Jul 17, 2015
Executives
Dave Hafner - Director of IR Kevin Thompson - President and CEO Jason Ream - EVP and CFO
Analysts
John DiFucci - Jefferies Matt Williams - Evercore Steve Ashley - Robert W. Baird Scott Zeller - Needham & Co.
Jim Warren - FBR Capital Markets Greg McDowell - JMP Securities Sanjit Singh - Morgan Stanley Tim Klasell - Northland Securities
Operator
Operator
Good afternoon and welcome to SolarWinds' Second Quarter 2015 Earnings Call. This call is being recorded.
All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer period.
At this time I would like to turn the conference over to Mr. Dave Hafner, Director of IR.
Please go ahead, sir.
Dave Hafner
Thank you, Johnny [ph]. Good afternoon everyone.
Welcome to SolarWinds' second quarter 2015 earnings call. With me today are Kevin Thompson, our President and CEO, and Jason Ream, our Executive Vice President and CFO.
Following prepared remarks from Kevin and Jason, we'll have a brief question-and-answer session. Please note that this call is being simultaneously webcast on our Investor Relations website at ir.solarwinds.com.
Please remember that certain statements made during this call, including those concerning our financial outlook, our expectations regarding growth and profitability, our market opportunities, areas of investment and focus for our business, our sales and marketing efforts and the results of those efforts, and our product plans and releases are forward-looking statements. These statements are subject to a number of risks, uncertainties and assumptions described in our SEC filings, including our Form 10-K for the fiscal year ended December 31st, 2014, which was filed on February 23rd, 2015.
Should any of these risks or uncertainties materialize or should any of our assumptions prove to be incorrect, actual company results could differ materially and adversely from those anticipated in these forward-looking statements. These statements are also based on currently available information and we undertake no duty to update this information except as required by law.
Precautionary statements regarding these forward-looking statements are further described in today's press release. In addition, some of the numbers during this call will be presented on a non-GAAP basis.
Our using calculation of these non-GAAP financial measures are explained in today's press release and a full reconciliation between each non-GAAP measure and its corresponding GAAP measure is provided in the tables accompanying the press release. Each non-GAAP item in our forward-looking financial outlook that we will provide today has not been reconciled to the comparable GAAP outlook item because we cannot reasonably or reliably estimate future adjustments.
Lastly, given the recent volatility in foreign currency exchange rates, we will be discussing certain results on both a reported and constant currency basis. During today's call we will indicate each time the constant currency metric is referenced.
If we do not indicate the basis on which we are reporting a particular metric, you should assume we are referring to the reported figure. A reconciliation of revenue on a constant currency basis to our reported revenue is also provided in the tables accompanying our press release today.
With that, I'll now turn the call over to Kevin.
Kevin Thompson
Thanks, Dave. And thanks everyone joining us for our second quarter 2015 earnings call.
During the second quarter, we had very good performance in a number of areas of our business, including strong license sales growth from U.S. federal, Asia Pacific and our installed base sales teams.
In addition, we had solid subscription bookings growth from our MSP business and our cloud management products. We also drove improved customer retention rate in both maintenance and subscription customers.
In addition, we delivered non-GAAP operating profit above our expectations, and record operating cash flows. However, our performance in the second quarter had a higher level of variability across our business than we have experienced in recent quarters as we saw sales growth in our North American and EMEA never-consistent management product line slow as compared to the solid growth rate we have driven for these products over the four previous quarters, looking back to the second quarter of 2014.
Total revenue for the second quarter of 2015 was a record $119.1 million, reflecting year-over-year growth of 17% on a reported basis, and was $125 million, increasing by 23% on a constant currency basis. Our revenue growth in the second quarter was led by a 26% increase in recurring revenue to $80.5 million on a reported basis and a 34% increase to $85.3 million on a constant currency basis.
This growth in recurring revenue reflects a sequential improvement in customer retention rate, along with strong growth in new subscription sales of our MSP and cloud products. Our license revenue in the second quarter increased by 3% to $38.6 million on a reported basis and by 5% to $39.6 million on a constant currency basis.
Our operating model once again showed great leverage as we posted non-GAAP operating profit at $51.5 million in the second quarter or non-GAAP operating margins of 43%, both of which exceeded our expectations. In addition, we generated a record level of operating cash flows in the second quarter, reaching $55 million or 46% of total revenue, despite over $7 million in non-recurring operating cash payments related to acquisitions in prior quarters and a final payment on an office lease abandonment.
In summary, as I look across the performance of our business in the second quarter, I'm encouraged by the strong performance of our MSP business, our cloud management products, our installed base sales efforts, and the significant sequential improvement in the performance of our Asia Pacific region. However, I am disappointed with our license sales growth in North America and EMEA.
Looking forward to the areas of our business where our performance in the second quarter or exceeded our expectations, we had a very good quarter of annualized subscription bookings growth for our MSP business and cloud management products, delivering year-over-year growth of 47% on a reported basis and 56% on a constant currency basis. As it relates to our MSP business, you may recall that on our first quarter-end call, I mentioned that we had an exciting new release of our core MSP product in N-central 10.0 planned for the second quarter.
Thus far, this release has lived up to its billing, as customers and prospects have given us very positive feedback on this release, and even more importantly, we have seen our average transaction size for N-central rise by approximately 10% in just a few months since the release of N-central 10.0. We believe the MSP market will continue to grow rapidly as more and more small businesses that are light IT approach to their IT infrastructure turn to the MSP to manage most if not all of their internal and external IPSS.
We believe that opportunity exists to create a very large and fast-growing business in the MSP market and plan to increase our focus in investment in this area of our business. The strong performance in sales of subscriptions to our cloud management product was led by sequential growth in bookings of our premium products and the addition of Paper Trail in the second quarter.
We are also beginning to see acceleration in the revenue growth rate of Librato driven by initial changes we have made to create clear messaging on how Librato allows developers and IT pros to manage the performance of a WS-based infrastructure. Global core products in new subscription sales growth in the second quarter was 11% on a constant currency basis.
Breaking this down in a little more detail, year-over-year core network and systems management product license sales growth was 6% on a constant currency basis, while growth in the annualized value of new subscription sales was an impressive 56%. On a geographic basis, our U.S.
federal business continued to perform very well, delivering year-over-year growth in new business sales of 13%. We believe that we have created momentum in the U.S.
federal market over the last four quarters and are well-positioned to deliver a solid growth quarter in the third quarter of 2015 despite a strong growth compare from the third quarter of 2014. Our international business core product in annualized new subscription sales grew by 15% on a constant currency basis in the second quarter, led by a record quarter in new license sales from Asia Pacific and strong performance by our MSP business and cloud products.
This growth was negatively impacted by a slower-than-expected growth in new license sales of our network and systems management products in EMEA. The Americas, which is our largest geography, struggled the most against our expectations in the second quarter.
Total core product in annualized new subscription sales growth for the second quarter of 9% was lower than the range of our expectation. This outcome was driven by weaker-than-expected core product license sales growth as our new subscription sales growth in the Americas for MSP business and cloud management products was strong in the second quarter, growing by 58%.
Our average core product license sales growth in the Americas over the last four quarters has been in the range of 20%. We believe that with solid demand capture execution, which we back at these levels by the end of the year.
Turning to the two questions I know are top of mind for all of those on the call. First, what happened in the second quarter that caused license sales growth of core license products in North America and EMEA and resulting license revenue to be lower than our expectations?
