Equinix Inc (EQIX) 2006 Q3 法說會逐字稿

Hello and welcome to the Q3 2006 conference call. [OPERATOR INSTRUCTIONS] I would now like to introduce the host of today's conference, Mr. Jason Starr, Director of Investor Relations. Sir, you may begin.

Good afternoon, and welcome to our Q3 2006 results conference call. Before we get started, I would like to remind everyone that some of the statements we will be making today are forward-looking in nature and involve risks and uncertainties. Actual results may vary significantly from those statements and may be effected by the risks we identify in today's press release and those identified in our filings with the SEC, including our form 10-K filed on March 16, 2006 and form 10-K filed on August 14, 2006. Equinix assumes no obligation and does not intend to update forward-looking statements made on this call. With us today are Peter Van Camp, Equinix's Chief Executive Officer; Keith Taylor, Equinix's Chief Financial Officer; and Margie Backaus, Equinix's Chief Business Officer. At this time, I'll turn the call over to Peter.

Thanks, Jason. Welcome, everybody, good to have you on the call today. As you saw in our press release, it's been another great quarter for Equinix while we're carrying a great deal of momentum as we approach the end of the year. Of course, you may have also seen in a separate release that I've made an important decision which I'll talk about in a moment. But first, let's get to the highlights of the quarter.

Total revenue for the company was $73.7 million. This represents a 27% increase over same quarter last year and just under 8% as compared to Q2. Cash gross margins were 60% while we saw $24.9 million in EBITDA in line with our expectations. This result doesn't necessarily reflect on our revenue over performance in the quarter. A key reason for this was $1.2 million of expense for our audit committee's investigation of our stock options practice of which $600,000 was not anticipated in our guidance for this quarter.

Equinix closed 92 new customers in the third quarter while the installed base continued to contribute well over half of our new bookings. Also our Silicon Valley four IBX in Sunnyvale opened in early August. We've begun to install our initial customers there while our third Silicon Valley IBX approaches its bookings capacity.

Also in the quarter we made significant progress in our first Greenfield build in the Washington, D.C. market. And we now have clear visibility to an expected completion in January allowing us to onboard customers in February. We're pleased with our ability to execute here and we already see a strong pipeline for this new IBX.

You may have also recently seen a release highlighting an agreement with the Chicago Mercantile Exchange, which enables our financial exchange customers in Chicago to directly connect with the Merck. Many of our recent Chicago wins are directly related to this.

Then on a more personal note, as you probably saw in a separate release today, I am planning a transition to a new role with Equinix over the course of the next year. As you might guess, this was a tough decision but I believe a very good one for Equinix and my wife and me.

As mentioned in the release, my wife is a three-year breast cancer survivor. We recently had a setback that required additional surgery. And I'm very pleased to say that it went quite well. Yet it's also the type of thing that brings a certain amount of perspective to life and has made us more thoughtful about certain choices on how we spend our time.

So as a result, my plan will be to transition to a role of executive chairman over the course of the next year and once my successor is chosen. This will allow me to continue with the responsibilities of chairman of the board while I'm also planning to stay involved with aspects of the company such as ongoing strategy, the communication task with all of you as our investors and key customer relationships. Our plan is to conduct an external search for a new CEO. And we'll keep you updated as this unfolds.

So now let me give it to Keith, and he'll take you through the specifics of the quarter.

Thanks, Peter, and good afternoon. I'm pleased to provide you our third quarter results along with some insights into our expectations for the next quarter and fiscal year. But let me first start with our quarterly revenues. Our Q3 revenues came in at $73.7 million, an 8% increase over the previous quarter and up over 27% compared to the same quarter last year. Our U.S. revenues were 86% of total revenues.

This strong increase in our revenues line outpaced our expectations and was due to accelerating installations and our recently opened expansions in Chicago and L.A. as well as continued strong results in other key markets. The three expansion IBXs generated $2.2 million of revenue in the quarter, up from $358,000 in the previous quarter. Looking forward, our backlog and installation schedule remained at their highest levels, which will continue to drive strong growth into Q4.

Recurring revenues in the quarter were $69.9 million, a sequential 7.4% increase over the previous quarter and a 29% increase over the same quarter last year. This growth was the result of strong bookings activity over the first nine months of the year.

Looking into Q4 we expect to see sequential recurring revenues increased above the current quarter's level, both in dollar terms and as a growth percentage. Non-recurring revenues for the quarter were $3.8 million derived primarily from installation fees and professional services.

Looking at MRR churn and cabinet churn. In Q3, our cabinet and MRR churn with 3.1%, an increase over the prior quarter yet in line with our guidance and expectations. Looking forward into Q4, we expect cabinet and MRR churn to decrease over current levels, although we will continue to manage the optimization of our IBXs. As a result, looking forward we expect churn to approximate 2 to 3% a quarter.

Moving on to gross profit and margins, the company recognized a gross profit of $24.6 million for the quarter or 33% compared to $23 million in the previous quarter and $17.1 million for the same quarter last year. Our cash gross margins were 60% consistent with the prior quarter. Cash gross margins were 57% same quarter last year.

Our same IBX cash gross margin, which excludes revenues and costs from our three recently opened IBXs and the Greenfield projects, was 63%, including the seasonal - pardon me, including the impact of the seasonal fluctuations in our utilities expense line. Quarter over quarter our utility expense increased $2.3 million to about 15% of revenue. $1.3 million of this increase was related to our existing IBXs.

