阿卡邁科技 (AKAM) 2005 Q2 法說會逐字稿

Good afternoon.

At this time I would like to welcome everyone to the Akamai second-quarter 2005 earnings conference call.

Mrs. Smith, you may begin your conference.

Thank you Shayla.

Good afternoon, and thank you for joining Akamai's investor conference call to discuss our second-quarter 2005 financial results.

Speaking today on behalf of Akamai will be Paul Sagan, Akamai's President and CEO, and Bob Cobuzzi, Akamai's Chief Financial Officer.

George Conrades, our Executive Chairman will also be available during the question and answer portion of today's call.

This conference call we'll discuss information about Akamai's future expectations, plans, and prospects that constitute forward-looking statements for purposes of the Safe Harbor provisions, under the Private Securities Litigation Reform Act of 1995.

Actual results may differ materially from those indicated by these forward-looking statements, as a result of various important factors including but not limited to, unexpected network or service interruptions that cause loss of revenues, increased expenses or diversion of resources, failure to increase our revenue, retain our significant customers, or keep our expenses consistent with revenues.

Inability to maintain the prices we charge for our services, inability to realize the benefits of our net operating loss carry-forward, inability to effectively integrate Speedera networks into our existing business or realize expected benefits of the acquisition, delay in developing or failure to develop new service offerings or functionalities, and if developed, lack of market acceptance of such service offerings and functionalities, and other factors that are discussed in our annual report on form 10-k, our quarterly reports on form 10-Q, and other documents periodically filed with the SEC.

In addition, any forward-looking statements represents our estimate only as of today, and should not be relied upon as representing our estimates as of any subsequent date.

For we may elect to update forward-looking statements at some point in the future, we specifically disclaim any obligation to do so, even if our estimates change, and therefore, you should not rely on these forward-looking statements as representing our estimates as of any date subsequent to today.

During this call, we will be referring to some non-GAAP financial measures that we believe are helpful to an understanding of our financial results and operations.

These non-GAAP measures are not prepared in accordance with generally accepted accounting principles.

Under the news and publications portion of the investor relations section of our website, we define these non-GAAP terms, and reconcile our non-GAAP financials to promote most directly comparable GAAP financial measures.

Now let me turn the call over to Paul.

Thank you, Sandy, and thank you all for joining us today.

We had a record quarter at Akamai.

Our best in top-line revenue and bottom-line earnings.

Even before the positive effects of the Speedera acquisition that we completed in June.

In the second quarter, on a combined basis with Speedera, Akamai generated revenue of $64.6 million, and normalized net income of $17.1 million, or $0.12 per diluted share.

Net income according to GAAP was $15.9 million, or $0.11 per diluted share.

These result includes 20 days of activity from Speedera networks, following the closing of our acquisition on June 10.

It's important to note that our normalized results exclude the impact from any amortization of acquisition related intangible assets, and equity-related compensation costs.

Our stand-alone results, even before the benefit of the acquisition were terrific, and included top-line revenue growing to $62.1 million, a 3% increase over the prior quarter, and a 22% improvement over the second quarter of last year.

Normalized net income grew to $16.1 million, up 12% over Q1.

We also added 71 net new customers, again, before the benefits of Speedera.

Of course, one of the reasons we're so excited about the Speedera acquisition, is that it brings an additional 305 recurring revenue customers to our platform.

With Speedera, our total customer count at the end of the second quarter was 1,736.

I'm very pleased with how the Speedera integration is progressing, especially in light of our twin goals of retaining the vast majority of Speedera's customers, and their valuable employees.

I'll be back in a few minutes to give you more details on how we see the transition shaping up, and the combined outlook for the rest of the year.

Now let me turn the call over to Bob, so he can provide more detail on our financial results.

Bob?

Thank you Paul.

As Paul just highlighted, we had another great quarter, even before we consolidated results with Speedera.

As I review financial results in detail, I'll provide you with our consolidated financial results for the quarter, which includes 20 days of Speedera activity.

