iRobot Corp (IRBT) 2018 Q2 法說會逐字稿

Good day, everyone, and welcome to the iRobot Second Quarter Financial Results Conference Call.

This call is being recorded.

At this time, for opening remarks and introductions, I would like to turn the call over to Elise Caffrey of iRobot Investor Relations.

Please go ahead.

Thank you, and good morning.

Before I introduce the iRobot management team, I'd like to note that statements made on today's call that are not based on historical information are forward-looking statements made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995.

These forward-looking statements are subject to risks and uncertainties, and involve many factors that could cause actual results to differ materially from those expressed or implied by such statements.

Additional information on these risks and uncertainties can be found in our public filings with the Securities and Exchange Commission.

iRobot undertakes no obligation to update or revise these forward-looking statements, whether as a result of new information or circumstances.

During this conference call, we may also disclose non-GAAP financial measures as defined by SEC Regulation G, including adjusted EBITDA, which we define as earnings before interest, taxes, depreciation, amortization, net merger acquisition and divestiture income and expenses, gain on business acquisition, restructuring expenses, net intellectual property litigation expenses, and non-cash stock compensation expense.

A reconciliation of GAAP and non-GAAP metrics is provided in the financial tables at the end of the second quarter 2018 earnings press release issued last evening, which is available on our website.

On today's call, iRobot Chairman and CEO Colin Angle will provide a review of the company's operations and achievements for the second quarter of 2018, and Alison Dean, Chief Financial Officer, will review our financial results for the second quarter of 2018.

Colin and Alison will also provide our business and financial expectations for fiscal 2018, then we'll open the call for questions.

At this point, I'll turn the call over to Colin Angle.

Good morning, and thank you for joining us.

We delivered second quarter revenue growth of 24% over Q2 2017, slightly ahead of our expectations.

EMEA and Japan saw significant year-over-year growth of 51% and 31% respectively.

EMEA growth was in part driven by the revenue uplift from the acquisition, which closed at the beginning of Q4 2017.

Shipments to support Q2 holidays in the United States drove domestic revenue growth of 15% over last year.

Second quarter profit was higher than we expected due primarily to the high revenue, a slightly higher gross margin, and some delayed marketing expense.

Given our Q2 results and our outlook for the rest of the year, including our second half product launches, we are increasing our 2018 full-year expectation.

We now anticipate full-year 2018 revenue of $1.06 billion to $1.08 billion, which is year-over-year growth of 20% to 22%; full-year 2018 operating income of $90 million to $96 million; and full-year EPS of $2.30 to $2.50.

The anticipated EPS increase reflects several factors favorably impacting our tax rate, in addition to the core business uplift reflected in the revenue and operating income increases.

Alison will discuss these items in more detail later.

Now I'll take you through some of the highlights of Q2 '18 and our business expectations for the rest of the year.

We are excited to be selected for the fourth straight year as a featured product for the U.S. Amazon Prime Day.

We once again sold out, doubling our Prime Day sales volume, as we have each year since the event began in 2015.

The Roomba 671 was ranked number one in robot vacuum cleaners, number one in all floor care, and number three in all home and kitchen for the U.S. Prime Day event.

This is particularly significant when considered in the context of increasing availability of cheaper, lower-quality competition.

While our performance on Amazon Prime Day in the U.S. isn't material to our Q2 results, it has become an important gauge for second half expectations in the U.S. We expect a competitive landscape to be similar to last year's landscape and have a high level of confidence in meeting our expectations for the full-year U.S. revenue growth.

Building on our relationship with Amazon in the U.S., we were featured on Prime Day in EMEA and Japan this year for the first time, and the results of those regions were great.

An element of our rationale for acquiring overseas distributors was to optimize the e-commerce channel in those regions, beginning with a strengthening of the relationship with Amazon.

We are very pleased to be included in the events in EMEA and Japan, and look forward to further expanding our relationship with this key partner.

Our coordinated global Amazon Prime Day further strengthened our global brand.

During Q2, we announced a favorable initial determination from the International Trade Commission, ITC, regarding our patent infringement claims.

The finding recommended an exclusion order to bar the importation into the United States of certain robotic vacuum cleaners made and/or sold by ILIFE, bObsweep, and Hoover.

The initial determination and recommended remedy are now before the ITC commissioners, who are expected to issue a final recommendation by October 25, 2018.