[On to] utilization rates. At the end of Q3, our operational cabinet capacity was approximately 30,900, which reflects the additional 1,900 cabinets added from our recently opened Silicon Valley four IBX in Sunnyvale. At the end of Q3 with our ending quarter and our weighted average cabinets billing, were approximately 16,200 cabinets or about 52% up from 15,400 and 15,100 respectively from the previous quarter. Breaking down the weighted average utilization on a regional basis, the U.S. was about 49% utilized, and Asia was about 65% utilized.

Looking at revenue per cabinet on a weighted average basis, our average monthly recurring revenue per cabinet increased to $1,475, up 2.3% over the Q2 level of 1,442 and up over 8% compared to last year. This reflects the continued strong pricing trends, benefits from our IBX optimization strategy as well as incremental volume purchases of non-cabinet items such as power, cross connect, the 10 gig ports on our exchange platform.

Moving into Q4, we expect pricing to increase to approximately $1,500 per cabinet. On a regional basis, our weighted average price per cabinet in the U.S. was $1,559, up from $1,543 last quarter. In Asia-Pacific, this result was $1,110 per cabinet, a quarter over quarter increase of 7% as we continue to benefit from increased interconnection activities in our core Asia markets.

SG&A expenses for the quarter, including stock-based compensation expense of $6.2 million, were $26.1 million. Our cash SG&A expenses were $19.1 million for the quarter, a $1.4 million increase over the prior quarter, greater than our expectation and up compared to last year's cash SG&A of $15.1 million.

Cash SG&A was higher than anticipated as our professional fees related to the audit committee stock option investigation were $600,000 greater than expected. Looking into next quarter, we expect our cash SG&A to approximate $18 million, which includes an additional $1 million of costs attributed to the stock option investigation. For the year, we're increasing the estimated costs for the stock option matter to be $2.7 million from our initial $2 million estimate.

Moving on to EBITDA,our EBITDA was $24.9 million for the quarter, in line with our expectation of approximately $25 million. This represented a flow-through rate on incremental revenues of 17%. Effectively the benefit to drive - pardon me, the benefit derived from the higher expected revenues were offset by the increased professional fees noted. We still remain on track to hit the mid-point of our original 2006 EBITDA range despite increased costs incurred during the year, including about $2.7 million of professional fees and about $1.3 million of incremental costs related to our Secaucus expansion effort.

Also during the quarter the company recorded an additional $1.5 million expense related to a 2004 restructuring charge. We continued to review our specific assumptions related to our restructuring charges. And to the extent additional information becomes available or assumptions are changed, we reflect the adjustment to the restructuring charge in the quarter the decision was made.

Specifically in Q3, we entered into an extended sub-lease for the second phase of the original New York metro IBX in Secaucus, which reduced our original charge by $500,000. Yet this was fully offset by a change in assumption related to the seventh floor of our initial L.A. IBX whereby we extended the time it would take to obtain a sub-tenant for this particular property resulting in a $2 million increase to the restructuring charge.

Our net loss for the quarter was $5.2 million or $0.18 cents per share. Excluding stock-based compensation expense and the restructuring charge, we would have had non-gap net income of $3.2 million.

I'd also like to provide a brief update on our expectations for cash, cash expenses over the next two years. Taking into consideration our NOL position, tax deduction benefits attributed to our stock option activity and our active construction projects, we do not expect to pay any meaningful cash tax in either 2007 or 2008.

Turning to our balance sheet, our unrestricted cash balances totaled $166.3 million, including the $40 million drawn under the Silicon Valley Bank line of credit, which has since fully been repaid in October. CapEx was $46.6 million. Breaking the details down, our expansion CapEx for our new centers was $39.6 million, while our ongoing CapEx was $7 million.

Next moving to our operating cash flows, our net cash generated from operating activities was $20.7 million, a $4.5 million improvement over the prior quarter. Our working capital balances ended as anticipated with our accounts receivable balance back to an expected level as our DSOs returned to less than 30 days.

Cash used from investing activities was $49.9 million, which was comprised of $46.6 million of capital expenditures and a $3.3 million reduction in the accrued property and equipment line. Cash generated from financing activities was $47.5 million primarily attributed to the line of credit drawn down of $40 million and $8.2 million of proceeds from our employee stock plans offset by debt repayments of $700,000.

Finally, with respect to our equity balances outstanding, we had approximately 29.2 million shares of common stock outstanding. This number excludes the 2.2 million shares related to our convertible diventures and the 3.6 million shares related to the employee stock plans and other warrants, the majority of which vest over the next two to four years.

Let me turn the call back to Peter.

Thanks, Keith. Having outlined our Q3 results, we've now got pretty clear visibility in the full year of 2006. With this we see continued momentum which has us increasing our annual revenue guidance to exceed the high end of our previous range with a new mid-point of $286.5 million. Of course, as we outlined earlier this year, we expected a strong second half ramp in our growth which this new guidance certainly supports.

As you all understand, in a recurring revenue business, our exit to '06 also creates a solid foundation to build on for another year of strong growth. In fact, as we consider all our efforts over the course of 2006, there's been a great deal accomplished that positions us very well for next year.

