Illumina Inc (ILMN) 2015 Q2 法說會逐字稿

Hello everyone, and welcome to the second-quarter 2015 Illumina Inc.

earnings conference call.

(Operator Instructions)

As a reminder, this call is being recorded for replay purposes.

I would now like to turn the call over to the Vice President of Investor Relations and Treasury, Rebecca Chambers.

Thank you, operator, and good afternoon, everyone.

Welcome to our earnings call for the second quarter of FY15.

During the call today, we will review the financial results released after the close of the market, and offer commentary on our commercial activity, after which we will host a question-and-answer session.

If you have not had a chance to review the earnings release, it can be found in the Investor Relations section of our website at Illumina.com.

Participating for Illumina today will be Jay Flatley, Chief Executive Officer; Francis deSouza, President; and Marc Stapley, Senior Vice President and Chief Financial Officer.

Jay will provide a brief update on the state of our markets, Francis will comment on product performance, and Marc will review our second-quarter financial results, as well as provide updated guidance for 2015.

This call is being recorded, and the audio portion will be archived in the investor section of our website.

It is our intent that all forward-looking statements regarding our expected financial results and commercial activity made during today's call will be protected under the Private Securities Litigation Reform Act of 1995.

Forward-looking statements are subject to risks and uncertainties.

Actual events or results may differ materially from those projected or discussed.

All forward-looking statements are based upon current information available, and Illumina assumes no obligation to update these statements.

To better understand the risks and uncertainties that could cause actual results to differ, we refer you to the documents that Illumina files with the Securities and Exchange Commission, including Illumina's most recent Forms 10-Q and 10-K.

Before I turn the call over to Jay, I would like to let you know that we will participate in the Morgan Stanley healthcare conference in New York the week of September 14.

For those of you unable to attend, we encourage you to listen to the Webcast presentation, which will be available through the Investor Relations section of our website.

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

Thanks, Rebecca, and good afternoon, everyone.

I'm pleased to report that Q2 was another strong quarter for the Company.

Order receipts exceeded our expectations, resulting in a record exiting backlog.

Revenue increased 21% year-over-year to $539 million, slightly above the guidance provided on last quarter's earnings call.

The leverage we delivered this quarter was also remarkable, as non-GAAP earnings per share grew 40%, compared to the second quarter of 2014.

These robust results were due to strength in sequencing revenue and solid operational execution.

Second-quarter total sequencing revenue grew 28% year-over-year, driven by record consumables, and strength in services.

Shipments to clinical and translational customers grew 45% year-over-year to approximately 40% of total shipments, further demonstrating the expanding utility of genomics.

Oncology was a significant driver of this mix, as use of NGS expanded in hospital laboratories and commercial molecular diagnostic settings.

Shipments to all segments of oncology grew approximately 35% year-over-year.

Over the past several quarters, we've shared some of our strategies to penetrate this market, which include our OncoPanel and Liquid Biopsy programs.

We're on track to deliver -- release one of the OncoPanel as an RUO product by the end of the third quarter.

This initial release includes 15 genes critical in oncology, and sets the groundwork for our OncoPanel products that will be CE/IVD-marked and FDA-approved.

We're committed to delivering the enhanced 170-gene version of the OncoPanel to address the needs of the research market early next year.

We will also be providing this product for use by our pharma partners in their clinical trials.

Our Liquid Biopsy program continued to make progress during the second quarter.

As part of this effort, a study was recently published in the Journal of the American Medical Association, which showed specific examples where discordant NIPT results were detecting maternal cancer.

Our development teams are optimizing assays and exploring their clinical power, for use in this exciting frontier of genomics.

Our market development strategies in reproductive health also contributed to the higher mix of clinical and translational customers in the second quarter.

Sales to IVF reference laboratories increased 25% in Q2, with approximately 20% of sales now coming from VeriSeq PGS.

As a reminder, VeriSeq PGS is our sequencing-based pre-implantation genetic screening solution used as laboratories transition from the older generation arrays.

Total NIPT revenue including products, services and test fees grew more than 50% versus the second quarter of last year.