If approved, the exclusion order will go into effect immediately upon publication regardless of further appeals.

We are very pleased with the initial results.

We will be carefully considering next steps with respect to our IP litigation strategy.

I know you are all excited about our second half product launches, as are we.

Although I can't give you any specifics, what I can tell you is that we are planning multiple launch activities.

The first will be happening in a few weeks.

It will be a soft launch for limited stores initially where we will introduce our next offering of high-end innovation at a lower price point.

Our next launch event will take place later in Q3, so stay tuned.

Braava growth momentum continued into the second quarter, driven by international demand, where Braava revenue grew roughly 50% over Q2 2017.

As we've commented previously, growth in the category has suffered historically at the expense of investment in Roomba [wares] campaigns.

Therefore, additional advertising in Braava is critical.

Lack of awareness and articulation of the product's value propositions are the hurdles we will overcome to drive this growth.

One of the components of our 2018 incremental sales and marketing investments we discussed in February is executing Braava marketing programs for all major markets.

In the U.S. in Q4, we will run a national TV campaign similar to the one we ran last year at the same time.

In addition, there is a TV campaign planned for Japan later this year and additional marketing campaigns in EMEA.

Now turning to our second quarter performance by region, we delivered 15% year-over-year revenue growth in the U.S., while international revenue increased 34%.

In the U.S., sales were robust through the Mother's Day and Father's Day holidays, as consumers continued to appreciate the value proposition of our products.

In EMEA, revenue grew 51%, where we saw strong demand, also benefiting from the revenue uplift associated with sales in the EMEA countries, where we now sell direct.

Japan revenue grew 31% in Q2 over 2017 without any benefit from the revenue uplift from SODC's acquisition, which we completed at the beginning of Q2 2017.

We still expect full-year 2018 growth of low 20% for the U.S., mid-teen growth in Japan, while our growth outlook for EMEA has increased to 30% to 35%.

And finally, a comment about the third round of proposed tariffs announced a couple weeks ago, which included Roomba and other RVCs.

We're currently exploring our options and the potential implications if these tariffs are enacted.

Our current full-year 2018 financial expectations do not reflect any impact from the proposed tariffs.

In summary, with half a year behind us and a very successful global Prime Day just concluded, we are well on our way to meeting our increased full-year financial expectations.

I will now turn the call over to Alison to review our second quarter results in more detail.

Thanks, Colin.

Our second quarter performance was ahead of our expectations.

Quarterly revenue of $226 million, an increase of 24% from Q2 last year, was driven primarily by growth of our Roomba 900 and 600 robots.

Operating income for Q2 was $13 million compared with $4 million for Q2 2017.

EPS was $0.37 for the quarter, compared with $0.27 in Q2 2017.

For the first half, revenue was $443 million compared with $352 million in the first half of 2017, operating income was $39 million compared with $26 million last year, and EPS was $1.08 compared with $0.85 last year.

Revenue growth of 24% for Q2 included domestic growth of [15%], EMEA growth of 51%, driven by stronger demand and ASP uplift from the Robopolis acquisition, and Japan growth of 31%.

The gross margin of 52% in the second quarter was slightly ahead of our expectations, driven primarily by favorable promotional expenses and returns.

We now expect gross margin for the year to be at the high end of our previously provided range of 50% to 51%.

The 300 basis point year-over-year increase in gross margin in Q2 was due to higher margins in Japan this year compared with last year, due to the SODC acquisition, and COGS improvements partially offset by the Robopolis intangible asset amortization.

Keep in mind that quarterly gross margin can fluctuate wildly based on mix, seasonality, and promotional activity among other things.

Q2 operating expenses were 46% of revenue compared with 47% in Q2 last year, and slightly below our expectations.

Q2 R&D and G&A were in line with our expectations, but as Colin mentioned, we had about $2 million of marketing expense originally planned for Q2 that is now planned for Q3.

We also benefited in Q2 from an earn-out from one of our investments of $600,000 reported as non-operating income.

Relative to tax, I'm excited to announce that we implemented a UK principal company effective July 1, 2018.

As you may recall, at our 2015 Analyst Day, I said we were exploring the establishment of an international principal company to optimize our global operating structure.

We recognized then that our U.S.-based operating model would likely change as we increased our international business operations.

During 2016, we assessed various operating models and set plans in place to establish our global operating structure centered around our London office.