First, in thinking about our business execution at a sales and marketing level, we're enjoying our best year ever in results. In our operations teams we've correspondingly onboarded this growth while maintaining greater than five nines reliability across our IBXs. This also includes scaling to manage three new IBX openings, which represents our largest increase in cabinet capacity since we acquired our way into Asia four years ago.

For our peering customers, we successfully rolled out support for 10 gig ports on our exchange while we've applied important strategic focus to key programs such as customer optimization and the right pricing to get customers more aligned with market. Also in 2006 Asia began contributing in a meaningful way to our EBITDA performance.

Our '06 execution in all of these areas provides confidence as we consider an outlook for 2007. Now equally important to this year's success in our '07 outlook has been our expansion execution. The completion of three Equinix centers positioned us for our strong second half growth and will play an important role next year.

Added team resource and focus this year on our expansion strategy have now enabled us to acquire the right properties and secure a long-term platform for incremental future growth in our key markets. And something that doesn't necessarily show up in the numbers is the hard work that's required to scale the company to the size of opportunity in front of us. We've made investments in people and the reengineering of systems and process to strengthen the back end engine for our future growth.

This is challenging stuff. We're putting our employees and customers through a fair amount of change. But this is necessary for our ability to support the service levels and response times our customers deserve. We'll continue to have a high degree of focus on this into 2007. With these pieces in place and again, the strong exit rate expected in the fourth quarter, we are well positioned for next year.

I'll get to the specific guidance in a moment, but I'd like to spend a little more time talking about expansion. We really feel that the expertise we've developed in this area over the past few years has become a strategic differentiator and an advantage as we see a disproportionate share of market growth.

Of course, in 2006 we wrapped up the equinization phase of our expansion, bringing three new centers in on plan in Chicago, L.A. and the Silicon Valley. As you saw in today's press release, we now have a specific expectation for opening our first Greenfield build in the D.C. area. Completion is expected in January with customer installations beginning in February.

With added visibility in the project, I should note we've increased our expectation for the cost of this by $5 million to approximately $60 million. Yet as a first Greenfield effort, the expected delivery of this IBX ahead of our original timeline is a demonstration of our execution and ability to respond to market opportunity.

These are complex projects which begin with our unique and proprietary IBX design followed by the property acquisition, the procurement of critical equipment such as generators now on a 60-week lead time and other elements such as securing long-term power for a dramatically increased customer demand as well as fiber and access to networks. We believe this has become a core competency. And it doesn't end here. Very significant is the strong project management required around these builds, particularly now as we've entered our Greenfield phase, while the operating cash flows we generate and our execution track record has attracted support from leading Capitol sources.

And of course, none of this is worth anything without the strength of our customer base and the ability to generate strong returns from these investments. This whole process really begins with our customers and a linkage to their growth plans. For instance, the customers have already weighed in strongly in the pipeline we built for our D.C. Greenfield, which is outpacing our expectations.

I should note that this strong pipeline and our continued demand analysis indicates further strength in the D.C. market. We're taking a hard look at this right now as D.C. could support further expansion in 2007.

And just another potential expansion note, as we've previously mentioned, Tokyo has been a strong market for us. Based on high utilization levels there, we see a path to expand that would look more like an equinization of an acquired site while achieving our targeted returns.

With our expansion strategy and the execution pieces now in place, let's go to the outlook for the remainder of this year and 2007. We're increasing our expectations for revenue in 2006 to now be $286 million to $287 million, which, of course, places our fourth quarter expectations between $79 and $80 million.

Cash gross margins are expected to be between 60 and 61%. Cash SG&A is expected to be $70 to $71 million. We're tightening our expectations for EBITDA, which is now expected to range between $102 and $103 million. We're also tightening our CapEx guidance to now range between $175 and $180 million, of which $143 to $148 million is for expansion and approximately $32 million is ongoing CapEx.

Within this, our Chicago expansion expenditure will be approximately $50 million, which reflects some timing issues within the project shifting an additional $10 million into 2007. This does not affect our expectations for a Q3 opening. However, this shift is partially offset by the increase in our expectation for our D.C. Greenfield of $5 million. Again, with a pretty good view of 2006, let's move on to outlook for next year.

In 2007, we expect revenues to be in the range of $352 million and $362 million. EBITDA is expected to range from $137 million to $143 million. At the mid-point of revenues and EBITDA, this points to a flow-through of approximately 53% of incremental revenues to the EBITDA line. This guidance also reflects $6 million of net costs in our expansion.

So let's look at capital expenditures in '07. We expect our CapEx to be in the range of $230 million to $245 million. This represents approximately $30 million of ongoing CapEx and $200 million to $215 million in expansion CapEx. This includes the additional move of $10 million for our Chicago build from 2006 into next year while we've now created a range around our announced expansion plan. We feel this is appropriate based on the complexity of these projects and the incremental costs we experienced in our D.C. build.

So as you can see, a strong exit to '06 with the operating pieces in place has positioned us for a strong year ahead. However, as in past years, it really is all about execution as we have a lot of hard work ahead. Three new Greenfield IBXs to open, continued development of the team, systems and processes to scale for the size of our opportunity while maintaining a high level of customer focus. My role will be to continue to lead our execution success in all of these areas to ensure a successful transition to my successor over the next year. I want to ensure everybody that I'm fully engaged in leading this effort.