And second, what is being done to address the issues that negatively impacted our license sales performance in the second quarter. To address the first question, we believe there's one primary factor that resulted in lower-than-expected license sales growth of our core license products in North America and EMEA in the second quarter.
After four consecutive quarters in strong demand capture volume growth at a consistent level of overall demand quality across our key core products and geographies, our demand capture engine faltered in the second quarter. To set the context, it is important to remember that most of our sales cycles are between 30 and 45 days.
This means that we typically enter a quarter with 35% to 50% of the pipeline we need to achieve our bookings target for that quarter. The remaining 50% to 65% of pipeline necessary to achieve our bookings target is created and closed within the quarter based on demand that we capture during that same quarter.
In the second quarter we experienced a drop in the overall quality level of the demand we captured for our core Net Man and Sys Man products in North America and EMEA, which was not offset by high enough volume growth and the volume of demand captured in these two geographies to allow us to reach our license booking targets for the quarter. The impact of the drop in quality of demand captured during the second quarter was lower than projected conversion rates to opportunities in the pipeline across a number of our core license products.
As a result, our sales team had a fewer number of opportunities and less dollar value of total pipeline created within the quarter to work with than what was built into our second quarter plan based on our demand capture forecast at the beginning of the quarter. This resulted in our finishing the second quarter being softer than our historical trends and our expectations.
Turning to the second question. What are we doing to address the issues we encountered during the second quarter?
First, it's probably worth noting that we remain focused on a fully digital marketing and selling from the inside approach to attacking the enterprise software market. In doing so, we are continuing to blaze a new path in enterprise software.
We have already successfully used this model to reach almost $500 million in annual revenue run rate. What this means, however, is we must constantly be aware of how our unique go-to market model scale as we grow.
As I've indicated previously, one of our goals is to become the best company in the world of reaching our potential buyers using digital marketing, which includes our website, community sites, social media and mobile. While we feel that we have been making solid progress towards that goal over the last 12 months, we also know we can do much better than we did in the second quarter.
And in general, progress has not been happening at the pace that I would like to see. To that end, we have recently added a new marketing executive to our team.
Jason Marshall, former VP of E-Commerce and Digital Marketing at Cost Plus World Market, has joined us as SVP of Marketing and CMO. Jason brings a wealth of experience at building and running a large and rapidly-growing e-commerce sites where the website was the business, as well as a depth of knowledge in modern digital marketing which I believe will allow us to make more rapid progress toward our goals and address successfully the demand capture issues we encountered during the second quarter.
Jason Marshall and the team are already in the process of making modifications to our digital marketing strategy and are further enhancing how we measure and think about quality and quantity of demand. We believe these changes will quickly have a positive impact on the overall volume and quality of demand we are capturing across all of our core products and specifically will improve the quality and volume growth rate for core license product demand we are capturing in North America and EMEA.
Second, we are in the process of reorganizing our product management and product marketing team for greater focus, attention and investment on areas in our product portfolio where we believe we have large opportunities for high growth. These teams are working to create more compelling content about the problems in our products individually and collectively sold, which our demand capture team can use to get in front of IT pros who are actively searching the web for ways to improve the performance of your infrastructure.
Third, we have made changes in our sales leadership structure and sales management processes over the last several weeks to increase the level of detail management intention to the sales process in the goal of driving higher conversion rates from the current volume and quality level of demand we are capturing. We are committed to delivering higher conversion rates as we move forward, and believe that these changes will have a positive impact fairly quickly.
We are taking this multi-pronged approach to accelerating our demand capture and in-quarter pipeline creation growth as we are focused on growing demand captured and pipeline created for our core license products at a meaningfully higher rate than what is reflected in our outlook for projected new business sales growth for the second half of 2015 which Jason will discuss in his remarks. We intend to accomplish this, are reallocating dollars of investment that were already in our plan for the second half of 2015 into various forms of content creation and demand capture activities.
We believe that it will take approximately 90 to 120 days of consistent improvement in demand capture quality and volume growth to get us back to the level of performance necessary to drive new business growth in the range you've seen us deliver over the last year. In the discussion of our outlook for the second half of 2015, which Jason will cover in his remarks, you will be able to see that we have given ourselves a little additional time to get our additional -- our demand capture engine running at this speed as our outlook assumes we are not back to the levels we have driven for much of the last five quarters until after the end of the fourth quarter.
As I look at the second half of 2015, I'm encouraged by the momentum that we feel we have created and maintained in a number of parts of our business over the last four quarters, including U.S. federal, installed base sales, our MSP business, and our cloud management products.
We expect these areas of our business to continue to grow at a fast pace during the second half of 2015. In addition, we have several core products which have been consistent contributors to growth over the same period, such as SolarWinds Virtualization Manager and SolarWinds Network Performance Monitor.
And we are focused on taking the advantage of these trends as we look forward. Over the first half of 2015, we believe we have also created positive momentum for [inaudible] resource monitor as we saw a strong North American commercial performance in the first quarter for license sales of SolarWinds SRM, install a global performance in the second quarter.
We expect SolarWinds SRM to be a contributor to our global commercial growth in the second half in 2015. I will now turn the call over to Jason Ream who will walk through our second quarter 2015 financial results in more detail and will provide an update to our full year outlook.
Jason Ream
Thank you, Kevin. In my portion of today's call, I'll give some further detail on our revenue for Q2, talk about a few financial highlights from the quarter, and provide the key metrics that we give you every quarter, and we're providing our outlook for Q3 and the remainder of 2015.
First, total revenue for the second quarter was $119.1 million. Within that number, license revenue of $38.6 million was below our expectations for the quarter, as Kevin discussed in his remarks.
However, maintenance revenue had $67.6 million, was actually above the level built into the high end of our outlook, with the out-performance driven in large part by a stronger quarter of renewals worldwide than we had expected. Subscription revenue of $12.9 million was also above what was built at the high end of our outlook range, primarily due to stronger-than-projected MSP customer retention rates but also bolstered by strong cloud management bookings than we have forecasted.
Now, for a few highlights in the quarter. Total revenue of $119.1 million is the record high for us, driven by growth in recurring revenue, both maintenance and subscription.
Year to date, our revenue has grown by 19% on a reported basis versus the first six months of last year, or 25% on a constant currency basis, growth that is comparable to some of the fastest-growing public software companies. Total recurring revenue of $80.5 million represented 68% of total revenue for Q2 and 26% year-over-year growth, or 34% on a constant currency basis.
Subscription revenue led the charge but I want to point out that maintenance revenue accelerated from last quarter, growing year over year by 23% on a constant currency basis. We had a very strong quarter of earnings in the second quarter, delivering non-GAAP operating profit of $51.5 million or 43% margin and non-GAAP EPS of $0.52, all of which were well above expectations.
Lastly, we had exceptional cash flow this quarter, and $55 million of operating cash flow represents a record high, despite $7 million of non-recurring operating cash payments that occurred in the second quarter. This result was driven by high operating profit, a record quarter of collections, and a relatively linear bookings quarter.
I do want to dig in a little on our profit for the quarter and point out that our total spend of approximately $68 million was below what was implied by our original outlook for the quarter. We did under-spend a little relative to budgets as we often do, but the majority of the delta was driven by decreased variable compensation in the form of bonuses and commissions as our compensation plans across every level within the company, not just sales, are heavily tied to performance consistent with the high end of our revenue outlook.