This entity will now be responsible for managing all of our EMEA-direct business, and as such some of our operating profit will shift into the UK, which currently imposes a lower statutory tax rate than the U.S. Our 2018 tax rate was already benefiting from the U.S. tax reform, and now we are further benefiting from this operational change.

Our Q2 actual effective tax rate before discrete items was 16% versus our Q1 rate of 27%, benefiting from the implementation of our new UK structure, regional profit shifts, and a higher estimated R&D tax credit.

Additionally, we booked a discrete charge for exit charges associated with the establishment of our UK principal company, partially offset by a small discrete stock compensation windfall benefit, resulting in an overall effective tax rate for Q2 of 30% versus 21% last quarter.

Our second half and full-year ETR before discrete items is expected to be 24% to 26%.

We expect to more fully benefit from the UK principal in 2019 and beyond.

Comparing Q2 2018 EPS to Q2 2017 EPS is challenging.

The Q2 2018 EPS of $0.37 was a $0.10 increase over Q2 2017.

Key items in the Q2 2017 EPS were the $7 million of acquisition-related adjustments for SODC, which had a negative $0.16 impact, stock compensation windfalls, which had a benefit of [$0.15], and the higher effective tax rate before discrete items of 32%, which was prior to the U.S. tax reform.

We have included a schedule to the financial statements providing the actual first and second quarter 2018 tax rates before and after discrete items and the anticipated effective tax rates for the third and fourth quarters, as well as full-year 2018 before discrete items for your reference.

We can't forecast the discrete items, so we don't provide any estimates for those in the schedule.

We ended Q2 with $127 million in cash, down from $184 million at the end of Q1.

During the quarter, we completed the previously announced stock repurchase program, authorizing the purchase of up to $50 million of our common stock.

The plan became effective on March 28 and we repurchased a total of 800,000 shares at an average price of $62.57.

In addition, we announced an increase to our revolving credit facility to $150 million and an extension of the term to 2023.

This facility gives us additional flexibility to execute on our growth strategy.

We expect to end the year with approximately $150 million of cash.

Q2 ending inventory was $115 million or 97 days as expected, compared with $83 million or 81 days last year, driven by the acquisitions and the need to hold inventory for direct-to-retail sales in Japan and more than 50% of EMEA.

As we have previously said, due to the structural change in our business model, we expect DII to be approximately 100 days, plus or minus, on average in 2018, with our typical quarterly fluctuations.

In Q3, we expect inventory to increase well over the 100-day average in support of the Q4 holiday season before declining well below 100 days by year-end.

We continue working to identify operating efficiencies to improve working capital and reduce DII over time.

Now I'd like to provide you with additional detail and some of the underlying assumptions for our full-year 2018 financial expectations.

As we have previously discussed, we manage our business from a full-year perspective.

Likewise, our 2018 financial expectations should be viewed on a full-year basis, as quarterly year-over-year revenue growth rates and overall results will vary greatly by region due to a number of factors, including the impact of the acquisitions we made in 2017.

For 2018, we now expect full-year revenue of $1,000,000,060 to $1,000,000,080, with roughly 60% of the year's revenue being delivered in the second half of the year.

In addition to the traditional second half seasonality of the business, 2018 will be positively impacted by the inclusion of revenue uplift from the European acquisition through the third quarter.

Our revenue expectation range contemplates yen and euro exchange rates roughly in line with current rates, plus or minus 5%.

We continue to expect double-digit year-over-year growth in each quarter and for revenue to increase sequentially throughout 2018, with the highest year-over-year growth rate in Q1.

The smallest sequential quarterly increase is expected between the first and second quarters, and the largest sequential quarterly increase between Q3 and Q4.

In the third quarter, sales and marketing as a percent of revenue is expected to decline from Q2, but not as significantly as in prior years due to our third quarter product launches and the expense that moved from Q2 into Q3.

We still expect full-year R&D of 13% of revenue, but Q3, like Q2, will be closer to 15% of revenue before declining as a percent of revenue in the fourth quarter.

Profitability will be the lowest in Q2, driven by our sales and marketing expense increase over Q1, as is typical, to support the Q2 selling season.

Q3 profitability will be higher than Q2, but only modestly.

For the full year, we continue to expect operating expenses to total roughly 42% of revenue.

Our full-year guidance still assumes reinvesting the incremental gross margin provided by our cord integration primarily into higher sales and marketing spend.