So that concludes our prepared comments. Operator [inaudible - background noise]

Very well, sir. Our first question comes from Tom Watts with Cowen and Company.

Congratulations on another good quarter, Peter.

Thank you, Tom.

Just a couple of quick questions. One, in terms of your executive search, are you considering internal people as well?

No, obviously gave a lot of thought to this. And between the board and the executive team, myself, of course, we've decided that the best path on this is an external candidate.

Okay. And then also, clearly, you have a very large CapEx program and going into '07 will have a lot of new capacity to be online by year end. As we look out through - I know you've spoken in the past about expansion in '08 and '09 being at much slower levels. Shall we still look for that? And what markets could you see that you might be needing to expand in the '08 timeframe?

Too early to tell on any '08 expansion in terms of specific markets, Tom. But, yes, we are still looking forward at '08 and seeing it being a year of expansion and just incremental chunks, if you will and certainly a meaningful cash flow in '08.

Okay. Thanks very much.

Sure.

Thank you, sir. Our next question comes from Jonathan Schildkraut with Jefferies. You may begin.

Thank you. First I just wanted to say, Peter, we've appreciated your leadership of this company. And our best thoughts are with you and your family during this time.

Thanks very much, Jonathan. And certainly, you're going to see me around for quite a while.

Well, I'm glad to hear that. A couple of questions here. The first question is I just wanted to get an understanding of the audit costs going forward. Should we be thinking there are any audit costs in 2007? Or is your expectation to wrap this up during the course of this year?

Too early to tell. I mean, there's been no change on that front for us and really been little contact with the associated agency. So it wouldn't be right to give you any real forecast on it, Jonathan.

All right, great. Could you run us through some metrics in terms of customers at the end of the period, cross connects, ports on the gigi exchange, a number of 10-gig ports, things like that?

Sure. I'll start at the end. I'm looking for the actual net customer count right now. Keith's getting it. But from a peering standpoint, we're at 389. There were 11 new ones and seven upgrades last quarter. Good progress still on 10-gig as well. And we've even got new customers on the 10-gig platform. And VeriSign and Yahoo being moving to that for the first time. Cross connects are, well, right now approximately 14,500. And the net new customer now, Keith, is --

Is 1,258.

1,258. Great.

Net 92 adds.

All right. Could you give us a little color on what's going on in the market in terms of pricing? The MRR per cabinet definitely was ahead of our expectations for the quarter. And even as we look out into next quarter, based on the kind of cabinet ad run rates, it looks like there's a stronger pricing environment than we had expected, at least coming into the back half of this year. Can you add any color?

Well, I don't think our strategy's really changed with respect to it. I mean, as I've always said, that we're in the install base doing more modest increase as customer term comes due and more planning over time moving the install base. But I think what's starting to show up is new business deals are priced significantly more than what the incremental increases in the install base has been about.

And I've said this at conferences before. But really the sales force is living under a new rate schedule with even authorities being raised in the sales organization for how far they can discount. The bar has been set higher. And this, I think, is showing a big impact in the pricing as we're just onboarding new bookings.

All right, great. Keith, I was wondering if you could - in the past you've pointed to positive EPS for the fourth quarter. Net of the $1 million of audit expenses, I am looking at a positive EPS quarter. But then my expectations are that you go EPS negative again next year due to higher interest expense and increase DNA. Can you give some color around those items that you talked about in the past?

Certainly. And in the end when you referred to sort of the positive net income, clearly we feel we're on a path to achieve positive net income in the quarter. You're absolutely right that the $1 million in incremental costs will have some effect on what we - given our previous position on the quarter, but still we believe we'll be positive net income in the quarter.

As we look into 2007, though, I think one of the points you had a discussion with me at the analyst day. We had two sort of projects sort of on the table. And at that time we still felt that we had a pretty good look at 2007, there was the potential to turn net income - remain net income positive for a period of time. But given the fact that we've added a third project and some debt financing, it's going to put some more pressure on the year to remain - to be net income positive.

Having said that, I think what we're going to do is on the next quarterly call we're going to give you a full strip-down P&L like we've done previously in our quarters on what you should expect for the 2007 timeframe. But it is certainly fair to say there is going to be more depreciation and amortization than anticipated, originally anticipated and more interest expense.

Okay, great. Thank you very much for taking my questions.

Thanks, Jonathan.

Thank you, sir. Our next question comes from Erik Zamkoff with Morgan Joseph. Please begin.

First of all, Peter, I'd like to second my colleague's wishes for you and your family, wish you nothing but continued health and hope everything gets better there. In terms of question, on the competitive environment, there's certainly been renewed interest in expansions out of a number of competitors. However, you shared with us in the past sort of the state of your data center and what it takes to get this up and running. In particular, if generator lead times are still at 60 weeks, how does that impact the ability for competitors to ramp capacity?

Well, I can't see inside of any of their plans, of course. But depending on where they are in the planning stage of this, could have meaningful impact. I think one of the points we were trying to make in this call in terms of the dedicated energy we put around expansion, that engine's in place. And we really do feel it gives us an edge in a [inaudible] market.

And one follow-on, can you give us some color on general growth in the peering exchange, any progress [inaudible] and talk about how growth in video impacts the equation there?