To put it more clearly, apart from lower bonus and commissions, we did not reduce our spending during the quarter as our performance until late in June was at a level that we believe could beat the revenue within or above our outlook range. Now I'll walk through some of the key metrics that we have committed to provide each quarter.
First, core commercial license transaction volume grew by 3% in Q2 on a year-over-year basis. The average size for those core commercial license transactions was approximately $8,200, approximately flat with Q2 of last year.
Total revenue from network management products was $67 million, reflecting 11% growth from Q2 of last year, or 16% on a constant currency basis. Within network management, our tools products grew slower than the total, although they did increase in revenue year over year.
Total revenue from systems management products was $36 million, reflecting 14% growth from Q2 of last year, or 18% on a constant currency basis. Within systems management, core products grew faster than the total and revenue from systems management tools was approximately flat year over year, which is an improvement compared to the first quarter of 2015.
Total revenue from MSP and cloud products was $16 million, reflecting 72% growth from Q2 of last year or 84% on a constant currency basis. Sales by our installed base team grew year over year by 41%.
Substantially all of the growth year over year was driven by our installed base teams in North America where we have the most scale and experience with this selling motion. We are focused, however, on driving much stronger growth in installed base sales in EMEA and Asia Pacific over the second half of 2015 and have made some organizational changes in our international sales leadership team to accomplish it.
And lastly, 26% of our revenue in Q2 came from our international subsidiary. Our percentage of revenue from international was slightly up from last quarter but down from 28% in Q2 of last year.
We do believe we are making more progress internationally than these metrics imply as this result was impacted significantly by foreign currency exchange rates. To highlight that point, if we look at our results on a constant currency basis, international revenue would have been approximately 3% higher as a percentage of total revenue.
Turning to the balance sheet, we ended the quarter with approximately $256 million of cash and investments, with approximately 56% of that balance held in the U.S. Our strong operating cash flow in the second quarter was partially offset by the acquisition of Paper Trail for $41 million in cash.
Despite this, we increased our cash balance by approximately $5 million during the quarter, reflecting the cash generating power of our business model. That cash balance, along with access to our available line of credit and other financing alternatives, provides us with additional degree of flexibility to continue the pursuit of our hybrid cloud management strategy.
Turning to our financial outlook for the third quarter and full year of 2015, I'm going to provide a little background color before I give the full details on our outlook. First, given our review of the metrics behind our performance in Q2, we are adjusting our expectations for new business sales during the remainder of the year.
We expect our U.S. federal business to have a solid growth quarter in the third quarter of 2015 as it is the end of the federal government's fiscal year.
Our pipelines are strong, we are well-positioned to continue to deliver the strong growth we have seen from our U.S. federal business over the last four quarter.
We also expect that the strong retention we saw in Q2's renewal bookings is sustainable and will drive maintenance revenue for the remainder of the year. Lastly, we believe that our subscription businesses, both MSP and cloud, should continue to perform well and grow quickly based on our second quarter performance.
But until we have fully addressed the demand capture challenges we faced in our license business in Q2, we are moderating the expectations that we are building into our outlook for the remainder of the year. Second, while the euro to USD exchange rate has been a little bit more stable for the past few months, we still expect difficult year-over-year compares and a potentially unfavorable environment.
So we'll continue to provide our outlook based on a specific euro to U.S. dollar exchange rate assumption.
As of now we are assuming a euro to U.S. dollar exchange rate of $1.10, consistent with what we experienced on average in this past quarter.
Now for the details on outlook. For the third quarter, we currently expect our reported revenue to be approximately $130 million to $134 million, reflecting growth of 15% to 19% over Q3 of last year.
As I mentioned above, in our outlook we are assuming a euro to U.S. dollar exchange rate of $1.10, which compares to an average of approximately $1.33 in Q3 of last year.
If we were to assume a euro to USD exchange rate of $1.33, consistent with the third quarter of last year, our expectations for revenue in Q3 of this year would be approximately $135 million to $138 million, or 19% to 23% growth. As you think about individual components of your model, I want to remind you that we have historically experienced a jump in license revenues from Q2 to Q3 as a result of spending by the U.S.
federal government at the end of their fiscal year, and another sequential increase from Q3 to Q4 with the end of the commercial fiscal year. When you model maintenance revenue, please note that the sequential quarter-over-quarter increase in maintenance revenue has historically been much larger in Q3 than in other quarters.
This has been primarily due to catch-up revenue on late renewals from U.S. federal customers, contributing revenue to Q3 instead of Q2.
This year, however, our federal renewals have been more generally on time with revenue hitting in the quarter that the renewal was due. As such, we currently expect the sequential increase in Q3 to be more similar to the increase we saw in Q2 than it has been in past years.
From a spending perspective, we continue to invest in product development, focusing on improving the usability and integration of our products, beginning the development of a number of new products for core net man and sys man markets, as well as creating the connection between our on-premise and cloud management technologies. With that spending plan, we expect to generate non-GAAP operating margins in Q3 of 2015 of approximately 41% to 42%, which is an increase in non-GAAP operating margin compared to what was built into the outlook we provided in April on our last call.
Lastly, from the third quarter of 2015, we expect our non-GAAP tax rate to be approximately 26% to 27% and to have approximately 77.75 million weighted average diluted shares outstanding, driving non-GAAP earnings per share of $0.49 to $0.53. Turning to the full year of 2015, we currently expect our reported revenue to be approximately $502 million to $512 million, reflecting growth of 17% to 19% over the full year of 2014.
Again, in our outlook, we are currently assuming an average euro to U.S. dollar exchange rate of $1.10 for the remainder of the year, compared to $1.33 average we experienced during the full year of 2014.
If we were to assume a euro to USD exchange rate of $1.33 for the full year of 2015, consistent with the full year of 2014, our expectations for revenue in 2015 on a constant currency basis would be approximately $521 million to $530 million or 22% to 24% growth. From a profitability perspective, our full year outlook for non-GAAP operating margins in 2015 is 41.5% to 42%.
This is an increase of 1 to 1-1/2 points from the midpoint of the initial outlook we provided for 2015 in November of last year, as our business is delivering more leverage than we expected at that time. We expect our non-GAAP tax rate for the year to be approximately 26% and to have approximately 77.6 million weighted average diluted shares outstanding, driving non-GAAP earnings per share of $2 to $2.08.
With that, I will turn it back to Kevin.
Kevin Thompson
Thanks, Jason. While the second quarter of 2015 did not go exactly as planned on all fronts, we believe that our results for the first half of the year have positioned us well to deliver a solid year of growth in a year of expanding profitability and operating cash flows despite the dilutive impact of the Librato and Paper Trail acquisitions.
We have delivered strong growth in total revenue of 25% on a constant currency basis for the first six months of 2015, coupled with non-GAAP operating margins of 42%, with operating cash flows of $92 million. During the first half of the year we had success in a number of significant areas of our business in driving strong growth rates in new business sales.
These areas include our MSP business, the U.S. federal market, installed base sales, as well as our cloud management offering.
We also had several key core products which have performed well across our key geographies. We plan to build on this success in the second half of 2015 by increasing our investment in these areas, while we are addressing the issues we encountered in our demand capture machine that prevented us from growing total revenue at our planned rate of over 25% on a constant currency basis in the second quarter.