1/3 of the incremental spend is a full-year of expense associated with our two acquisitions, and the remaining 2/3 from marketing expenses associated with our 2018 product launches and continued investments in the Roomba and Braava awareness campaigns to drive continued worldwide household adoption.

As a result of all of these items, we now expect full-year operating income of $90 million to $96 million and EPS of $2.30 to $2.50.

We are very pleased with the first half results and our ability to increase our expectations for the year.

I am also very excited about having delivered an improved global operating structure that will benefit iRobot and its shareholders in the future.

I'll now turn the call back to Colin.

Thank you, Alison.

We have delivered the strong first half we expected, and I'm very excited about the rest of the year ahead.

In the second half, we will launch products that will reinforce our leadership in consumer robotics.

With a positive initial ruling from the ITC, we have validated the strength of our technology and IP portfolios, which will further widen our competitive (inaudible), and we will leverage the Braava family growth momentum of the first half to build the category and further develop our revenue diversification.

We believe that consistent execution of this strategy is the most effective way to drive sustainable growth and shareholder value.

With that, we will take your questions.

(Operator Instructions) Our first question comes from Frank Camma of Sidoti.

Colin, I know you don't want to give any details about the product launch or about the product itself, but could you give any color on the way you're launching it?

You mentioned -- the next one, I'm talking about, the first one.

You said limited stores.

Is that to give your partners some exclusivity, or is that to gauge demand?

So by soft launch, we mean we're not having a big singular event, but we're rolling it out in a controlled way with our partners.

And I think that that's the color I'd like to give.

But will it be available on your online store as well, or is it just in bricks-and-mortar, I guess?

That's really where I'm going with the question.

Frank, it will be available in limited stores in August, and then it'll further roll out after that point elsewhere.

And no, there will not be a press release.

There won't be a press release, okay.

But you do have sales built into your guidance for both products.

Absolutely.

Yep.

And further on to that, I was surprised you didn't call out any incremental build for inventory for the new products.

Don't you have to stock up on them before you launch, or is that more of a Q3 event, like as far as the inventory goes?

Yeah, as I indicated in the call, Frank, we do expect inventory to be higher at the end of Q3 as we support the second half of the year, but remember we do have significant quarterly fluctuations anyway in our business.

Okay, so it's not that -- even though it's two new products, it's just not material enough given the seasonality already.

Okay.

That's not the right interpretation, and we are building up our inventory in Q3, and the new products do count in that inventory.

So that, again, our inventory is ramping, as is normal for that time.

We're just shifting what we're building from the older models to the newer models.

And so while that inventory ramp might not look extraordinarily different from prior years, the makeup of that inventory build will be.

Okay, I got it.

That's clearer.

I got it.

And then my last question, just in the script, and I don't want to read into this too much, but the way Alison described sort of the revenue from the 900 and 600 series of being sort of the drivers, is that a little different from the past, given that -- so I guess the question, really where I'm going with this, is what makes a consumer -- I can see why consumers would choose a 600 series given the price point, but why are you finding that consumers are going all the way up to the 900, given the price point?

I mean, it's much more expensive than the middle series.

What does your research say, that somebody will choose that 900 series over the 800 series?

I know it's got more bells and whistles and stuff, but it seems like it's incrementally more to the consumer.

So Frank, in my comments, that mix between 900 and 600 really has been consistent with what we've seen the last several quarters, so as we've talked about for a while, we see a strong amount of demand at the 900 end and the 600 end, and that is -- been very consistent over the last couple of quarters, so there was nothing really new in how that revenue split between those two models.

I think, to put a little color, the Prime Day success demonstrated continued strength down at the entry-level price points, which I think is very material in that we are not simply getting pushed into premium only, we continue to have strong performance across price points, and I think that that's a very material result and probably the most important message coming out of Prime Day.

Got it.

And that, in fact, you can obviously be competitive at that low end and still maintain margins, right?

Correct.

Our next question comes from Troy Jensen of Piper.

Couple questions from me.

First on ASPs, it looked like they were down on a sequential and year-over-year basis.

I guess my assumption is, is potentially promotional activity kind of fleshing out some of the older products?

But would just love to know your thoughts on the ASP trends here.

Yeah, I mean the gross ASPs were actually up, net were down a little.

Part of that is driven by normal promotional activities.

There really wasn't anything unusual.