And where appropriate, I'll also provide you with Akamai's stand-alone, or our Organic results for the second quarter.

Our total revenue for Q2 was $64.6 million.

That includes $62.1 million from Akamai, which keeps us on track for our organic growth goal of at least 20% for 2005, and includes $2.5 million from Speedera.

On a consolidated basis, that's up 8% over first-quarter revenue, and up 27% over the same period last year.

Consolidated revenue from international accounts was 21%, up slightly from 20% in the first quarter, and up from 17% in the same period last year.

Reseller on a consolidated basis accounted for 25% of our total revenue in the second quarter, unchanged from the prior quarter, and down slightly from 26% in Q2 of last year.

No customer accounted for 10% or more of our quarterly revenue.

Speedera also did not have any 10% of revenue customers.

As you know, we've had three consecutive quarters of strong network customer adds.

As you'll recall, new customers tend to (--)smaller, and grow as they stay with us.

A Police report, that was a strong new customer performance, average revenue per customer or ARPU on a stand-alone basis, was down only 1% to about $14,700 per month.

Consolidated GAAP gross profit margins were 80% for the quarter, and we expect to generate similar gross margins over the balance of the year.

Consolidated operating expenses for the quarter were $34.7 million, which includes amortization of intangible assets, and stock compensation charges related to the acquisition.

Adjusted EBITDA was $22.7 million, up 35% of revenue on a combined basis.

That's up from our 34% adjusted EBITDA margin in the first quarter of this year.

In the second half of this year, we expect adjusted EBITDA margins to continue to expand, with full-year margins in excess of 35%.

Total depreciation amortization for the second quarter was $4.9 million.

That number includes $500,000 for amortization of intangibles, related to the Speedera acquisition.

Net interest expense for the quarter was $800,000, down slightly from the first quarter.

On a consolidated basis, our GAAP net income for the quarter was $15.9 million, or $0.11 of diluted earnings per share.

Our GAAP net income includes amortization of acquisition related intangible assets of $500,000, and $700,000 of charges for equity-related compensation.

Excluding these expenses on a consolidated basis, our normalized net income was $17.1 million or $0.12 of diluted earnings per share, exceeding first call estimates for the quarter.

Akamai's normalized net income, excluding any impact from Speedera, was $16.1 million.

That's up 12% from $14.3 million in Q1, and up 55% from $10.4 million in the second quarter of last year.

We ended the quarter with $138.8 million, up million shares outstanding.

Of the $12.3 million shares issued in connection with the Speedera acquisition, $10.6 million are outstanding and included the $138.8 million shares.

The other $1.7 million shares were issued in the form of auctions.

Weight average shares for the second quarter was $130.1 million, and diluted shares were $150 million.

We expect diluted shares outstanding for the third quarter to be approximately $160 million shares.

Given, that we'll be including a full quarter of the newly issued shares for the Speedera acquisition.

Now let me review some balance sheet items.

We ended the quarter with $130.7 million of cash, cash equivalents and marketable securities, which includes $4.5 million of cash from the Speedera acquisition.

That's up from $118 million in the first quarter of this year.

On a consolidated basis, we generated $14.7 million of cash from operations during the second quarter, as compared to $18.7 million in the first quarter.

Included in second-quarter cash outlays were some acquisition-related costs, and payments associated with the increased CapEx spending in the first half of the year.

On a year-to-date basis we've generated $33.4 million, as compared to $21.1 million over the same period last year.

We've spent $9.8 million in CapEx in the second quarter.

While we expect to redeploy some of Speedera's network assets onto the Akamai platform next year, we will continue to expand our network this year especially to accommodate the migration of over 300 Speedera customers.

As a result, we are now anticipating CAPEX spending for the year to be around 12% to 13% of revenue.

Today sales outstanding for the quarter on a stand-alone basis were 51 days, up from 49 days in the first quarter.

Before I turn the call back over to Paul, let me take a moment to discuss the valuation allowance on our net operating loss carry forwards.