And as Jason indicated in his remarks, while we have lowered our expectations of total revenue growth a bit for 2015 to a range of 22% to 24% on a constant currency basis as a result of the impact of our second quarter results on the full and adjusted expectations of demand capture for the second half of the year, we still believe, with solid execution, we have the ability to drive our total revenue growth for the full year back within the range of outlook we provided on the first quarter earnings call. In addition, we feel confident that our non-GAAP operating margins will actually exceed the outlook we provided on our first quarter call despite the small decline in projected total revenues.
We also have laid out an organic product development plan for our network and systems management product line for the next 18 months, we are very excited about. We believe there's several important opportunities to more effectively solve a number of significant problems within network, server and application performance caused by the increase in highly distributed and virtualized environments for our existing customers, as well as potential new customers.
We believe we are well-positioned to solve these problems for IT pros and, as importantly, that our customers expect us to solve these problems for them. We are planning the total level of investments in products to remain consistent with our prior plan for 2015.
We are in the process of making meaningful changes in our -- in the levels of investment on an individual product basis, as well as prioritizing development on several new products that we believe will open up large market opportunities for us within the IP pro community where our brand is already very strong. We expect to hear more about these opportunities and the new products we plan to develop to attack them over the next several quarters.
In addition, as we continue to invest in our MSP business that I spoke about in April, we have another exciting new product release on the deck for this business that is planned for later this year, which we believe our MSP customers will be as excited about as we are. As we get closer to the release of this new cloud-based offering, we will provide additional details.
I think it is also important to note that our revenue model continues to shift, as recurring revenue, which is comprised of subscription revenue and maintenance revenue, has risen to 66% of total revenue for the first half of 2015, compared to 63% in 2014. The recurring revenue has also grown at a rate of 32% on a constant currency basis.
We expect the recurring revenue will continue to be a larger and larger percentage of our total revenue as we look into the future as we drive fast growth in our subscription products and maintain our high customer retention rate. In addition, the leverage in our model has continued to be strong with non-GAAP operating margins in the second quarter of 2015 at 43%.
And finally, the cash generating power of our business is increasing, as cash flow from operations has increased to 46% of revenue and actually would have been 5 percentage points higher excluding $7 million of non-recurring payments made during the second quarter. We have a number of initial initiatives underway that we're enthusiastic about and that we believe will contribute to our future growth.
We believe we will quickly address the demand capture issues we encountered in the second quarter and that we have made or in the process of making additional changes to compensate for any demand capture shortfall over the next several quarters. With that, we will open up the call for questions.
Operator
Thank you. [Operator Instructions] And we'll go ahead and take our first question from John DiFucci with Jefferies.
Please go ahead, sir.
John DiFucci - Jefferies
Thank you. Kevin, you said the issue in the quarter was a drop in the quality of demand capture.
And I -- so I kind of understand kind of what that means, and you went through some things you're doing to rectify it. But I still don't understand like why you saw a drop this quarter after four quarters in a row where you said it was strong.
So I'm not sure how you can rectify it. [Inaudible] but could you, I guess, tell us like why you saw the drop this quarter?
Kevin Thompson
Yes. So, look, I think the key thing to recognize, John, is, as we are growing the volume of demand that we're capturing using the web, right, that's how we capture demand is we use the web, we use various forms of digital media to reach potential buyers, and try to find them in the places that they already are on, on the web, looking for information.
And as we grow that volume of demand, we are adding new sources on a quarterly basis, so, finding new ways, if you will, to find that buyer wherever they happen to be on the web, in social sites, in community sites, on mobile devices. So as we add new sources, we are finding out what the quality of those sources are going to be.
And quality is a bit of a backwards-looking indicator for us because we don't know what the quality is until we've captured the demand, put that demand through a sales process, and look at how much in dollar comes out at the other end of the sales process. So it takes us, when we add a new source of demand, it's going to take us anywhere between, you know, 45 days and 120 days to figure out what the level of quality of that source is.
Once we know what the quality level is, then we can build that into our plan and we can project both quality and volume that we're going to get from that individual source. So, part of the challenge for us as we grow and as we add additional sources, we try new forms of social media, all of which we all see every single day, to try to reach our potential buyer, we don't know when we drive those new sources exactly what the quality is going to be.
Now, we analogize what we think that quality is going to be based on similar sources. We do testing before we roll out a source in large, you know, to capture large quantities of demand.
But, you know, that process is not going to be perfect. And I think what happened in the second quarter is that too many of the new sources that we have been using, you know, really starting to hear too many of those new sources didn't end up having the level of quality we expected them to have.
I would have liked us to catch it quicker than we did. We really saw the impact of it in the month of June, particularly the last two weeks of June.
As Jason indicated in his remarks, going into that, you know, the back half of June, we still believe, based on the pipelines we had, that we'd be able to get to a number that was within or above our outlook. We just didn't see the pipeline close at the rates that we would have expected it to based on our historical metrics.
So what are we doing? One, we have done a lot of work to understand the quality of the sources that are new, and some of the sources we're not using any longer, because we've decided that the sources is not of high enough quality compared to the cost, to continue to use it, and it's putting noise into our sales engine, and that noise is causing the engine not to perform at the level that it normally has.
And so that's part of what we're doing. We also are increasing our focus on both older sources and some new sources where the quality of demand was high.
And we're going to drive higher volume and demand from those sources in Q3 and Q4 than we did in Q2. So, really rebalancing the sources we're using in order to capture demand.
And then the last thing we're doing, I mentioned in our remarks, is Jason Marshall has come onboard. He comes out of the B2C world where he's been running a business that is all digital and has some additional ways we can look at quality of demand and volume of demand that we've not been using.
And so he's going to bring some expertise and knowledge we don't have to the table. Now, I started looking for Jason before the last two weeks of June.
So it's like I found him, hired him, and he's here, that quickly. So it's something I knew we needed to add to the team.
I knew that we didn't know everything we need to do lots of times. We're the best company in software at reaching the web, reach [inaudible] buyer, but that's a very low bar statement because most companies' software don't do it at all and the few that do do it poorly.
And that we can be much better than we are and that we will be much better than we are. So hopefully that provides a little bit of additional detail without going into a lot of terms that probably a lot of people in the call are not familiar with.
John DiFucci - Jefferies
That is helpful, Kevin. It is.
And it also really emphasizes the -- how dynamic your process has been, which -- and I think we all realize and we kind of push the envelope -- pushing the envelope with all the digital marketing and selling. But if you could, just because just -- could you just give us an example of a source that didn't -- like just it may be a little helpful to me anyway, given where the limit does --
Kevin Thompson
Yeah. There's a lot of ways to capture demand on the web.
Google has a number of different methodology you can use other than just straight paid search. So they've got lots of different things you can do, and I'm not going to name all of them, but lots of different avenues you can use to try to reach a potential buyer in some level of volume other than just straight SEO, SEM around terms.
And we've tried a lot of those. So, whether it's the Google banner ads or some of the Google community sites.
There's also IT pro community sites that have been created over the last several years. There are things we do inside of mobile, whether we're using Twitter or whether we're using Facebook or we're using different forms of social media like that to reach potential buyers, we do all of those things.
We're using LinkedIn as a marketing vehicle. All of those we use, and some of those have much higher quality than others.
Straight paid search actually has pretty high quality. Some of the other means that Google allows you to use to try to reach potential buyers, the quality has not been as high.