Yes, Erik, this is Margie. Actually, I'll take a stab at that one. Yes, there's no question that video is driving a fair amount, especially as we see 10-gig. So as those file size grows, that's why we see a lot of the upgrades to 10-gig. We are even seeing a fair amount of customers who are joining the exchange for the first time going directly to 10-gig.

And so, that's across video sites such as the [U-tubes] of the world and the content guides as well as a fair amount of that video is flowing through the CDNs. So given most of the CDNs are on 10-gig on our exchange today, that's where we're really seeing the uptick. So that is what is specifically driving 10-gig growth.

I think somebody asked earlier the number of ports. So we have 36 10-gig ports now, which is significantly outpacing what we thought. And that is primarily being driven by video.

On the [sip] side, again I think as we said last quarter, look for real progress there in '07. We've completed the development that we needed to do there in terms of some things we want to measure to push that forward for some of our customers who are already on the platform. But I think you'll see some more significant expansion of that as we go into '07 and it gets budgeted properly in our customers budgets.

Well, I'm glad I can ask a question that brought you into the fold, Margie.

Thank you, Erik. Thank you very much.

It wouldn't be a call without you.

She was pulling on my arm. Thank you, Erik.

Thank you. Our next question comes from Michael Rollins with Citigroup. You may begin.

Hey, Mike.

Michael Rollins Good afternoon. Just a couple quick questions. First, just from a high level perspective, you think about capital structure and the way that you're positioned in the marketplace. Do you ever think about whether you could qualify as a REIT and whether you'd have ever interest in going that route since a majority of your revenue comes from rent?

And the second question I had was just in terms of as you look at the -- what looks to be an acceleration in revenue growth in your guidance, where does that incremental revenue come from. Is it from international and some of the repricing benefits you've had there? Or do you see it more in the domestic business? Thanks.

To our first -- I'll take a shot at the REIT question. Keith may fill in if he has anything to add to it. But certainly we actually heard that question from a number of investors, too, Mike. But just from an overall recognition of what it takes to qualify you for that status, there is just the service streams and different things we do do get in the way of that from all things we've been able to consider.

This is just my high level understanding. I can't give you specific details on it. But it is something that's there and one we've looked at just as we look at any strategic review for issues that face us. But we haven't found a pathway of why that makes sense or even if we would qualify.

And, Mike, I'd just follow-on and say one of the things that I mentioned in sort of my small tax section was the fact that we are an NOL company and we do have the benefit of tax shields for certainly the next two years. And looking forward not knowing exactly where that will be as we look into the turn of the decade, it certainly feels that we've got some momentum where we're going to continue to expand the business and take advantage of the continued growth that we see in the marketplace until we can continue to shield our tax for at least a period of time that we're comfortable with. But like anything, it's a strategic question, as Peter said. And we'll review all strategic matters as appropriate to maximize value for the shareholders.

Mike, I think your other question was just about where our incremental revenues are coming from and some of the growth we've been talking about here. You were asking domestic and international. I will say both. Interesting note that Asia actually had their best bookings quarter ever this quarter.

If you recall from our last call, we had said as a company total bookings, our bookings in Q2 were our best ever. And we came close this quarter, had another good bookings quarter. But it wasn't our best ever. So second to that last quarter. But Asia as a part of bookings this quarter actually was their best. So solid booking continue.

One thing I will note is that in Asia we're getting to a higher utilization level, which is really just making sure that we have a growth path there. We mentioned Tokyo, an opportunity we see, work to be done in getting it done, but an opportunity to expand there. And then even incrementally there's augmentations we could do in other markets like Singapore and Sydney that are more add-ons to the properties we're already in. But we do see some avenues to good head room out of Asia going forward, too. So I think both can contribute to growth going forward.

Michael Rollins Thanks very much.

Thank you, sir. Our next question comes from Mark Kelleher with Canaccord Adams. Please begin.

Thanks. On the three new facilities, are those completely built out? Are all the cabinets included in the total cabinets available? Or can you add more cabinets from those facilities?

You're talking about the ones we opened sort of the mid-point of this year?

Right, the Chicago, the L.A., the Silicon Valley.

Yes, yes.

So those are all fully in the numbers? So how many new cabinets might we expect coming online this quarter?

There are no new ones coming online this quarter. We've had these three openings, so that covered the new capacity coming online.

Okay. And the new customer tickup in the quarter - I assume that's because of the new facilities opening and new customers going into those facilities?

Yes, I think there is some impact there. But back in that very good quarter that Asia had, we saw a lot of new names out of Asia as well contributing to it. But as we've said in the past, you think about it going forward, we are trying to steer new customers towards new facilities, save head room in the current centers for current customer growth.

Okay. And in the past you've said that you're constrained on your power to the amount of about 80%. Is that still a good number?

Yes, yes. I think we've raised it a little bit, 75 to 80. We do a little better in some of our full centers, a little less in others.

All right. And my last question - the turn was up a little bit this quarter. Anything unusual going on there?

No. I think it's more continued optimization as we've talked about that program in the past. And so, nothing significant there.

Okay, great. Nice quarter.

Thank you.

Thank you, sir. Our next question is from Ari Moses with Kaufman Brothers. Please begin.

Hey, good afternoon. A couple things here. First, Keith, the comments - I don't know if it was you or Peter who made the comments about the flow-through implied by next year of 53%, EBITDA flow-through. I know you've talked about it on a kind of longer term goal of 70%. You also mentioned the $6 million net cost of expansion next year.