As we mentioned on our last call, we continue to evaluate the status of our NOL carry-forward.

As we've previously indicated, we currently have $1.4 billion of NOL carry-forwards, which results in a $300 million growth deferred tax asset. against what we have applied a full valuation allowance.

Because of our continued profitability, it is likely that in the second half this year, we will no longer need the valuation allowance.

As a result, we expect to release the valuation allowance, resulting in an immediate, one-time, income tax benefit of approximately $300 million.

This could happen in the third or fourth quarter.

After the release of the valuation allowance, we expect to record a tax provision.

That would be a non-cash charge to earnings of approximately 40% of pretax income.

Due to the non-cash nature of the tax charge and the significant side of our NOL, we are excluding the release of the valuation allowance and the tax charges for our 2005 normalized earnings tax guidance.

We believe that excluding the tax charge from our normalized earnings will give a clearer view into the operating performance of the Company.

As for our second-quarter results, we're on track with our goals for organic growth, and we're very excited about the accretive nature of the Speedera acquisition.

Now let me turn the call back over to Paul.

Thanks Bob.

As just heard, we had a great quarter, capped off by the Speedera acquisition on June 10.

As many of you know, Speedera built a business that was very similar to ours.

It was founded in '99 to provide content delivery services to business web sites.

And since then it's developed a great base of more than 300 accounts worldwide.

Speedera also assembled a very talented team of people, and developed some unique features to build a strong customer loyalty.

Our integration is moving forward very well.

We've notified Speedera employees about their status with Akamai, placing about 80% of them in comparable jobs here.

We're now in the process of migrating Speedera customers under one unified Akamai platform.

And we're optimistic that we will retain the vast majority of Speedera's customers and revenue during the migration, which we expect will take us into early next year.

We're also very excited about extending some of Speedera's innovative customer reporting capabilities to all of our existing Akamai customers.

In short, we continue to believe that this transaction will be the win-win we expected.

Most importantly, we are already seeing the positive impact to our top-line, and expect to see a positive impact to our bottom-line starting in Q3.

We will not be breaking up Speedera's contribution going forward, because we're fully integrating the companies as operating one brand with one set of services on a single operating platform.

Previously, we were forecasting normalized earnings this year of $0.46 per diluted share, and we said we expected the Speedera acquisition would be accretive.

We are now raising normalized earnings guidance by about 10%, and expect that earnings will increase by another $0.04 in 2005 to at least $0.50 per diluted share on a normalized basis.

To be clear, we're defining normalized earnings to exclude the impact from any amortization of acquisition related expense, equity related compensation costs, and any impact from the potential reversal of our NOL's.

For the full year we're now raising revenue guidance to $276 to $280 million.

That will represent another 30% year-over-year increase.

Our full-year projections include an expectations of organic growth of more than 20%, plus Speedera's revenue contributions.

We have anticipated that some Speedera customers will not transfer over to the Akamai platform, and we've,accounted for this possible turn in our projections.

Of course, we're working hard to minimize any such churn.

For the third quarter, we expect revenue to increase to $74 to $77 million.

Earnings per share on a normalized base for the third quarter are expected to be $0.13 to $0.14 per diluted share.

You can track our progress as you always have, by measuring our top-line growth, and our increased profitability in earnings per share on a normalized basis.

ARPU will be adjusted in Q3 as a result of bringing on Speedera's customers which were slightly smaller on average than Akamai's.

So you should expect that ARPU next quarter to decline 2% to 3% from our Q2 level.

But at the same time, we're very pleased with the opportunity to expand our customer base through this acquisition, and offer these clients our comprehensive array of services.

Now let's switch topics.

Another bright spot for us is the launch of our web application accelerator service.

As you recall, we mentioned it on our Q1 conference call with you, and we just began selling the service in Q2.

Web application acceleration gives us the opportunity to reach out to new prospects in new industries, such as manufacturing.

Additionally, it gives us a way to reach deeper into existing accounts.

Let me give you a couple of examples.