Some of the advertising we've done on certain IT pro sites, whether those are more journalistic sites or online magazines, that kind of stuff, have been very strong. Some of the advertising methodologies we've used have not been very strong.
And some of the sites we've used have not been very strong. So it's a lot of different ways you could reach a potential buyer, but those are the ones that probably people understand the most.
We use all of those. And they have varying levels of quality.
And then we use a number of things that are probably a little less known than that, to try to reach potential buyers. I think the key is we've got to identify more quickly than we did and more quickly than we are right now when a source is going to have a lower quality than we expected to.
Like I said, we test them but you don't really know until you roll them out in volume. So we have to be better at that, and that's what I'm confident that Jason will be able to help us do, and he and his team are already doing some things right now to change how we're thinking about quality and quantity of demand, as I think we will be better in Q3 than we were in Q2.
But, you know, I've said lots of times, what we're doing, nobody else has done in enterprise software. And so the things we're trying haven't been tried by almost any software company in the past.
And as we scale, we're going to run into situations where things don't work exactly the way we expect them to. But I think what we've shown over the years is we quickly figured out what didn't work and then we find new things that will work and we get things rolling at the rate we want them to.
So, yeah, we ran into a bump I'd rather not -- would rather not have run into, but I have tremendous confidence that we're going to figure that out, just like we had the other times when we struggled on the demand capture side as we scale. Because we're a very large enterprise software company now that is driving a lot of demand use in the web and, you know, I don't know of any company that's driving 30% of the demand enterprise software using the web that we are.
John DiFucci - Jefferies
Kevin, that's really helpful. And the numbers now look more attainable for this year, still decent numbers.
But just a quick follow-up for Jason, and then I'll turn it to someone else. But Jason, the Paper Trail was $41 million, you mentioned it.
On the cash flow statement there's this $50 million in cash expenditures for acquisitions. And I just wonder what that other $9 million.
Did you buy something smaller that you didn't mention?
Jason Ream
Yeah. So we did a small IP purchase.
It's really a tuck-in product. And well, a tuck-in to our products and some of the stuff that Kevin was alluding to that will be in further announcements throughout the year.
John DiFucci - Jefferies
Okay. Thanks a lot guys.
Kevin Thompson
Thank you.
Operator
And we'll take our next question from Kirk Materne with Evercore. Please go ahead, sir.
Matt Williams - Evercore
Hi guys. It's Matt Williams actually on for Kirk.
Kevin, you kind of just touched on this, I'm just curious if you could speak to it a little bit. I think you guys have gone through a few sort of hiccups around demand capture historically, and I'm just wondering what you saw this quarter that is sort of different from what you've gone through in the past, and maybe I guess another way, sort of flipside of that, would be how similar it is, and then, yeah, in terms of lessons that you've learned in the past, how applicable are some of those to what you're dealing with right now.
Kevin Thompson
Yeah. Look, I think it's both different and it's the same.
The same meaning, you know, I think we're brought into these at different points in time as the company is scaling and as we're growing to a larger and larger size and as we're trying to drive larger dollars of growth every quarter obviously of a larger base number. And as we've done that, we are pushing the envelope in terms of the types of the things we're doing on the web to reach an enterprise software buyer.
And when I say pushing the envelope, meaning -- I mean trying new things, finding new ways to capture the attention of a potential buyer, finding new ways to communicate with a potential buyer. And we're going to find those buyers sometimes in different stages of the ready-to-buy cycle if you will, meaning certain sources when we find a potential buyer, they have a burning problem, they're ready to buy today.
Certain sources we use like maybe Google display network and others that the buyer may be interested in learning something, maybe that buyer is not quite ready to buy yet. And so some of our historical metrics around timing from interest to close of deal may need to shift for those sources.
And so that's really the same. The sources are different because, as you know, there's a bunch of sources that are available to us today and things that, you know, places people are going in the digital world that they weren't going to two years ago.
So while the sources are different, and that's what's different, the issue tends to be somewhat the same. And I think we've gotten better at testing new sources.
I think we've gotten better at identifying issues than we have in the past. I think we respond more quickly.
But clearly in the second quarter we didn't respond quite as quickly as we would have liked to. And that is because no matter what we do, quality is a backwards indicator.
We don't know what quality was until we've had time to measure it. And it's going to take anywhere between 30 days to 90 days for us to really understand quality.
And typically that's going to be closer to 90 than 30 because we have to get enough volume through a full cycle before we can really look at the quality of new sources. So I think the key is, one, bringing -- continuing to upgrade the team we have, and that's why Jason Marshall is now part of our team, and bring people in who know digital marketing better than we do, people who have been in the B2C world, they've used sources we've never even thought about, they've used sources we're beginning to try, and they have a view and experience as what the quality happens to be.
So in terms of addressing the issues, it's going to be, you know, the same type of activities we've done before, but I think we've got a team that's more experienced and has more knowledge to address them than we've had in the past.
Matt Williams - Evercore
Got you. And maybe just one quick follow-up for me, but if you could talk a little bit about what sort of headwinds or how much of a headwind you're seeing from the public cloud vendors, and is that impacting demand or how much is it impacting demand from what you guys are seeing?
Thanks.
Kevin Thompson
Yes. So, look, overall the public/private cloud is driving a lot of growth for us.
We have $16 million of subscription revenue in this quarter, you know, growing very, very quickly. And a lot of that's being driven by the concept of cloud, if you will.
Not all public cloud vendors, but cloud. So that's absolutely being driven by the public cloud vendors.
Librato's [ph] revenue and Paper Trail's revenue, is all being driven by technology that's in the public cloud and people's need to manage that technology. So, definitely driving growth for us.
We don't see that as a headwind in our on-premise license product business right now. We don't really expect to over at least as far out as we can see right now, because that technology [inaudible] on-premise needs to be managed.
The technology in this on-premise is actually not reducing, it's still increasing, and I think we are the best vendor in the space right now of managing on-premise technology, and we're investing I think more heavily than anyone else. Also, we talk about this a lot, we really believe ultimately the trick's going to be, can you deliver the ability to manage the hybrid IT world?
Can you manage the assets that sit on-premise? Can you manage the assets that is sent in public cloud and private cloud?
And can you manage all the connections between the user and wherever that asset happens to fit? That's what we're working to create, that's what our technology strategy is about, it's what our acquisition strategy has been about, and now we're going to bring all that technology together to give ourselves the ability to manage that hybrid environment that I think we and a bunch of other technology companies believe is where we're going to be for, you know, at least as far out as we all can see.
So what I view going on in cloud is an opportunity. Now we have to capitalize on that opportunity and we have to make sure that our on-premise products remain incredibly relevant.
We have to make sure that our cloud management products solve the problems that need to be solved in the public cloud to private cloud, and that we then create that connection, which we've not done yet, we're working on it, and that we do it in a way that customers can still solve individual problems, whichever individual problem they happen to have, but they also then can connect our technologies together and solve that hybrid connected problem when they need to.
Matt Williams - Evercore
Great. Thanks for taking the questions.
Kevin Thompson
Thank you.
Operator
And we'll take our next question from Steve Ashley with Robert W. Baird.
Please go ahead, sir.
Steve Ashley - Robert W. Baird
Thank you very much. I just wanted to talk about funnel activity and if there was any crowding out, by putting [inaudible] chaffs [ph] into the funnel, did it actually help reduce the demand for the core products?