I was wondering if you could kind of walk through that and give us a sense of when you start rebounding to that 70%. Is that kind of a longer term trend? Is that - once the major expansion of next year kind of fall off, we get back there real quick?

Secondly, if you could walk through the cabinet adds in '07 as far as the timing and quantity and which quarters they're coming on and more specifically, how many from the new builds.

Okay, great, Ari. I think generally speaking our objective as a company prior to our expansion effort was EBITDA flow-through on that incremental revenue dollar of 70%. That still holds true today. And the reason you're seeing it certainly over the last, certainly, '06 and '05 as an example and as we look into '07 being something less than that is any time you're in an expansion phase you're introducing new costs to a business plan without getting the requisite revenue attributable to those costs.

An example of that would be the rent costs. You hire staff in advance. You incur other out-of-pocket costs. You open up an IBX. So until we get to a point where either the size of the organization absorbs the impact of that incremental build - and as Peter alluded to, the build will be smaller going forward - or alternatively you're not expanding as aggressively as we've expanded, we're going to continue to have an EBITDA flow-through less than that 70%. But recognizing we're still going more than 50% of that to the EBITDA line even with the expansion that we're seeing today.

As it relates to expansion, we have roughly - if my math does me right here - just under 5,000 cabinets that are going to come online next year. Certainly, the most significant one - pardon me, the most significant one that we're going to see in the short term is the one that Peter alluded to. It's roughly 1,700 cabinets in the D.C. market. And then in Q3 we're going to introduce the remaining cabinets both in the Chicago market and then the New York metro area.

Okay. A couple of other ones. First, in terms of debt levels, can you point us in the direction of - I know obviously it's some financing activity going on related to the build. As far as the timing, when we see - has that kicked in now as far as the higher interest. When do we - what kind of levels are we looking at for interest for '07? And the other question is kind of a more high level strategic one.

We saw in the quarter or recently some news coming out of myspace, your anchor tenant in L.A. about Fox starting to bring some of its CDN activity in-house, which would suggest to me that there is some more - possibly some more connectivity needs that have to review. I don't know if you can comment on that. Or if you can comment generally as far as whether you're seeing some more activity -- people bringing services in-house and therefore starting to require greater levels of connectivity from you.

Great. So why don't I take the financing one, and then I'll push the myspace question over to Peter? Generally speaking, as you're aware, Ari, we've said that we're comfortable with a debt load of 3 to 4 times EBITDA. And so, we still maintain that position today.

We did announce this recently that we were going to raise an incremental $150 million of which $110 was attributed to the Chicago expansion and $40 million was augmentation for our existing Ashburn facility financing. Neither of those two deals have closed to date. And hence there's no incremental interest to be - that has been charged on those two facilities. We do anticipate that they will close in the not-too-distant future. And so, at that point in time, you'll start to see us draw down against those facilities.

The one thing I will say - and we'll give more clarity as we move into '07. But when we draw down these facilities and we are in a construction phase, part of that interest does get capitalized to the cost of the project. And we'll give you a further update on that as we have more clarity. So let me push the other question over to Peter on myspace.

Okay. In myspace - and a number of you saw this in an announcement that they had done. But they were speaking specifically to really moving more to a peering model in the way they distribute their content. By being in our data centers and being next to all these networks they can readily connect over through cross connects or connect through our exchange switch and immediately get their traffic onto downstream network, which is really what the CDNs do for them across the general broad Internet.

So it was a case of myspace -- and in one sense I think the way they described it was just taking the application in-house. But really they're managing their content distribution at all these Internet properties that sit in our data centers do for multiple sites, putting their content directly on the network. So really they were just outlining a broader peering strategy in the way they accomplished that.

But also certainly all the major CDNs are customers as well and sit in our facilities. And they take advantage of the proximity of networks to allow their customers also in our building often being the case to get onto these downstream networks. And so, they also can get you to networks that aren't across the room like they are in our specific model. So that really gives you the most complete coverage of your content distribution strategy when you're going to a CDN who will get you to the other regions of the Internet. But really myspace's announcement was all about the value of peering an Equinix facility.

All right, great. And just - and echoing the comments, Peter, the best of luck, the best of health.

Thanks a lot.

Thank you. [OPERATOR INSTRUCTIONS] Our next question is from Colby Synesael with Merriman. You may begin, sir.

Sure, thank you. You guys talked about how in some of your older centers you can only get to about 80% utilization because of power constraints. What's the difference in economics when looking to potentially just upgrade those to actually reach the full 100% utilization versus actually going and building a new center, an adjacent center and expanding that out?

It's a great question. Unfortunately I didn't like the answer when I first asked it. Basically once the data center is in place, designed and operating, full of customers, it becomes very difficult to augment a current operating facility. And certainly, construction and changing the design of not just the power feeds across the building to reach all these customers and give them more power, but there's a commensurate amount of cooling required to obviously serve these customers and keep them in the right temperature range.

So that issue just makes it extremely difficult. We've done some minor augmentations in certain sites that we've been able to get a little bit more out of them. But at the end of the day, that is not a strategy that will get you that remaining 20% of space. We're downsizing here in the way we look at the current sites.