One exciting new customer win, for example, comes from the chemical manufacturing sector.

A major U.S. chemical company signed up for our application acceleration services, because it wanted to improve the global performance of its procurement software.

With Akamai's web application accelerator, clients such as this one can see global application performance improvements of 40% to 60%.

Another new client for our services, a well-known company known for its cutting edge audio technologies.

The company relies on a secure portal to offer rich, timely information to its business customers.

With a user name and password, businesses accept specialized technical information and product support, without picking up the phone.

Using Akamai's new service, this customer is able to ensure reliable portal performance to business customers, regardless of whether they access the site from San Francisco or Singapore.

We believe application acceleration is an important, new category, and we're pleased with the early customer interest in our new offerings.

While it's too soon for us to provide specific metrics on this new service, we believe it's a direct extension of our core content delivery capabilities, building on the same platform in internet expertise that we're known for.

So we've got a lot to be excited about at Akamai.

We closed the Speedera acquisition, and integration is well under way.

We believe the launch of our new web application accelerator service is off to a successful start, and we continue to see growing opportunity in our core content delivery business, which would benefit from some benefit macro trends including the expansion of broadband access.

More and more success stories are on online entertainment, and the emergence of new mobile and other consumer entertainment devices, that we think will drive demand for a more robust, interactive online experience.

Akamai excels at helping companies meet this sort of demand.

And I'm confident that we will continue to help our customers take advantage of these exciting internet opportunities.

Now Bob, George, and I will be happy to take your questions.

So operator, the first question please.

Thank you.

Your first question comes from Michael Turits of Prudential Equity Group.

How are you?

Good, how are you?

Good.

A couple of questions.

First, I know you've -- your guys are over 20% on a core basis.

Any chance you could be a little more specific about that organic growth.

If that's changed, if we can define that a little bit?

And also, if you could maybe talk about the integration process in terms of the network, what you're doing in terms of how the networks will be combined, and how the sales forces will be combined.

Sure.

So I'll take those in the reverse order.

We've had a full integration team up and running since we -- on the Akamai side since we just signed a definitive agreement, claiming from our perspective the best way to integrate the companies.

After the deal closed, we were able to work collaboratively with our new colleagues.

And we've a team that's focused on figuring out how to merge the technologies, how to bring the customers onto one platform, where there are any gaps in Akamai offerings that we need to make available to Speedera customers, and what's the best product portfolio that we can bring to the market going forward to keep our existing customers happy, to grow those accounts, and to bring on new accounts.

We'll make a few tough choices about futures that we may not keep, but we'll always do that with a mind toward maximizing customer attention and happiness going forward.

Many of the Speedera employees who are now Akamai employees are fully integrated into Akamai.

So the sales reps that we kept are going through training and now selling one Akamai set of services.

On the operations side, there are people who are now, if you will, straddling both sides, because they have to learn the Akamai system as well as maintain Speedera systems until we have one unified formula.

But for operational efficiency, for robustness, for security, we want to wind up with one platform, with one set of processes and procedures, because we think that's best for our customers and for the Company.

And that will take into next year as we migrate all of the customers over.

In terms of

Go ahead.

building the business, there really is no difference anymore between Akamai and Speedera.

One platform, one set of customers, so it's really hard to track going forward.

Certainly, much more than a few months out, which customers are which.

Because once you're on the platform, our goal will be to renew those contracts, sell more features, and bring more things to new customer that's they didn't buy before.

Frankly, whether they came to us through Speedera or whether they started as an entry-level Akamai customer.

So it's pretty easy for us to track through Q2 Organic growth because those were Akamai customers going forward.

It's going to become very difficult.

Frankly I don't think really very well which is why we won't try to break out those metrics as we go forward.

But we had a very clear picture of how we were growing through the first half of the year, what was in our pipeline going through the second half, our expectation around new offers, and their uptake, existing business and pipeline.

So it gave us confidence to renew our guidance that we felt the business was growing at or ahead of the expectations we had, even before we brought in the new business.

How about on the network side?