Kevin Thompson
Yes. So I think the way to think about what happened in the second quarter, Steve, is really not that.
I think we did put some demand into the funnel that was not of a quality level than we wanted it to be. But the volume growth, you know, wasn't high enough to offset the quality drop.
So the quality drop is really the issue. We had some processes employed that we put in place to deal with the chaffs [ph] as you called it, and I think we dealt with that relatively effectively.
We just didn't get enough demand into the top of the funnel. I actually probably would have taken even a higher volume of demand, even if the quality overall would have dropped a little bit, because I think we could have dealt with that, and we would have ended up with a little bit more bookings.
So it was really this quarter more that we didn't get enough volume and then the volume we got was a little lower level quality. And so, given that 30 to 45-day sales cycle, meaning the first half of the quarter we're closing deals that were in the pipeline when we started the quarter, the second half of the quarter we're closing deals that we created the opportunity from the beginning of the quarter to the middle of the quarter, that second half of the quarter, particularly the last part of June, is where we felt the impact of some of the opportunities in the pipeline were of lower quality than we believe them to be because the opportunities were created from sources that just weren't at the quality level that we expected.
Does that make sense?
Steve Ashley - Robert W. Baird
Sure. And then I think tomorrow, one of the questions we will be asked is about maturity and the maturity of their core products, that they're hitting the wall.
And I think you remember that a year and a half ago, that became a narrative, and then I think you proved through recent times that it had gotten better. How would you respond to and provide some confidence that the core products have not hit maturity?
Kevin Thompson
Yes. So, look, I think first thing I would say is we actually had a number of core products perform very well in the second quarter, you know, MPM, Virtualization Manager, Storage Resource Monitor, you know, all performed very well in the second quarter, and they've been performing well as you know for a number of quarters now.
So that'd be first part, is we had a number of core products perform well. The second thing I would say is that, you know, the issues we saw in the quarter, and this is where you have to be [inaudible] to know this, had nothing to do with the products and whether they continue to be relevant and whether the problems they solved need to continue to be solved.
But we just didn't capture enough of the demand that exists in the marketplace to get all the way to the license product, number one, and -- two. The other thing I would, you know, we're now -- we have a lot of products, like our MSP product, our cloud management product, that grew very quickly in the quarter.
And those products continue to perform well. And as we create those connections between our on-premise products and our off-premise products if you will, then the on-premise products I think become even more relevant than they've been historically.
So we're just not seeing that the products are relevant. Now what I would say, and I indicated this in my remarks, we are going to reallocate some investment dollars on the product side.
I don't think that I've had the investment dollars on necessarily allocated exactly the right way, because we have a number of products I believe can grow very quickly, and I think we've spread our R&D resources a little bit too much like peanut butter, if you will, on a piece of bread, and we've got to change that approach. And we've done that.
And maybe they've already done it or in the process of doing it, and that'll be done relatively quickly. Also we've identified, as I indicated, a number of additional problems we think needed to be solved a lot more effectively than are being installed today in the on-premise world, in network management and systems management where we are incredibly well-known, and we're going to bring some products to market to plug some of those gaps over the next year, year and a half, you know, as we nail those timelines down, I'll start to share what those products do and when we expect to have them.
But we have a number of ideas that we're beginning to work on, because we have customers asking us to solve those problems. So the on-premise problems continue to exist.
They'll continue to exist for a very long time, probably well past when I continue to be working, and I intend to work for a long time, and we want to make sure we're dominating in that market, and that as others abandon that market, we'll just continue to increase our market share. And we've shown time and time again, since I got here in 2006, that we can, will, and almost always have outgrown the market by a very wide margin, and we plan to continue to do that.
Steve Ashley - Robert W. Baird
Great. Thank you.
Kevin Thompson
Thanks, Steve.
Operator
And we'll take our next question from Scott Zeller with Needham & Co. Please go ahead, sir.
Scott Zeller - Needham & Co.
Yes, hi. Could you comment on two metrics you shared in the past?
One is the percentage growth of inbounds into sales, like the volume of opportunities year on year. And the other thing would be bundling effectiveness of sales, just color commentary on the bundling effectiveness.
Kevin Thompson
Yes. So in terms of volume demand, we did see volume growth year over year by a decent amount, but we actually saw quality fall.
And it's really the combination of those two metrics to turn into additional value inside the pipeline. So in general we saw overall conversion rate be about where they had been.
And so we just didn't -- at the level of quality we drove in Q2, we needed volume to grow faster. So, volume grew and it grew in double digits, but quality fell and almost offset the volume growth.
So the conversion rates by the sales team were relatively consistent. So as you look at our total core product growth of 6%, it gives you a relatively reasonable, it's not exact and never will be, but approximation of the combination of the volume change year over year and the quality change year over year.
As it relates to bundling, I wouldn't call a bundling, I'd call it attached, and then cross-sell, upsell. So, attach, and let's define attach as someone comes in, downloads a product that's not a customer of ours, and then we get them to buy another product.
Or they come and download two products, we give them to buy two products. The attach rates of additional products and the sale were relatively consistent with what we've seen over the last four or five quarters.
The installed base team on the other hand, as Jason indicated in his remarks, grew by 41% year over year globally, and North America actually grew faster than that. So we're seeing really good success, continuing to see really good success on the installed base side.
We're going to continue to push that. We think we can continue to grow our sales at our installed base at at least the rate they're growing now, and the team knows, I believe we ought to be able to grow it faster because the opportunity is so large.
And so we're going to -- you're going to see us focus on that a lot over the next couple of quarters. Feel okay about attach rates.
They can be higher. But they are consistent with where they've been for the last four or five quarters and I do think they'll improve, but they were pretty good in the second quarter.
Scott Zeller - Needham & Co.
And just a follow-up. I guess, Kevin, given the fact that it's been several quarters since you'd had what's called a demand quality issue.
Is it possible that you're facing an environment where channels or marketing strategies have a shelf life of a certain time period and you may just have to get used to specific ends, you know, a lack of runaway, let's say, for certain channels?
Kevin Thompson
Look, I'd probably put it a slightly different way. So I think certain marketing channels have a limit to the amount of volume you can capture through them in a 90-day window.
So without a doubt, certain marketing sources, even in digital media, there's a limit -- now limit really depends on what you're willing to spend, meaning if you're trying to drive a certain ROI inside a 90-day window, there is a limit to the volume you can capture. You can do two things.
You can drop your ROI bar and then you can capture more volume. Or you can add new sources and keep your ROI bar roughly where it's been.
Our strategy has been, we don't drop our ROI bar much. We'll drop it a little bit, but you can see inside our operating margin, we don't drop it much.
And instead we generally try to find new sources that we believe will have ROI at the level of our average or better because we generally have found that over the years to be a more productive way to grow demand and thus grow the business. So I think the key for us is to know, one, we need to drop the ROI bar window [inaudible].
And as we add new sources, we have to much more quickly understand the ROI of those sources. And that's what we're working on right now, is to make it less and less backwards-looking and more and more real-time in terms of our ability to understand what is happening inside of quality.
And it absolutely can be done. It's not impossible.
It will get done overnight and it will get better as we move forward, because there are some tools that exist today that didn't exist even a year ago that will allow us to do more effectively, and more tools will exist a year from today, because as you guys know, lots of people are investing in lots of different marketing tools these days to help digital marketers have a better view of what's really going on. So that's the way I'd phrase it, is a little bit different.