This result was due in part to a doubling of test send-out samples, as well as an increase in product sales to NIPT providers globally.

Our partnership with Berry Genomics, the second-largest NIPT Company in China, is progressing well, after the recent Chinese FDA approval of Berry's assay and the NextSeq CN 500.

This has resulted in an increase of shipments for resale to Chinese hospitals that are bringing NIPT in-house.

During the quarter, we signed a similar deal with Annoroad, the number-three NIPT player in the region, and will jointly bring together Illumina sequencers with Annoroad's assay development capability.

We believe the sustainability of NIPT revenue growth will come in part with the availability of our next-generation solution, VeriSeq NIPT.

We're in the final stages of the European regulatory process for the software, and hope to have CE mark shortly.

We will follow with a CE mark on the VeriSeq NIPT product for distribution as an IVD solution.

During the second quarter, we signed five new partnerships outside the US, and now have a total of 19 VeriSeq NIPT customers working on installing the protocol in their respective laboratories.

Moving now to arrays, we continue to witness growth in sample volumes, but at lower prices.

In the second quarter, Infinium volumes increased 29%, yet total microarray revenue declined 11% year-over-year.

Despite this result, we have positive trends in the array market to share.

First, total genotyping orders were up 16% versus Q2 2014, driven in part by strength in agriculture, due to the continued adoption of genomic selection in livestock, for which our Bovine arrays are well-suited.

For the year, we're projecting arrays to decline in the mid to high single digits compared to the prior year.

We expect the second half to be positively impacted by new custom products that will launch as a result of our increased R&D investment in arrays.

This expanded product line includes an updated ImmunoArray, with an additional 140,000 markers of new content, as well as a DrugDevAarray, which combines available genomic data with advanced computational tools that enable researchers to test targets earlier in the therapeutic development process.

Lastly, I want to highlight our sample-to-answer solutions for forensic and HLA markets, which continue to gain traction.

The number of customers evaluating our TruSight HLA solution has close to doubled since Q1.

We continue to build broad awareness for our MiSeq FGx solution, and recently held a webinar for more than 400 target customers to further educate them on the importance of NGS in the forensic setting.

Encouragingly, we now have at least one major law enforcement customer in every region of the world, an important first step to broader market adoption.

Overall, Q2 was another solid quarter for the Company.

Our momentum continues, driving demand across our diverse customer base.

As we move into the second half of 2015 and beyond, our relentless focus on market development and innovation in our sequencing portfolio will be critical factors in our ability to penetrate the large and untapped opportunities ahead of us.

I'll now turn the call over to Francis, who will provide a detailed overview of our product results.

Thanks, Jay.

And good afternoon, everyone.

I'm pleased to provide an update on the performance of each of our product families in the second quarter.

As Jay mentioned, total sequencing revenue grew 28% over the second quarter of 2014.

This included sequencing instrument growth of 13% year-over-year due to demand for HiSeq and NextSeq, offset in part by lower HiSeq shipments.

Again this quarter, we saw strong interest from new to Illumina next gen sequencing customers, who accounted for more than 45% of instruments ordered.

This class of users is primarily interested in gaining exposure to sequencing through the MiSeq and NextSeq bench top platforms.

With this in mind, we have introduced a MiSeq promotion called new to NGS.

This starter bundle includes training, sample prep and core consumables, as well as the instrument and service plan needed to get a new sequencing lab up and running.

We expect this to not only build the funnel of new customers, but also promote our entire sample-to-answer MiSeq solution in a cost-effective manner.

Order dynamics for our bench top instruments are largely unchanged from the trends seen in prior quarters.

Across this portion of our portfolio, more than 45% of instruments ordered came from clinical and translational centers.

As a result of these factors, MiSeq and MiSeq DX sales in the quarter continued to be strong.

Rounding out our benchtop franchise is NextSeq, which saw 16% sequential increase in shipments.

Development activities in reproductive health and oncology markets drove demand, which together accounted for close to 40% of quarterly NextSeq orders.

Oncology customers grew close to 50% compared to the prior year, as we saw a large multi-unit order from a Liquid Biopsy company, as well as orders from customers focused in gene panels.