You have what, 14,000, 15,000 servers?

How many Speedera has, have -- what are you doing in terms of bringing the two networks together geographically?

Are there overlaps?

Can you tell us a little bit about the technical process of network integration?

Sure.

First, we had a, I think, just over 16,000 servers at the end of the quarter, Akamai servers.

Speedera brings about 1,500 over.

What we need to do is build out the Akamai side, and if you will, decommission what you might see on the Speedera side.

Now that may in cases mean physically moving, and in some cases it simply will be switching over code and control into one operation center.

And so, the network group is looking at whether there's overlap in deployment or contracts, where there was deployment that we'd like to add to the system, and therefore, we might leave it in place, but move it under the Akamai network control.

Or decommission some places where we may have already had coverage that we thought was better or sufficient and wouldn't need.

So that's an evolving process that will take some number of months to complete.

We're confident that we could use most of the hardware though, so there won't be waste overtime.

So we'll redeploy it, whether it physically moves or not.

Also, we'll try to leverage greater buying power because we're now an even larger user of internet resources.

So our network group will be looking overtime to renegotiate contracts either early or as they come up, to try to increase our economies of scale.

Okay, Paul.

Thank you very much.

Thank you Michael.

Thank you.

Your next question comes from Henry Naah of Lehman Brothers.

Hi, Henry.

Hey, how you doing?

Good.

Two questions, if I may.

Net add 71 customers during the quarter.

Nice, strong number there.

So I was wondering if you could give us a little detail of where churn was, and whether you think that spike in the number there was more due to an improvement churn or good customer signings?

And second question in ARPU.

Some people probably point to a decline in ARPU this quarter.

I don't think it's a big deal, but if you could, just kind of help us understand the decline after three quarters of increases.

That would be great.

Thanks.

Sure.

So churn continued its strong trend of trickling down.

It was under 4% in the quarter, which was great.

As you know, our goal is to be low single digits on a quarterly basis, and we're almost to the acceptable range.

A little more work to do there.

But we're very pleased with where that is.

That certainly helps fuel the net adds.

But it was at least as much on the other side, which was bringing on more customers, both traditional customers and some early adopters of the web application accelerator service.

As you've seen, it was up -- net customers were up 40%.

So I've been very pleased with that, and it's not all coming from reducing churn, because there's not that much more to reduce.

It's really I think, strong demand in the market for our services.

And you've seen strong customer adds.

They tend to come on smaller as -- as smaller customers.

That clearly if you will, slows down ARPU, which is why it was off a percent.

Speedera's customers are on average a little smaller than ours.

But frankly I think not so much as I think people were expecting, which is why we normally don't guide you on ARPU, but we did want to give you a reset for Q3, and expect that now that we're combined and running as a complete Company for a whole quarter, you should see ARPU go down 2% to 3% in Q3 over the Q2 number, and then we'll work on it from there.

So I've been very pleased with the trends there and the impact of Speedera on the business.

And one follow-up if I may.

In terms of customer Adds, sounds like you're pretty pleased with the gross adds there.

Wonder if you could give us some commentary in terms of trends, in terms of verticals or applications that you saw customers kind of accepting?

We continue to see strong traction in all the ones you usually think of with Akamai.

Technology and software, M&E, commerce, government, travel, automotive.

What we're seeing now though, with the web accelerator services, is that we are getting early traction in manufacturing, not the vertical you would normally think of as a big user of the web.

But in fact, manufacturing is a heavy user of the web for things like portal adoption and procurement systems, and there those are often global companies with poor global performance.

And what we're seeing is that this service allows us to entree to a market where we haven't been before.

Now it's early, we don't have a lot of experience selling there, but I take it as a very encouraging early sign.

Great.

Thanks, guys.

Thank you.

Thank you.

Your next question comes from Doug Campbell of Spirit Capital[ph].

Hi, Doug.

Hi, Paul.

Congratulations.

I'm just curious if you could put some parameters on the web application, web acceleration business in terms of the components of cost both to your customers and what kind of performance they can expect, and the components of cost to you as the service provider.