That said, as we've now gone to building these new sites -- and as we said before, we're going to increase by more than two x the amount of power and cooling we can get to each cabinet in one of our sites. And so, when we talk about cabinet capacity going forward, in those sites we're really going to talk about a number of sellable cabinets. Because we feel we can get 95% of these buildings utilized. And so, we'll always talk about sellable cabinets.

And I think you asked a little bit about the pricing. As I said, fixing current sites is a challenging task. But I will say, as you see the CapEx numbers we're outlining here, that to get this extra power and cooling, that's really showing up in the cost of building these sites. Where our original footprint when original IBXs were built were somewhere in the $500 per square foot range to build the full site, maybe $600. It depends on the site.

What's happening now is we're seeing costs closer to $800 per square foot, really over that if you think of the co-location floor we're serving customers, it's in the $1,000 per square foot range. So a big increase in the cost of building this out. And it really is the electrical work, the power and the cooling. And a big cost is just labor to build these out. So that's the radical change in really building a data center that we think is rightly designed for what the market demand is now.

Thanks. That's actually very helpful. And the second question - you hear a lot about WAN optimizers these days and actually equipment per service. What gives you guys so much confidence that the technology going forward won't improve to a point where the actual value added going to someone such as an Equinix isn't as strong as it is today? And then that would lead into my next - I guess the second part of that question, which would be what makes you guys so confident that you guys expanding in 2007 and into 2008 is the proper model opposed to just trying to get the free cash flow off of the business that you have today?

Well, just your first question - I think it really stems for customers. And certainly, win optimization and some of the techniques being used now to really look at enterprise networks and make sure that the financial applications get through on time, that you're doing the quality of service and routing techniques to ensure that you're getting the most out of your bandwidth are all helping the process.

But at the end of the day, the economics and direct access to all the networks where end users are if you have any critical mass of scale, that is absolutely going to be the highest performing solution to get there and the economics of being able to directly cost connect versus paying on a per-packet basis - it's just still extremely attractive. And I don't see us upsetting our momentum in that at all, particularly with video and just where all these Web properties are going with high content that they're distributing.

The other question was --

So, Colby, the other question - just as you relate to why not maximizing our cash flows to date, I mean, in the end obviously we're reaching on our existing facilities capacity levels that are fairly high. And as a result, based on the demand that we have in the market - and as you're probably aware, it's a very under-supplied market, or we believe is an under-supplied market today, we have the ability to continue to build out and get the returns that we're going to get, which we anticipate will be preferable to our investor base versus just trying to maximize our existing cash flow today.

And so, from our perspective, we do detailed demand studies per market. Margie Backaus is very active in that. And we look at all aspects of ascertaining what's needed in a given market. And we feel very comfortable that what we're building today will meet the demands of the customers for tomorrow.

Okay. Thanks a lot, Keith.

Thanks.

Thank you. Our next question is from Jonathan Hoopes with ThinkEquity. Please begin.

Thank you. A couple questions. If you mentioned it, I missed it. What was the cabinet churn?

The cabinet churn was 3.1%.

Okay, great. Regarding Asia, is that what we should be modeling as the profit driver here in the near term?

Coming out of Asia?

Yes, profits coming out of Asia, given the expansions in the U.S. I'm just trying to - profitability-wise, where is that coming from mostly in the next four, six quarters.

Just so I can understand maybe the question a little bit better, Jonathan, are you talking about on a cash basis, an EBITDA basis or on a net income basis?

Well, answer any two of those three.

Peter has already - we've already set aside the Asia market. We believe it's roughly a $45 million business today without expansion and generates EBITDA margins in the range of sort of 25 type of EBITDA margins. Clearly, when we look at the U.S. on a blended basis, we're looking at we want to continue to expand the U.S. market and we want to be tracking better EBITDA margins than the Asia market. But Asia only represents 14% of our revenue today and probably a little less than that on an EBITDA basis. And so, from our perspective, recognizing we're going to have some drag from the U.S. market, we believe the majority of the growth is going to come out of the U.S. market.

Jonathan Hoopes. Great. That's super. Thanks for the response there. On the 60-weeks lead time for the generators, how's that lead time trending? And how does that compare with the lead times you've experienced with some of the utilities that you need to do major infrastructure upgrades?

Well, yes, I think this all speaks to the challenge in building data centers and getting them to market. But that is a specific supplier and one that we use and has been key to our design. But it has been trending up. And so, this is just something we've seen out of demand specifically for generators. And frankly, we ordered a number of them just to stay ahead as we knew our build plans would constitute the need for it. So rather than just in time provisioning, we ordered a block of generators ahead of time to ensure we were in good shape there.

Okay. And with regard to the generators, specifically are you - do you feel there is much risk that you'll have unused orders there? Or what kind of penalties there might be if you decide to slow down?

Well, if we did. But I think we announced within our expansion play clearly we don't need those immediately in our New York build. But they're ready and available. So we're well positioned for that. And I think you had another question.

I do. Total number of customers at the end of the quarter. I missed that if you mentioned it.

That number is 1,258.

Okay. And then finally with the strong demand and good prospects for profitability, what's to keep deep-pocketed competitors from ramping up supply? And I'm specifically referring here to Suns Data Center in a box announcement. And I wonder if you could just elaborate on what kind of impact you might think their modular data center approach could have on a competitive landscape.