Reproductive health customers also bolstered demand, with the June availability of NextSeq 550, which enables both sequencing applications and microarray scanning for cytogenetic testing on a single platform.

These data points are a strong indicator of broader adoption across a diverse set of customers.

We have made great progress improving the success rates on NeoPrep, and as a result, the product is now in full commercial shipment with the TruSeq Nano prep kit.

NeoPrep is very well matched to the benchtop sequencer market, and we're pleased with the feedback received from early access sites on the instrument's performance and reliability.

We plan to release the TruSeq mRNA kit shortly, and are actively working on building out the application menu for the system to further drive interest in this platform.

Moving now to our high throughput sequencing instruments, in the second quarter, demand for the HiSeq family of products profited from the impact of HiSeq 4000 and the HiSeq X Five.

As expected, shipments of HiSeq X instruments were lower sequentially and year-over-year.

We continue to expect HiSeq X orders and shipments to remain lumpy quarter to quarter, and dependent on the timing of funding and customer readiness.

Interest in the X platform remains high, as we continue to add new customers in Q2, raising the total to 22, as well as shipped systems to existing customers to increase their capacity.

Exiting the quarter, our HiSeq X order funnel remains healthy, resulting in an expected quarterly order rate of 20 to 30 HiSeq X units.

We have recently introduced new products and services to further improve the performance of HiSeq X. This includes the launch of an improved V2.5 flow cell, which has been made broadly available to our installed base, as well as the associated software and firmware updates.

We also launched Illumina SeqLab, an integrated solution to help our customers maximize quality and throughput of their sequencing facilities.

Through a partnership with GenoLogics and Hamilton Robotics, the Illumina SeqLab is designed to standardize and optimize the whole human genome sequencing workflow, reducing the time lab personnel need to spend operating the instrument, and ensuring quality results.

This solution includes the LIMS system, liquid handling, robotics, data analysis, and the consulting services a lab would require to get up and running as an X customer.

We believe the addition of these offerings will increase the consumable utilization of our customers, by moving them more quickly up the adoption curve.

Momentum grew through the first half for the expanded HiSeq family of instruments, including the HiSeq 3000 and 4000.

As expected, we saw an increasing fraction of orders migrate from the HiSeq 2500 to the HiSeq 4000, which just one quarter after launch, became clearly the leading HiSeq platform sold.

Interest in this instrument fueled higher shipments, both sequentially and year-over-year.

Demand was driven in part by multi-unit orders from genome centers, and commercial customers standardizing on the platform.

We remain excited about this expanded portfolio, given it offers tremendous flexibility to address various budgets and scientific priorities.

Moving now to an update on informatics, I am pleased to share that we recently hired a new general manager of our Enterprise Informatics business unit, Sanjay Chikarmane.

Sanjay joins us from SAP, and brings with him over 20 years of experience in enterprise software.

I want to thank Nick Naclerio for the terrific work he did as the GM of this group, and we look forward to him being able to focus full-time on our opportunities in corporate and venture development.

We saw meaningful progress in the adoption of BaseSpace during the second quarter.

We now have 65 applications available, and app launches increased 45%, compared to the first quarter.

We saw a 17% increase in runs uploaded sequentially, and close to 50% of our accounts are current users of BaseSpace.

We continue to focus on more fully integrating our data solutions to further build the informatics ecosystem surrounding the Illumina portfolio of sequencers.

I'll now turn the call over to Marc who will provide a detailed overview of our second-quarter results.

Thanks, Francis.

As Jay mentioned, total revenue grew 21% year-over-year, slightly exceeding our guidance.

On a constant currency basis, revenue grew 25%, bolstered by incoming orders that exceeded our projections, resulting in an impressive book-to-bill ratio of 1.1.

This strong order trend will benefit shipments over the coming periods, as the vast majority of the backlog is shippable within the next four quarters.

Shipments in the Americas grew 32% year-over-year, due to strength in HiSeq instruments and HiSeq X consumables.