And give us an idea of overtime as this builds up, what kind of margins we should be looking for?

Not exactly, but just the directional suggestion on that.

Sure.

Well, it's too early to give a lot of metrics.

Also I'm not eager to expose too many competitive advantages.

But I would say, first of all, it's very compatible with our existing business.

It's built on the same platform, on the same basic network where we've added technology on top of what we had so we can scale it very well and we can offer it economically.

We can also offer it to our customers in the way they think of the value of their business.

Not as bandwidth, but as what's the value of a visit to your site, and therefore, what would you pay to make that visit more valuable to enhance your business?

And we can scale it for customers.

One of the interesting things and important things that we're seeing is that application acceleration is already an established marketplace in the product space.

There are a lot of people who sell appliances that people try to put in data centers to improve application performance.

And what they're finding is, that often doesn't work the way they want it to.

It doesn't scale when you try to deal with scores or hundreds of thousands of partners who may access your site.

And you can't go put boxes into 1,000 partner or prospect locations.

So a service model is ideal, and we come in with that alternative, which is you don't have to buy boxes to maintain them, you can use our service and scale on demand over a period of time.

So I expect to see that it will have economics that are consistent with or better than our services to date over the last five years.

I think it's very consistent with what we're doing.

It's already a large and existing market in the appliance base.

So we're not trying to create demand.

We're actually I think going out and trying to satisfy it with what in many cases a better answer for customers.

I think it's very compatible with where their mindset is.

It often allows us to go into customers where we have a great reference, but go down the hall to a different buyer with a different problem and sell there as well.

And in terms of improvement if you will, what they see, it's very similar to CBN or content delivery, in that they look at it and measure it as performance improvement, reliability and roundtrip time, if you will.

And there we're often seeing 40% to 60% improvement or greater, that can translate into 5 to 10x improvement, say between the US and Asia.

So I think that it has a great selling proposition.

I think customers understand the problem, and they -- I think respond to our internet expertise and see us as not expert who now brings them a unique solution, because by and large, their alternative is to buy somebody else's box, and hope that it works, rather than trying our service and seeing for themselves and then letting us manage it.

A logical extension of what you've been saying is that it should lead to pretty widespread adoption over time, I would think.

Is that something you anticipate?

I hope so.

Not ready to predict it yet, but we are -- we try to target our product roadmap around things that we think have large addressable market opportunity, where we can get large customer adoption.

Clearly our model is one of scale.

And now over 1,700 customers.

I don't know what the number will be for web acceleration over time, but it is something that we think could have wide adoption and allow us to go very deep into major industrial sectors.

Thank you.

Thank you.

Your next question comes from Jeff Van Rhee, of Craig-Hallum.

Great quarter, guys, how you doing?

Good.

Thank you.

A couple of questions.

First, while you were on the application acceleration side of the business, can you talk about these couple of wins that you gave a little color on.

Just specifically, were they competitive?

And in fact, did you have a chance to see sort of a head to head against some of the hardware vendors?

And can you give us any color on some of the comparisons other than the obvious service versus hardware that you may have seen in those competitive deals?

Interestingly, we're not seeing them in quite a large number of trials.

By and large, we find people who seem to be in a lot of pain without a good solution.

And so we're not even seeing it as a tight bake-off at all.

These are people who are say Look, I can't find a way to make this work.

Show me what you can do.

And we've gone in and improved it and started to close some deals.

So I think that if you will, customers understand the pain.

A market expectation has developed that there ought to be a solution, and they're not seeing solutions that work for them, and they recognize the service as a software solution model that we represent.

On -- I guess second question.

On the ARPU side, fantastic new customer caption n umbers.

That does distort the ARPU somewhat.

But if you do some sensitivity on it, it looks as though maybe the ARPU is flat.

Maybe slightly down should the new customer number have come in with historical norms.

With that in mind, as you're out there, can you give us some color on the pricing environment?