It's not so much the sources die and go away; it's more that they have a limit in the 90-day window at a certain ROI, in some cases we may need to drop our ROI when they're really good sources, because the more volume it drives, the more expensive it gets per unit captured. In other cases, I just want to make sure we're doing a better job of identifying sources quickly that are of high quality and identifying sources quickly that are low quality.
Scott Zeller - Needham & Co.
Thank you.
Kevin Thompson
Thank you.
Operator
And we'll take our next question from Daniel Ives with FBR Capital Markets. Please go ahead, sir.
Jim Warren - FBR Capital Markets
Great, thanks. This is actually Jim Warren for Dan.
Could you just talk about your appetite for more M&A and how the integration of your recent acquisitions are going, and then how you might be thinking of focus areas going forward?
Kevin Thompson
Yes. So, look, I think we continue to look for technologies that allow us to achieve our strategy, right?
Our strategy is to be able to manage all things in the hybrid IT world. So it's a pretty simple strategy in five or six words it takes to say it, but a much more complicated strategy when you think about the things you have to do.
Because the other key part of that is we need to be able to do that for organizations of all size. And so the acquisition we've done in the past, MSP acquisition, our database acquisition, our cloud management acquisitions, are all, you know, at all good stages in the integration process.
Some fully integrated and have been for a while; others are newer to the fold and are still in process of integrating. And we integrate them in different ways dependent on what the acquisition is, where it fits into our product strategy, where it fits into our go-to market strategy, what sales team is going to sell it, is going to determine how we integrate it.
So I think we are at a good stage in integration of the acquisitions we've done. We're at a stage that if we have a desire to do additional deals, we're ready and capable to do additional deals.
I think we're going to continue though, and you've heard me say this a lot over the last 15 months, is the acquisition you see us do moving forward are going to be strategic, acquisitions that we believe have large market opportunities that fit into that strategy of managing all things IT, for organizations of all sizes in the hybrid IT world. And those are the type of deals we're looking for.
We're not looking for more small tools acquisitions, we're not looking for, you know, we're not looking to acquire customer bases anymore, I think we've done enough of those. And we haven't done really any of those over the last six quarters.
And you should expect that the strategic acquisitions, if we do them, acquisitions we believe have large opportunities on their own, and they create large connected opportunities across the rest of our product portfolio. So it is about creating connected opportunities to manage that hybrid IT world so that I capture a potential customer when they have an individual problem, because that's how we're going to find them.
But then I give them the ability to very quickly put together a number of our technologies when they want to, to solve larger problems that span different areas of the IT infrastructure. And I think one of the things we don't get enough credit for is, while we can solve very small problems for very small companies, we also have the ability to scale and solve very large problems for very large companies, but we typically do that for large companies after we've solved a very small problem for them, and then they come back and buy a lot more technology from us.
So we really do have the ability to solve problem for an organization of any size and really a problem of any size, and that's what we continue to be focused on. We'll build a lot of that technology, and I talked about the fact we've got, you know, what we believe is a pretty exciting roadmap over the next 18 to 24 months for our on-premise technology, and also some additional cloud-based technology that we're going to develop, but we'll also keep our eye out for technologies that can expand our market opportunity or companies that have good foothold in spaces where we believe we should be.
Jim Warren - FBR Capital Markets
Great. That's it for me.
Thanks very much.
Kevin Thompson
Thank you.
Operator
And we'll take our next question from Greg McDowell with JMP Securities. Please go ahead, sir.
Greg McDowell - JMP Securities
Great. Thank you.
My question is also going to draw some parallels to 2013, to 18 months ago. And it's around some of the tweaks or adjustments you need to make.
I mean, would you characterize the tweaks or adjustments as being less drastic than the changes you made in 2013, more drastic, about the same? And I guess what I'm driving is, last time we went through this, four or five quarters pattern of decelerating license growth, I guess I'm just trying to get my arms around why it won't be like that again this time.
Kevin Thompson
A couple of things. Look, I think I would characterize it as less drastic given -- to answer your question simply the way you asked it.
But I would say is, it's less revolutionary, if you will, meaning we are not going to -- we don't need to add a bunch of new ways to reach customers. We just need to make sure that we can really measure and understand the ways we're using.
We need to improve our users' experience on our website -- websites for that matter. And we are better at this today than we were 18 months ago or 24 months ago.
We have a team that is -- got a lot more experience at it. We have leadership in place that had been digital marketers for a long period of time, and we didn't have that 24 months ago.
So we have a much more experienced team. I think the changes are less wide-sweeping than they were before.
And we know how to do it. I think we've done it before.
We're going to do it again. I think we've shown a track record of, when we hit a bump, that we know how to execute through that bump.
And the reality is all companies hit bumps. I'd like to never hit one.
I'd like to tell you I'm never going to get hit one, but I know that's not realistic. We're going to hit them.
The question is, how do we manage them? I'd like to manage them a little more effectively within the quarter than we did in Q2.
But I think we do a lot of things right in Q2. We had a lot of areas for business growth.
We generated tremendous cash flow, great profitability. So I think we shown the capability as a team to operationally manage through almost anything that we encounter.
Sometimes it has a little bit of a revenue blip, but it's, you know, at least to this date, never had a profitability or cash flow blip, because we manage our way through that incredibly well. So we know how to do this, we've done it before, we've got a really great team in place and a team that's getting better as we expand it, and I'm confident we'll get through it.
We've given ourselves six months to get through it. I hope to get through it faster.
I hope to end the year back inside of the revenue outlook we provided on the first quarter call. I will be happy if that doesn't happen and my team will be happy.
We're a competitive team. We don't like every coming up short, and we take these things incredibly personally and the team is working hard to make sure we address the issues.
I'm not worried about our ability to address them. We will address them.
It's just a matter of how quickly we can get it done.
Greg McDowell - JMP Securities
Thank you.
Operator
And we'll take our next question from Sanjit Singh with Morgan Stanley. Please go ahead, sir.
Sanjit Singh - Morgan Stanley
Thanks for taking my question. Jason, I appreciate that detail on some of the seasonal aspects of maintenance revenue.
I'm looking at your full year guidance, and for the balance of the year. I know you mentioned that it's going to take a while to see some comeback on the license revenue side.
So in terms of just sort of seasonality heading into the federal quarter and heading into the commercial quarter in Q3, Q4, any comments there on -- in terms of what we should be expecting in terms of license revenue?
Jason Ream
Yes. Sanjit, without guiding to the specific components of revenue there, since we just provided our guide at the total revenue level, I think it is still very instructive to look at the historical patterns.
We've typically had a pretty predictable seasonal pattern by quarter where we do jump up in license revenue in Q3 with the close of the federal fiscal year. And as you heard Kevin mention and, you know, as we've talked about on the last four or five calls, we've invested a lot in our U.S.
federal business both on the marketing and sales side, and we've seen good results, we've got a good team there, and I think we're very well-positioned to serve that market both in how our products work and how they're delivered. So we expect a good quarter in Q3 from our federal business.
And then in Q4, what you've typically seen historically is that we get a further sequential increase in license revenue there just because it is the commercial yearend for most companies, the fiscal yearend for most commercial companies. And while we're not a big deal company, we're not doing a lot of big deals on December 30th or December 31st, we do tend to get some "budget slash" benefit in Q4 of the year.