In fact, we had a little bit of pickup in Q2 as some of the promotional activities we had planned to run weren't as extensive as we had originally planned.

So I don't think there's really anything of particular note.

We've got some typical fluctuations in there in the ASPs.

Okay, perfect.

And then, Alison, for you, I think in Q1, you told us that the top five resellers in the U.S., a sell-through of about 29%.

I'd be curious to know what that was in Q2.

Our sell-through varied, obviously, by the different countries and regions.

I'd say it was very much in line with our sell-in rates for the quarter.

Okay, so I guess my other -- you may have just answered my next question, but channel inventory with the big retailers, I guess my assumption, it would be lower than normal given they're likely to know these new products are coming out, but it sounds like your comment about sales in and sales out being kind of the same implies that there's been no change in channel inventory.

Yeah, I'd say the channel inventory levels are where they normally are.

I don't think they've changed dramatically from where they were at the end of Q1 or last year.

Okay, I'll get one last question and I'll cede the floor here.

But for Colin, given the success you've had with this IP litigation, do you plan on targeting some of your larger competitors now?

Great question.

I think that we are considering our options very seriously as to what our next steps are.

We're, as I said, very pleased with this outcome as it validates the strength of our IP, and so definitely we have very attractive options as we think about what's next.

I can't really go further than that.

Our next question comes from Asiya Merchant of Citigroup.

This is Michael Cadiz for Asiya Merchant at Citi.

I just have one question.

Given that recent news that tariffs would most likely affect Roomba, would iRobot consider passing on any tariffs to the consumer, and how do you think this would most likely affect demand?

Sure.

I mean, there's a lot of remaining questions as to exactly what's going to happen and how we will react, and so that at this point we're looking at alternatives.

Realistically, either we pass it through or find a way to mitigate the impact, and we have not determined our strategy at this time.

So it's getting a lot of attention.

Our next question comes from Jim Ricchiuti of Needham & Company.

The growth in Braava in the international market was very strong, Colin.

Can you talk a little bit about whether that was balanced in both EMEA and Japan, or was it skewed more toward Asia?

We had growth in both Europe and Asia in Q2 for Braava, Jim.

Maybe a little bit more towards Asia, but it was representative across both regions.

And Alison, can you say what the growth was in Braava in the U.S. market?

Give me one second.

And then while you're looking for that, the other question I had was just with respect to the new products, you talked before about 20% to 25% contribution in 2018.

Is that still your expectation?

Sorry, Jim, just to get back on your original question, so our single-digit growth in Braava in Q2 in the Americas -- I'm sorry, and your second question was?

Yeah, the second question was just on the new products.

I think you've talked about 20% to 25% revenue contribution in '18.

Yep.

That's still the expectation?

Yes, that's still the expectation.

And I know you can't say a whole lot, but if we think about Q4, just trying to get a sense how widely available they're going to be.

Are these products going to be available in both domestic and international in Q4, or will there be more of a phased rollout?

There will definitely be a phased rollout.

We'll talk more about that as those products are launched, but as is typical, we won't be launching everything everywhere right from the start.

Okay, and final question for me is just, I wonder if you could say anything about with respect to your major retail customers, how they're thinking about the second half, the holiday season relative to maybe expectations last year at this time, mainly in light of just what clearly is an improved macro environment for consumers.

I would say that there is, at a very high level, continued enthusiasm about the category.

They've seen strong category growth continue in the first half of 2018, and again buoyed by iRobot doubling its sell-through on Prime Day, it's another bit of evidence that the market for robot vacuuming continues to grow strongly, and so that the attention and the enthusiasm they're placing against the category remains high.

Our next question comes from Mike Latimore of Northland Capital.

On the Prime Day success, I guess last year or just in the U.S. this year more broadly, does the success on Prime Day in Europe and Japan kind of have similar carry-through across those regions, across other retailers, as you saw in the U.S.?

I would say that it's a little early.

It's the first year that we actively participated in Prime Day outside of the U.S., and we had very good results, but we really don't have anything to benchmark our performance against.

And so we're very happy but I would be speculating to say whether or not people were paying attention and what effect it would be.

But it's certainly a great result we're proud of.

Great.

And then you talk about this first new product release having high-end features at a lower price.

In the past, what would have you defined as sort of high-end features?

We consistently try to bring features into robots that really are game-changing at the high end, and then have worked to bring those features down.