European and APAC shipments increased 16% and 2% respectively over the same period, as growth in these two regions was muted, due to the impact of currency.

APAC shipments declined $20 million sequentially as expected, due primarily to continued weakness in Japan, as funding remained constrained.

We expect a very slight improvement in the Japanese funding environment in the second half of this year, returning to normalized levels in 2016.

Instrument revenue grew 11% year-over-year to $155 million in the second quarter, particularly encouraging given the strong Q2 2014 comparison.

This increase was primarily driven by demand for NextSeq and HiSeq.

Overall, we are very pleased with the HiSeq family of products, as the lost sequential HiSeq X shipments were more than offset by the increased interest in the HiSeq 4000.

Overall consumable revenue in the quarter was $303 million, an increase of 23% compared to the second quarter of 2014, as higher demand for sequencing consumables and library prep was partially offset by a decline in arrays.

Consumable revenue represented 56% of total revenue, up slightly from 55% in the prior-year period, and lower than the 57% we saw in Q1.

Sequencing consumable revenue was very strong, growing 36% over Q2 of last year.

HiSeq X and NextSeq utilization exceeded the respective guidance ranges we have provided.

Quarterly shipments of NextSeq consumables were strengthened in part by increased adoption of the enhanced v2 reagents, which became available late in the first quarter.

During the quarter, 5 of the 22 HiSeq X customers accounted for approximately 70% of X consumable shipments.

As a result, we continue to see meaningful opportunity to raise pull-through of our lower utilization customers.

With the recent introduction of the Illumina SeqLab, we should be able to accelerate new customers' ability to implement large-scale sequencing workflows and increase pull-through as a result.

MiSeq utilization was in our projected range of $40,000 to $45,000, and HiSeq pull-through per instrument, excluding HiSeq X, was in our projected range of $300,000 to $350,000.

Services and other revenue, which includes genotyping and sequencing services, as well as instrument maintenance contracts, grew 35% versus Q2 2014 to $77 million.

This improvement was driven by growth in NIPT services, which benefited from increased test send out revenue, genotyping services, and extended maintenance contracts associated with our larger sequencing installed base.

Turning now to gross margin and operating expenses, I will highlight our adjusted non-GAAP results, which exclude non-cash stock compensation expense and other items.

I encourage you to review the GAAP reconciliation of non-GAAP measures included in today's earnings release.

Our adjusted gross margin for the second quarter was 72.4%, an increase compared to 72.2% in the first quarter.

Year-over-year adjusted gross margin expanded 150 basis points due to efficiencies in leverage and manufacturing operations, which were offset in part by currency headwinds.

Adjusted research and development expenses for the quarter were $85 million or 15.8% of revenue, higher compared to $80 million or 14.9% of revenue in the first quarter, due to the impact of additional headcount and outside services expense.

Adjusted SG&A expenses for the quarter were $103 million, or 19.2% of revenue, an increase compared to $97 million or 18% of revenue in the previous quarter, due to higher headcount and outside services associated with our ERP program.

As we have shared previously, as an organization, we have spent a tremendous amount of time and effort preparing to implement a new ERP solution, SAP, which we expect to go live next week.

To date, this project has spanned three years, and will, we believe, enable us to scale our operations more efficiently and effectively, to support long-term growth.

While this launch represents our single largest phase, we will continue to invest over the next few years in similar projects that enhance our ability to scale.

Returning to our results, adjusted operating margins were 37.4% compared to 39.3% in the first quarter, lower sequentially as we added 315 net hires in Q2.

Operating margin was higher compared to the 34.8% reported in the second quarter of last year, due to the impact of improved gross margins and operating expense leverage.

In the second quarter, our stock-based compensation equaled $33 million, up slightly sequentially.

For the remainder of 2015, we expect stock-based compensation expense to increase from the levels seen in the first half, due primarily to the impact of our annual grants and growing employee base.

Our non-GAAP tax rate for the quarter was 28.9%, compared to 29.8% in the second quarter of last year, due primarily to the benefit of increased manufacturing in Singapore.

Non-GAAP net income was $120 million for Q2, and non-GAAP EPS was $0.80.