Well, I think that approach can serve maybe an enterprise or a specific need. But that data center is also an island. And the value of our proposition is the multi-customer environment where the interconnection and exchange between customers is so important to who we are. So that doesn't - that picture doesn't fit our strategy and what we've continued to pursue and how we shaped our customer base and grown.

So clearly, it is supplies coming online in what is still an under-supplied market. But the overlap is not real significant with that announcement.

Great. Thank you very much. I appreciate you taking our questions. And best of luck, Peter.

Thank you very much.

Thank you. Our next question is from Greg Mesniaeff with Needham & Company. Please begin.

Thank you. A question on cabinet pricing - as pricing for cabinets in tier one markets continues to accelerate, are you noticing any of your largest customers looking to shift some of their co-location into tier two and even tier three markets to capture some of the lower price points? And if the answer is yes, is this a strategy that Equinix would consider in its expansion at some point in time?

Well, we're always looking at demand and second tier markets. And that's part of our demand analysis strategy in communication with customers as we look at where they're going over time. But I think it was the last call we talked a lot about who's building data centers and why. And specific to the question, I think everybody is aware that Google is building a fairly large facility up in Oregon as well as Yahoo is building now in Washington state.

And both of these guys because of their large demand and because of the strategic needs that control their destiny from a growth and support standpoint are moving to those type of implementations. And our business model has always been about serving the network strategy of these customers, less about just serving the large raw cooler footprint opportunity.

The other one thing we talked about on the call is REIT. And particularly we mentioned [DRT] as someone building - or have less build. Although they are still building now. But they did acquire a fair amount of footprint as we were also acquiring some distressed assets that they can make available through that large raw co-lo user.

So as we go forward in our strategy we're just very cognizant of this and continue to focus on customers that can get the best use of interconnection and of solving their networking strategy versus raw co-lo footprints. And as I said on the last call, we have turned down some fairly large customers who just had a big raw requirement and big power requirement because the nature of our model just put the higher value on what's going on in our centers. So it's something we're very cognizant of and something we've just seen in our strategy for some time.

Thank you, and best of luck to you.

Thank you.

Thank you. And our next question comes from Daniel Golding with Tier 1 Research. Please begin.

Hi, guys. Great quarter.

Hey, Daniel, thanks.

Hey, a couple of questions for you. First, I notice in the press release that you mention that there were 16,200 cabinets going at the end of the quarter and also [weighted] average of 16,200. I wanted to confirm that because that would be sort of unusual.

Yes, what we're doing there is this is approximate numbers. And in one case we're slightly above, and in one case we're slightly below. But just based on how the [churn] and the cabinet adds came in to play, we had this sort of numerical anomaly where we're very close. And so, that's why we've rounded it to approximately 16,200 in each case.

Okay, great. Another quick question was did you guys mention the number of 10-[gigi] ports you guys currently have installed?

Yes, 36.

Thirty-six outstanding. And one last question, which is a little more comprehensive is that right now you've got kind of two kinds of metro installs. You have a campus in D.C., that sort of approach. And then you're building the approach sort of like you have in Chicago where you have two facilities very close to each other and a third facility much farther away. It's like that in the bay area as well. Do you guys anticipate any problem selling the distant facilities to the peering networks versus enterprises?

Obviously it's much easier to do a peering interconnection when you're right next to the person you want to do it with. Do you guys have sort of a different cost structure for doing distant interconnections? Do you foresee something like a Chicago three facility, the new Chicago facility being more of an enterprise sale and then leaving your existing Chicago facilities more for interconnection? And how would you see that affecting revenue?

Yes, I think, Dan, you're right about that. And I think one of the things that we've spent a fair amount of time on this year is what we call customer optimization. And what that means is making room and ensuring we have the space and power available for our peering customers. Those are very important customers to us, both strategically and from a profitability perspective.

But when we look at a market like Chicago, which is kind of a good example here, the reason we chose this specific site in Chicago is in going to our customer base that's specifically there, they were very clear that they wanted an outside of a downtown alternative for both redundancy purposes and other things. And as you're aware, we've had just a tremendous amount of success with our FX product in Chicago continuing to be driven by the success we're seeing with the CME announcement. But you're right about that.

I mean, our strategy going forward is to kind of take those original IBX centers that have really become kind of the mother ship of peering and really continue to make room for them. Because as those enterprise customers want to expand in those original sites, we're seeing them move in a wholesale fashion over to our expansion. So it's kind of happening in a natural way. But, yes, the general plan is to kind of hold those original ones for peering.

But that said, we do put - we are putting 10-gig switches in all of our new centers that we are opening and making that as easy as possible for customers to facilitate that. And we've got a number of customers who use us in that way today, especially in the Silicon Valley.

Great. Okay, thank you very much. And again, great quarter.

Thanks, Dan.

Thank you, sir. And we have time for one last question, [Mr. Rod Ratliff] from [Sandford Group]. You may begin.

Very quickly, just a quick follow-up and clarification. Keith, the 5,000 cabs coming on next year, you said 1,700. Was that the first quarter coming on and the remainder in the third quarter?

That's correct.

Thank you, sir.

Yes.

Thanks, Rod. I think that's it, Operator. So thanks very much. And we'll see all of you on the next call.

Thank you for participating in today's teleconference call. You may now disconnect.