So I think the best way to think about modeling that is to look at our historical patterns, and that'll be a relatively good guide on the license side.
Sanjit Singh - Morgan Stanley
I appreciate that, Jason. Kevin, if I look at your businesses from a high level and look at the areas of strengths and the areas of weakness, the easy sort of takeaway would be like, wow, cloud is really doing well.
And so when it comes back to this demand capture or problem or the issue on the on-prem business, what's to say that your higher-quality prospects aren't the ones who are -- want a cloud product or is tied to more of a cloud use case? So that maybe some of these higher-quality process that you're seeking out, they may be diminishing in terms of population.
Kevin Thompson
Yes. So, look, I think there's a couple of things we look at that tells that that's actually not the case.
So, one, when you look at the amount of IT infrastructure deployed on-premise, I don't care who you talk to, whose study you look at, which CIO you talk to, we're still seeing growth in the volume of IT infrastructure deployed on-premise. In fact, for the last five quarters leading up to this quarter, we have been driving really good growth in our on-premise products.
And that just didn't change in the last 90 days. We also talk to thousands of IT pros in every single quarter.
So we know what IT pros are thinking about, we know what IT pros are doing. So, yes, are more IT pros dealing with cloud, be that private or public or hosted, if you will?
Yes, they are. Definitely more today than they were 18 months ago.
And they need some of those problems solved for them at the same time that you're still dealing with a larger IT infrastructure on-premise performance problem. They got -- they need to create the connection.
So [inaudible] as far as I can see right now, what we're going to have happen is cloud is going to grow quickly in terms of the amount of technology deployed within the cloud, and the so the management opportunity there will grow with it, and we're going to make sure we take advantage of that, and we're already taking advantage of that at some level. But on-premise management is going to continue to grow in terms of the overall market size I think for a number of years.
But even at the point that it's easy to grow and goes flat, I expect we will grow quickly in that opportunity because that infrastructure still needs to be managed and I think we are the best company in the space to manage that infrastructure. So that's what we look at, that's what we hear from our thousands, tens of thousands, more than that, end-customers that we have, and that's one of the I think advantages we have, is we have so many customers, so many more customers than most of our competitors and we are communicating with them on a daily basis on (Flac) and a couple of other things we're doing.
I don't know if you guys are looking, but if you were, we just I think finished today or maybe it finishes tomorrow, I can't remember, doing a great online thing we call (Flac Camp), which is a live -- it's a live event for our customers, which is really great, and you can get view if you go and watch that, the things we're talking about, the things our customer care about, and some of the questions that they're asking, so will give you a good feel of what's going on in the world as we see it.
Sanjit Singh - Morgan Stanley
Appreciate the answer, Kevin. Thank you.
Kevin Thompson
Thank you. And we have time for one more question.
Operator
Perfect. And we have our last question from Tim Klasell with Northland Securities.
Please go ahead, sir.
Tim Klasell - Northland Securities
Hey guys. Thanks for taking my question.
I have two quick ones. First, it looks like your marketing efforts, you mentioned B2C a couple of times.
And I always thought of you guys as sort of being like more B2B. Are you going that route because you want that skill set or is there something changing in your market where you want to sort of bring more of a B2C type approach to your marketing efforts to improve your conversion rates?
Kevin Thompson
Yeah. Look, I think it's a couple of things, Tim.
So, one, I think we've all, you know, we knew this well before anybody started talking about it, but the whole concept of the consumerization of IT, the fact that IT pros in their day jobs, inside the companies they work in, behave a lot like consumers as they're looking for things to buy, to solve the problems they're trying to solve. That's something that's been building over the last number of years.
We've been taking advantage of that ever since I got here, and even before I got here. The founders of our company were taking advantage of that behavior.
So we know the behavior exists and we want to make sure we take advantage of it. It's not that we're trying to sell to a consumer, we're not, but we know that behavior exists.
But more importantly, what we're trying to do, is in the B2C world, people that are using the web in the B2c world reach potential buyers are reaching a much larger volume of potential buyers than we even have. And they're using a lot more means to try to reach that buyer than we have used in the past.
And because we believe the IT pros behave like consumers when they're at the job doing their work every day, we need to make sure that we are leveraging all the forms of digital media that they are going to be using to potentially find the information, the knowledge, the products that they need. So it's really the skill set, the knowledge, the experience, and then using that to take advantage of the fact that we believe IT pros do behave like a consumer at work and searching for the same things that they need.
Does that make sense?
Tim Klasell - Northland Securities
Yes, it does. Thank you very much.
And then just finally, with the MSP and cloud business obviously outpacing the other segments of the business, what should the shifts be in the metrics that we should expect so we don't get surprised? I'm thinking things like renewal rates, ASPs, sales cycles, that sort of a thing.
Anything there that we should be prepared for?
Kevin Thompson
Yes. So, look, I think obviously in our subscription business, retention becomes even more important than it is in our maintenance business.
Now you know this because you've been following us for a long time, we have a very different kind of maintenance model where maintenance is a much larger percentage of the value -- of the license value we sold to the customer. So, license -- maintenance has always been a much bigger deal to us than any other license model software company.
So we put processes in place and procedures in place I think a lot earlier than most license software companies do to make sure that we have high customer retention rates. It's really important on the subscription side of our business that we have high customer retention rates.
In our MSP business, those retention rates are improving, and they have been really every quarter for the last year, and they're very strong. Can they get stronger?
Yes. Are we going to try to make them stronger?
We absolutely are, and the teams around the world know that that's something we need to be focused on. In our cloud management products, those are still relatively new.
We're not -- only through a year as of June 30 with one of those products which is Pingdom. And so we're still learning about retention on that side in terms of what are the rates and what do we need them to be.
They're good. They could absolutely be better.
And the numbers just aren't big enough yet for us to know how much better they can be, but right now we've assumed they're going to be consistent with where they are now. And then we're going to figure out a way to make them better.
But those retention rates are important, we'll talk about and we'll share them with you and make sure you know are they getting better, are they getting worse, and are they within the range of our expectations.
Tim Klasell - Northland Securities
Okay. And on ASPs?
Kevin Thompson
So, ASPs are interesting. So with our MSP business, our ASPs are not that much different than our on-premise license product business for the core products that we sell in our MSP business.
So those are very much the same. For our cloud management products, Librato, Paper Trail and Pingdom, a lot of those are monthly subscriptions right now, because that's the way those companies were selling subscriptions, mainly using the web, almost no sales people involved right now.
And so the key there is to make sure that your monthly subscriptions turn into multiyear customers over time. And so those are a little bit different.
So the ASPs there are very small. Those are not included in the numbers that Jason gave you, of the $8,200.
Those are really for our core license products. We've not included MSP or cloud in those numbers.
That's something when we get a little better view of kind of how that's going to trend, we'll probably start to share a little bit. But the ASP for MSP very much the same.
The ASPs for cloud much smaller because they're monthly. Now when you annualize them, so let's talk about them on an annualized basis for a minute, still a lot smaller than the $8,200 we're driving right now.
Tim Klasell - Northland Securities
Okay, very helpful. Thank you.
Kevin Thompson
And with that, we're done with our second quarter earnings call. Thanks for listening in.
Operator
That does conclude today's conference. We'd like to thank you for your participation.