This compares to non-GAAP net income and EPS of $85 million and $0.57 respectively in the second quarter of 2014.

The impact of foreign exchange lowered Q2 non-GAAP EPS by approximately $0.06 relative to last year.

We reported GAAP net income of $102 million, or $0.69 per diluted share in the second quarter, compared to net income of $47 million, or $0.31 per diluted share in the prior-year period.

Cash flow from operations equaled $171 million.

DSO increased to 62 days compared to 59 days last quarter, and inventory increased 9% in part to prepare for our systems transition.

Capital expenditures in Q2 were $41 million, due to spending associated with our ERP implementation, as well as other ongoing investments focused on scaling the organization, resulting in $130 million of free cash flow.

During the quarter, our Board of Directors authorized a new 10b5-1 share repurchase program of $150 million, which we will use to manage dilution from employee grants.

In addition, we have $96 million remaining under our previously-announced discretionary buyback program.

We ended the quarter with $1.5 billion in cash and short-term investments.

Turning now to our expectations for the remainder of 2015, we continue to project approximately 20% total Company revenue growth and 23% growth on a constant currency basis, based on current exchange rates.

We have increased our 2015 non-GAAP EPS projection to $3.39 to $3.45, up from $3.36 to $3.42.

Additionally, we continue to project a full year pro forma tax rate of 27%, which assumes 2015 federal R&D tax credit and other tax extenders are passed prior to our fiscal year end.

For modeling purposes, we feel it is appropriate to assume a Q3 tax rate of 29% given the expectation that the R&D tax credit will be passed in Q4, and at that time, we will record the full-year impact.

In conclusion, I would like to punctuate a number of key fundamental performance indicators related to the second-quarter results.

Firstly, robust instrument revenue growth driven by strong HiSeq and NextSeq shipments.

Secondly, sequencing consumables growth of 36% year-over-year, our eighth consecutive quarter of growth in excess of 30%.

An order quarter which exceeded our expectations, resulting in increased backlog across each major line of business, and our highest-ever exiting backlog.

And finally, significant outperformance in our Americas region, allowing us to manage through the impact of currency, as well as a weak funding environment in Japan.

We believe these fundamentals, combined with our technology leadership, new product introductions, and a focus on market expansion will enable us to continue to catalyze the broad adoption of genomics in 2015 and beyond.

Thank you for your time.

We will now move to the Q&A session.

To allow full participation please ask one question and rejoin the queue if you have additional questions.

Operator, we'll now open the line.

Our first question comes from the line of Tycho Peterson with JPMorgan.

Thanks, actually, Jay, I want to see if you can elaborate a little on your comments on the array market.

If we think about this, you stepped up your R&D there.

You talked about biobank contracts, and you did do the NeoGen agreement.

I'm just curious as to why you think that market's going to be down mid to high single digits?

The comps are tough on the array market overall, and a lot of it has to do with pricing, Tycho.

Our unit volumes continue to increase in general in the array business, but pricing is becoming increasingly aggressive there.

The biobanks to do the large scale projects that they want to do are only enabled at pretty aggressive price points.

So I think it's that combination that's causing us to predict the array business to be down overall.

We had a good order quarter in arrays but a lot of that is going to be in backlog and shipped out over the next two to four quarters.

Okay.

And then if we think about swing factors for the back half of the year, and think that's good upside on the revenue line.

You've got population sequencing, NIPT.

On the 3000, 4000 uptake, that's another thing that potentially could accelerate quite a bit.

Maybe just talk a little about the dynamics of the 2500 customers because you have rapid run mode on the 2500, but not 3000, 4000.

What percentage of those customers do you think pulled out, and what are you telling customers on 2500, in terms of phasing it out overall?

Are you making any commitment to servicing it for the next year, or what are those discussions like?

We have a long-term commitment to servicing any instrument that we put in the field, and we typically will do that for five to seven years after we stop shipping it, so even after we declare something no longer available for sale, we'll continue to support it for a long time.

So there's no concern there.

We did see in the quarter a great shift toward the 4000, which is what we predicted would happen.