Assuming stable new customer adds, so you don't have that drag.

Can you talk to pricing of alternative solutions, pricing of direct competitive solutions.

And your feel for core arpu and the ability to grow that.

Well, you know, we don't normally make ARPU predictions, but beyond the color I gave on Speedera is a one-time reset.

I don't want to get into future numbers, because I don't have that crystal ball.

But let me say the market remains very competitive.

We think we're a stronger Company than ever before, but we always say the challenge from the in-house offering of I'll would build it myself, and the competition of large, managed service providers.

We believe that we have our cost structure in line to be competitive in any profitable deal that we want to win.

And we just assume that the market technology will remain highly competitive.

But I'm very pleased with the ads, and I'm very pleased with the increased usage we see from our customers, and by and large, where the business has been going in the new accounts that we're getting, even though they come in smaller.

That meets with our expectation.

Let me bring George into the conversation for a minute.

He's been in the field a lot, meeting with prospects for the web acceleration product.

And he can give you a little more color on what we're seeing there.

Yes Jeff, you asked the question about what we offer that's more than a managed service, even though that's a very popular aspect today, because people trying to reduce boxes and software and licenses in the data center, and the management of that.

But what we're working on with the application acceleration service, is really the delivery of uncashable pages that you would find on a business-to-business portal, for example.

And that's why manufacturing is a great target industry for us.

It's not like CDN B to C. We're talking B to B. But where people are trying to gain the advantage of reducing call center costs, and encourage self-service of the portal.

And the differentiation of our service beyond being a managed service is that with the appliance folks, they're putting a box at each end, you know, at the origin, and at selected sites around the world.

We don't require that, because we've got computers all over the world.

But we deal with one thing that they cannot and do not deal with, and that is latency and reliability.

In other words, we improve the route speed over the internet, which is the most dramatic component of the performance improvement.

The boxes at each end don't see the internet like we do, you know, mapping over a million hits per second on the platform.

So we're able to route much more effectively at any moment in time than BGP will allow.

And so we dramatically reduce latency.

We improve as a result of that, reliability.

And we also deal with mobile.

Where the end points are mobile.

In today's world, you can't optimize the placement of receiving boxes, if you will, because of the great mobility of the business partners, because a lot of them are moving around with their laptops.

So we think we've got a very effective differentiated service.

Thanks, Jeff.

Thank you.

Your next question comes from Vic Grover of Thomas Weisel Partners.

Hi Vic.

Hey guys, congratulations on the quarter.

Thank you.

Most of the questions have been asked and answered, but I wanted to ask you to maybe talk a little bit about SG&A and R&D.

It looks like you had some good expense controls, although R&D was up.

So I'm just trying to figure out what's the baseline to model off of, and are there any more cuts we can see due to the integration of the companies?

And should R&D flatten off, flatten or even come down a little bit?

Thanks.

Well, I'll let Bob take most of the question, but we are manically focused on controlling costs, and making investments where we see leverage when they pay off.

We talked last year about investing in increased R&D and increased field force, because we saw opportunity for new things.

One of those was web acceleration, although we didn't tell you what it was at the time, but it was in the labs and getting ready to go.

So I think we made the right investments, and we're starting to see those results .

We'll be very focused on controlling costs, but we don't really target cuts, we try to invest wisely and grow into that.

So we're not looking at cuts, per se, but we are looking at operating really efficiently in deploying the resources that we have in the best possible way.

But I'll let Bob comment just a little more deeply.

Yes Vic, I think it's fair to say that operating expenses will continue to grow, but not at the same rate they have.

If you look at operating expenses as they relate to revenue, they're about 51%, 52% this quarter.

They will trend down over the balance of the year.

So even though it's growing, I think we'll get more flow through, or more scale ability. in the model, I think we'll come into being as we look out over the balance of the year.

So we'll continue to make investments as we see fit, but as Paul said, we're certainly managing that quite aggressively across the board.

Okay.

Well, I'll save my bottling stuff for offline.

Thanks a lot.

Thank you.