We brought Wi-Fi in at the high-end feature back a few generations ago, and now it's available all the way through our product line.

That would be one example.

We have the Dirt Detect technology throughout our entire line.

And so that without getting into the specifics of what we're doing, if you looked at some of the truly differentiating things we have in our products that our customers very much value, those are the candidates for cost reduction and pulling it down through the product line.

So I think that's where I'll stop on that one.

And just last one.

Any general guidance on just legal costs in the second half of the year?

I think that nothing out of the ordinary that's a change from what we've suggested.

Certainly, the litigation is not over, but it's winding down.

Our next question comes from Mark Strouse of JP Morgan.

So Colin, on the last call you mentioned more Chinese competitors coming into the shelves in Europe, and mentioned that that could potentially happen in the U.S. this year.

Reading your comments today, maybe I'm just reading too much into it, but when you talk about the competitive environment looking similar to last year, to me that sounds a bit more optimistic than last call.

First of all, am I reading that right, and if so, can you just talk about what has changed?

I think that we certainly saw an influx of new competitors in Q4, and it did change the dynamics in the marketplace somewhat.

We've seen that stabilize through the first half, and I think that, again, you look at Prime Day as an exemplar of, okay, there is significant availability of these low-end models.

How much share did they take, how much did they disrupt, and what did they do in Prime Day.

And the results are very favorable relative to iRobot demonstrating its ability to maintain strong performance at the low end of the marketplace while we maintain our strong performance at the high end.

And so that I think that there's a very good indication that, yes, the market was somewhat disrupted at the end of Q4, that the market growth rates certainly allowed for them to enter without substantially impacting the performance we had predicted and laid out, and the environment, while certainly very competitive, is not demonstrating new disruptions.

Got it.

Yeah, that makes sense.

Okay, thank you.

And then, Alison, just to follow up on the tariffs.

So I can appreciate you're still evaluating your options of mitigation, but just from a very high level, the math I'm kind of running through about a potential impact of a 10% tariff on your U.S. business, with the mix of your U.S. business and your overall gross margins, I'm kind of getting into like a 2% to 3% impact to gross margin.

Worst case scenario, understand again that you can try and mitigate that.

But is that kind of the way that you guys are thinking about it, or anything that you can do there to help?

Yeah, I'm not going to speculate as to the financial impact of those tariff changes.

As we've said, our guidance assumes no impact of the tariffs at this point.

As we learn more and understand if this is really something we're going to be dealing with, we'll do that assessment and share it with you guys.

But I'm not going to speculate about the potential impact of that now.

And then just real quick last one.

So we saw a trademark application for something called Imprint.

I guess, without asking specifically about these products coming up in the second half, is that something that we could see sooner rather than later, or is that more of kind of a long-term play that you're going for there?

We routinely trademark names that we think could be associated with features, and so that certainly, we think that's a very interesting name and have taken that -- worked to protect that particular term.

So I will leave the speculation to you as to how and when that might be used, but I would say that it's a term we like.

Our next question comes from Bill Baker of Garp Research.

My question is regarding really the second half in terms of the productivity of the marketing (inaudible) spend, and I'm curious.

It looks to me like right now, without a new product introduction, you're growing 15% in the States, and now when you look at the sales gains you're forecasting, you're moderately above that growth rate.

The company as a whole (inaudible) nice, but you look like you're ramping up marketing as a percentage of revenue, so I'm not sure about the productivity of that and whether you're just being conservative, and maybe you could tell us how you're changing the marketing, as it's changing as you work with Amazon (inaudible) approach and been more or less productive.

Maybe just shed more light.

Sure.

I think that -- a lot to unpack in your question.

There's definitely, first off, a cyclical nature to our marketing spend, where we generate momentum in Q3 and Q4, which drives our sales.

That momentum carries into Q1, quite nicely driving sales, and then we [goose] it a little bit around Mother's Day, Father's Day, with incremental spending in Q2, which we did successfully before, going quieter on air in the -- over the summer and then ramping up again, so that there's sort of a natural way that we go and apply demand generation dollars to our marketing flywheel that is not peanut butter across the year.

It has a definite cyclicality, and the entire year benefits from those injections of energy, and so that there's a pattern to it.

A second level of analysis on our market spend goes against what is our marketing efficiency, so what is the ROI on every incremental dollar spent.