The performance of that instrument, the price per base, the patterned flow cell road map are all important ingredients for people to move to the 4000.

You could almost think of it as a slightly derated X that's allowed to run all applications.

So that was, we think, largely responsible for that shift.

We do think over the next year to two, that people buying the 2500s will gradually decrease, and it will, in some sense, obsolete itself.

But the one feature that we have on the 2500 that customers like a lot is the rapid run feature, so we're taking a look at whether at some point in time we want to introduce that on the 4000.

But right now, we're not actively working on that feature.

Okay.

Thanks.

I'll hop back.

Our next question comes from the line of Doug Schenkel from Cowen and Company.

So I think the first thing I want to do is just try to align order and placement commentary with the Q2 revenue model.

Were there any timing dynamics that impacted the pacing of placements, such as, but not limited to funding?

Really, what I'm getting at is, were there some issues where customers wanted instruments, but maybe weren't able to pay for them, based on the timing of when NIH funding becomes more available?

Related to that, the 20 to 30 instrument order rate guidance that you provided on a quarterly basis, I think that's new.

You talked about shipment timing lumpiness.

Would you be willing to say if you were in the 20 to 30 placement range this quarter?

Yes, I'd say we were in that ballpark during the quarter, Doug, and we think that's going to be the rough range.

Some quarters will be a bit higher, some might be a bit lower, and that's why we made the lumpy comment in the script.

In terms of customer buying patterns, customers always want to buy more sequencers, if they had infinite money.

So it is the case that people who don't have money would want to get a sequencer earlier if they could.

But that's not a unique thing to the second quarter.

I think what we saw in the second quarter was very strong orders, but some labs not ready to take delivery.

That was true in fact in some Xs, where we received orders for complete X systems, X Tens, in fact, but the labs aren't ready to take those instruments.

And so some of those shipments will be spread out over the next two to four quarters.

That was true in the array business, as well.

We got some large array contracts that will ship over time, and I think that's why you see a pretty big gap between the order receipt rate and where the revenue wound up.

Okay.

And then the follow-up is really just sort of a simple guidance question.

If you beat your internal expectations for Q2, if order backlog is now at a record level, and if momentum built over the course of the quarter, I'm just curious, if you contemplated increasing full-year guidance and if not, why.

Thank you.

Well, we always contemplate what we do with the guidance, keeping it the same, raising it or lowering it, based on what our expectations are.

We do have a rich backlog, as we cited in the script, but we don't know exactly when these labs are going to be ready to take delivery, and customers move their delivery times in and out all the time, every quarter.

So we do have a little bit of uncertainty in the exact timing of shipments of parts of the backlog.

Having said that, we think the fundamentals of the business are strong, with the orders that we received.

We were really, really pleased, and as I said in the script, we exceeded our order plan during the quarter, and we're really pleased with that outcome.

Okay.

Thank you.

Our next question comes from Derik de Bruin from Bank of America.

Just a couple of quick ones.

Can you give us specific backlog dollar number?

We didn't.

Okay.

You said 1.1, though, is that what I heard?

Yes.

That was book-to-bill.

Book-to-bill.

Got you.

Right.

That's right.

So have you seen any issues in terms of any changes in the Chinese market ordering, and does some of the economic stuff going on right now impact how you're looking at that market going forward?

And then I have one quick follow-up after that.

No, I don't say -- we haven't seen any real changes in the Chinese market over the last couple of quarters.

The research funding was better in the Chinese market maybe 18 months ago than it has been over the past year.

That's been masked a bit in China, particularly in 2014, because we have such a large number of X placements that we didn't expect in China, so our performance was quite good in 2014 in China.

This year, it won't be quite that good, because we'll have fewer X shipments into China.

Great, that's helpful, and then just one quick one.

On the JAMA paper that you just did on the aneuploidy detection in pregnant women, and funding that, I guess, besides looking at chromosomal aneuploidy, what are you looking at in terms of either POI mutations or small INDELs, stuff like that?

Is your LiquidBiopsy technology capable of addressing those types of changes, and assaying those?

Sure, the question in liquid biopsy is what is the right assay method or combination of assay methods, as you know.