Your next question comes from Errol Rudman of Rudman Capital Management.

Hi, Errol.

Hi.

I wanted to chat with you about two things.

How do you suggest we measure the productivity of the new salesmen that you added last year and in the first quarter?

How will that show up, and how should we measure it in view of the merger with Speedera?

I think you should just model our productivity as revenue growth, and ARPU, and new customer adds over time We break out the number of reps so that I can shows numbers.

I think the number goes up by about 15 total, 13 direct reps and two channel reps.

So we go up by about 15 to, I think the number is about 95 now, off of about 80 which is where we are -- 93.

Okay.

And my next question is, our arithmetic suggests that you're projecting that you're going to lose about 15% of Speedera's revenues in the second half from what we think to be the annualized rate of Speedera revenues alone in '05.

If that assumption is roughly right, why would it be so conservative?

Why do you expect to lose 15%?

Well, I don't want to, as they say, confirm or deny any rumor.

You need to build your own model as you had.

We're not going to provide specific guidance.

But as you know, we've always been conservative about modeling our business.

We know what business Speedera brought over, we know how much of that, and we talked about this in two prior calls, was either dual-sourced customers, where we and Speedera already had business, or might be business that we wouldn't take for a number of reasons.

So we've modeled that in.

We also assumed there will be some churn from just a transition from someone losing a favorite rep or account manager to maybe going out of business, Or not renewing with us.

To create a model that we think is fair, and frankly, in line with exactly the same rules we use for forecasting future, quarter-in and quarter-out, and come up with a number that we think is about right.

We will work every day to try to do better than that going forward.

And we think that the combination of that plus the organic growth in the business ought to get us to $276 to $280 million for the year, which would be a 30% growth or more off of last year's results that we think would be terrific.

There's always the risk it could be lower, and if it is, we'll know why, and there's always the opportunity to do better, and we'll work really hard to get there.

And my last question relates to Speedera.

You had $2.5 million in revenues for the 20 days.

How much did you take in in expenses in those 20 days?

The operating expense, about $1 million for the 20-day period.

Okay.

And I just want add, my congratulations to you for not only this quarter, but for the strategic moves and excellent management that you've shown, and we thank you for your leadership.

Thanks, Errol.

Appreciate it.

Operator, any more questions?

Yes Sir.

Your next question comes from Jean Orr of Nutmeg Securities.

Okay, we'll make that the last one.

Thank you.

I wondered if you could just talk a little bit more about the competitive environment.

Whether or not this has changed in any way, whether you're seeing any new competition.

I'm sorry.

Which environment?

I lost one word you said there.

The competitive environment.

Oh.

It remains very similar.

Do it yourself remains the primary, big competitor for us.

Many people who still haven't made a decision to outsource because they run a large IT and network infrastructure, they're used to hiring people, buying software and machines, and running these infrastructure, and we have to show them the value of outsourcing to us, and the large managed service providers remain competitive, as well.

So I would say that the environment remains pretty consistent, and that we've done a good job of targeting various industries.

The competition is different by industry, and there are different solutions or potential competitors by industry, whether it's median entertainment or technology, software distribution, manufacturing.

And so we've tailored our solutions to what customers need and what the competitive alternatives would be in those verticals.

And I think we have a pretty good handle on it, and I'm satisfying with the progress that we've been making there.

Could you give us some idea of the amount of operating expense decline that -- not decline but decline in percentage that you would expect to see in the second half?

Or some kind of magnitude there.

I think we've indicated two or three data points.

One is that we expect growth of margins to remain at about 80%.

And we're at 35% EBITDA margins, and we expect those to expand, so that the full year there will be in excess of 35%.

So again, we expect operating expense to turn down from current the levels on a percentage basis up.

I think there's enough points to, without being any more specific than that.

Thank you.

Okay.

Thank you for your questions.

Thank you all for your call and your questions.

We look forward to updating you again in another quarter.

Bye-bye.

Thank you.

This concludes today's conference call.

You may now disconnect.