And that has a lot to do with the maturity of the product in a particular region.

And so as an example, sort of in a big picture example, we are more efficient spending dollars against demand generation on Roomba in the United States than we were a year ago, and much more efficient than two years ago.

But we're a global company, and so that there are regions that were less developed and less mature, frankly particularly in Europe and Japan, where you would say our ROI, while still positive, is not as strongly positive as it is in the U.S.

And so that as U.S. gets better and unlocks resources and opportunities to develop rest of the world, and then on top of that, as we invest in building a second material leg on the stool, we're spending money against building Braava, and so that we have a -- think of it as a giant spreadsheet of optimization, where we commit to our profitability targets and then within those commitments are moving marketing dollars around to develop emerging regions and develop the Braava category so that we can sustain the growth that we're committed to with our long-term financial model over time.

And so we have a very quantitative approach to marketing spend, and winning and increasingly making our marketing dollars work for Roomba in the U.S. creates opportunities to drive growth and diversification of product lines globally.

Is it still primarily television-oriented, and are you challenged by people not watching (inaudible)?

I would say that there's a high-level shift from -- a slow shift from TV to online and digital marketing.

We still, probably our biggest singular investment is TV, but two years from now that might not be the case.

We're definitely seeing strong performance from online marketing activities.

Our next question comes from Ben Rose of Battle Road Research.

Just some questions for Alison.

Looking at the non-GAAP G&A expense sequentially from Q1 to Q2, which now looks like it's down almost $3 million on a sequential basis, don't recall that kind of decrease seasonally last year.

May I ask, were there things in the March quarter that were one time in nature, or are there other expenses that are declining?

There's not anything really of one time in nature.

Certainly, our legal expenses in Q1 were quite high due to the litigation, but we will have legal expenses later in the year, but that was probably an unusually high quarter for us from a legal perspective.

But nothing else really unique going on in terms of G&A.

With regard to the share-back having been executed, may I ask what the company's appetite is for share repurchases going forward at a minimum to offset options dilution?

With this wrap-up of the program, we will readdress any other actions we want to take.

As we said, at this point we're projecting to end the year with about $150 million in cash, and that doesn't suppose another program.

But now that this program is concluded, we'll take a fresh look and decide whether we want to make any further uses of that cash in terms of share repurchase going forward.

And your comment on the revolver being increased in this past quarter reminds me of the fact that it's been quite a while since iRobot has done any kind of acquisition.

Could you just maybe refresh our memory just in terms of any kind of framework that you use in terms of either the size of the company to be acquired, the appetite for or the limitation for potential dilution, et cetera?

So just on your first point, actually we did two acquisitions last year, so in terms of it being a while, we don't have the same view.

I'm sorry, I meant from a product standpoint, I meant a la evolution robotics (inaudible) acquiring a product company.

Okay, understood.

No, with the increase of the revolver, obviously that gives us more opportunity to think about size of acquisitions.

We don't put a particular threshold on the size of companies that we're looking at, if we think they can be helpful in helping us deliver our strategy, they would be within our area of remit.

You shouldn't expect us to really go completely out of our core in terms of looking for acquisition opportunities.

But the facts and circumstances would be really related to the individual acquisition we're looking at, but certainly the increase of the revolver gives us a little more room to widen the lens a little bit.

Yes, and our criteria is largely unchanged.

Companies that have products that could benefit from our global distribution power and fit within our brand umbrella would be things we would look at, as well as strategic technologies that we think are going to be important in the emerging consumer robot space.

And so those are themes for I think your category of evolution-like acquisitions, and we have a active team that is looking at opportunities.

And again, certainly, it is part of our strategy to continue to look, and when we find high-quality targets, pursue.

Thank you for that color.

I didn't mean to overlook the distributor acquisitions.

No worries.

Sorry.

And then finally, just for Alison, could you just refresh us?

The amortization of intangible assets being $4.7 million in the second quarter, for how much longer does that go out in terms of quarterly impact?

So the largest driver of that amortization is from the Robopolis acquisition, and that's going to be with us through 2019, although the impact to 2019 for a full year will be slightly less than it is here in '18.

In '18, the Robopolis amortization impact is about $50 million evenly split across the four quarters.

Okay.

That concludes our second quarter 2018 earnings call.

We appreciate your support and look forward to talking with you again in October to discuss Q3 results.

That concludes the call.

Participants may now disconnect.