Aneuploidy will pick up some types of cancers.

Nobody quite knows yet the fundamentals of the biology, so we don't know exactly which ones will be picked up by that method versus POI mutations, versus fusions, versus methylations.

There's many things we're exploring there, and I think lots of companies are working on this to determine what the best methods might be, and I suspect the market's going to wind up with a whole host of different tests that will be complementary to each other.

Thank you.

Our next question comes from the line of Jon Groberg from UBS.

So Jay, if you were someone sitting from the outside, their first kind of look at this might be revenue relative to at least Street expectations has been -- the gap there has been narrowing, you have really tough comps in the second half, no real new instrument launch to speak of.

Can you maybe just -- I know you talked about your order growth and how excited you were.

Could you maybe talk about your comfort in the visibility that you have in the second half, and that you have with the guidance that you've outlined for the rest of the year?

And then as a follow-up to Marc, could you maybe explain a little bit -- you mentioned the inventory was up because of the ERP that was going to go live.

Can you maybe talk a little about the receivables, as well?

Thanks.

Yes, Jon.

I think our visibility is about as good as it has been, historically.

I wouldn't say we have a lot more or a lot less than we've typically had.

I think one key factor we're dealing with is the weakness in the Japanese market.

In Japan, even though it was weak, Q1 was an okay Japan quarter, and Q2 is always the worst Japan quarter, so certainly, that was a factor in the revenue in Q2, and we caught up on the backlog of HiSeq X shipments.

So now we're really in very much a steady state, in terms of where we are, in terms of balance between orders and shipments.

And I think that steady state gives us pretty good confidence going into the back half of the year.

The sales force is feeling really good about the overall momentum of the business, and a lot of the new hires that we're bringing in are in the commercial part of the organization, because we do feel very strongly that there's lots more business out there to be had, and as we expand our commercial reach, that we'll be able to incrementally bring in more revenue with the existing product lines.

We are continuing to launch new products, even though there's -- we haven't preannounced any new systems.

We have lots of products in the pipeline that will be announced virtually every quarter from here through next year.

And then Jon, on your question about inventory on consumables, you're correct, inventory, we talked about being up 9% and it was related to ERP implementation and our building of safety stock, to ensure a smooth transition there.

On the receivables, the DSO was up slightly from 59 days to 62 days, just based purely to the timing of shipments being more back-end loaded.

And I'd say the AR book is still very -- obviously very high quality, and that's a DSO figure that we're very happy with.

Great.

Thanks.

Our next question comes from Dan Arias, Citibank.

On the NextSeq system, Jay, I think back in, I believe January, you talked about the road show that you went on after that product was launched, and I believe at the time you said you had generated something in the neighborhood of 2,500 or so leads.

Any way you can talk to how much of that you've worked your way through or converted, just in order to try and get an updated view of the placement funnel there?

Well, I don't have that number at my fingertips, in terms of the size of the pipeline now.

We did that road show shortly after launch in the early part of 2014, and we've seen the benefit of that, because the NextSeqs are marching up every quarter.

I think the V2 chemistry has helped a lot.

So many customers who felt there was a slight difference in quality from NextSeq to HiSeq, that's now totally neutralized with the v2 kits on the NextSeq.

So we're really bullish on NextSeq.

I think particularly in the NIPT market, as we launched the VeriSeq NIPT broadly as we penetrate China with that platform.

As larger panels come into the market, where you need more throughput to run the panels, I think NextSeq is going to be a really, really important product for us.

Maybe as a follow-up on the X systems, I'm sure it's a combination of both, but maybe in general, I'm curious whether you're finding that with the discussions that you're having on buying an X Five versus an X Ten, is it more a function of capital and funding right now, or is it more based on throughput and capacity?

Thanks.

I'd say it's a combination.

So the X Fives that we're selling are to customers who will likely become X Ten customers at some point in time, but they want to walk before they run, if you will, and they're willing to pay a little bit extra on the price per genome for a while, until they really understand what their pipeline looks like.

So it's a combination of overall capital cost, and whether they think they're ever going to get to an X Ten or not.

