the impact for the back half of the year, are we talking just -- I guess a infrastructure of that couple million.

No it could be the majority, it is more than a couple million. It is less than 10 of monthly revenue, but more than two.

Okay.

I'm just trying to be fairly brood.

I hear you thank you very much

Thanks a lot.

At this time we have no further questions. I would like to turn it back over to Mr. Shafer for additional comments.

Again, I would like to thank everyone and appreciate the support. As always if anyone has follow on questions feel free to contact management. Thanks a lot now. Bye-bye.

Thank you for your participation. That does conclude today's conference.