To be clear here, we're also seeing the 4000, I think is a very important addition, right along that spectrum.

And so it's the case that some people who might have wanted to buy an X one or X two had we offered that, are now really happy with the 4000 product line, that allows them to run all the different applications that we have in the portfolio.

Okay.

Thanks very much.

Our next question comes from Amanda Murphy with William Blair.

So I just had a two part question on the HiSeq.

The first part is I think pre- 4000, pre- the launch of the 3000, 4000, you were shipping roughly 70, 80 platforms a quarter.

I'm curious, is that still the case, and is that what you expect going forward, just given that potentially the 2500 to 4000 conversion may not be exactly one to one, given the throughput differences?

And then the second part is, there was also a discussion originally around the 1,000 or so platforms that couldn't access the v4 chemistry of the 2000 and 2500 maybe upgrading to the 3000, 4000.

So curious if you're seeing any evidence of that either this quarter in shipments, or in the backlog?

Thanks.

I'd say the number you cited for the total for all the HiSeqs, 2500 up through the 4000 is in the range, and we're hoping the 4000 gives us a lift there, over the next few quarters.

So we're doing really well in combination.

The mix shifted very dramatically this quarter away from the 2500, and toward the 4000.

In terms of people doing upgrades, we don't have a lot of great data on that yet.

We're certainly seeing some 2500 customers who buy 4000s turning off their 2500s, and so we'll be -- the way we report our reagent pull-through takes into account a certain fraction of the 2500s that get turned off in the field.

And so that's how we do the accounting there, and at some point maybe we'll have to update the installed base of 2500s for you.

We're not prepared to do that today, but we'll probably do that at some point.

I guess that's all I had to say.

Did I answer your question, Amanda?

Yes.

I guess I was just trying to get at whether there was this kind of evidence of incremental demand, and I guess you talked a bit qualitatively, but incremental demand as the 2000 customers perhaps couldn't access the V4, and maybe they went out and bought the 4000 because of the cost per basis images.

Maybe they weren't planning on buying it, but the advantages on a cost perspective were so great, they opted to do that.

Yes.

That we're definitely seeing.

We had a couple genome centers that ordered multiple 4000s, and the reason they did is they could add a couple 4000s and turn off some of the older 2500s that they might have in the fleet.

We're definitely seeing that pattern happen in the field, and I think we'll continue to see it.

Got it.

Thank you.

Our next question comes from Isaac Ro with Goldman Sachs.

Jay, just hoping you could comment on, maybe broadly speaking, where you think we are in the adoption curve for X Ten?

Just assume -- and part of that question I guess, wondering if your guidance assumes the ending new population of sequencing studies get initiated in the second half of this year?

I'd say we're in the stable part of the X Ten lifetime right now.

So that's why we have sort of the quarterly ranges we should expect for your modeling purposes.

I do think any large-scale population sequencing projects could be incremental to those numbers, so throughout the rest of this year, we don't have built in any large purchases for singular population sequencing projects, and that's because it's hard to know when any of those will close, and they tend to be relatively long term sales cycles for us.

So that would be upside if perhaps somebody bought 20 or 30 for doing a large scale project.

Sure.

That makes sense.

Thank you.

Just a follow-up, maybe a longer term question on liquid biopsy.

I appreciate your comments on the maternal cancer study that was published.

Just wondering, as you think about the size of some of these markets and what it costs to develop an IVD that's got the sensitivity and specificity that you need to have, in maternal cancer, the type of product you that think is worth going after, or is it really an example, proof of concept and maybe there are other markets where liquid biopsy would be more compelling to start?

Yes, I think what we published really was just an example that was, I guess we'd call it incidental.

Right?

So the test was not designed to detect maternal cancer.

We got discordant results.

When we went back to investigate, in connection with those women that were in the paper, those 10, we were able to show that in fact what we were detecting was in fact cancer.

That isn't really how you would design a test, if your goal was to detect cancer, it would a very different test.

It would be deeper sequencing, you would probably include other kinds of markers, and as I said earlier, we're investigating lots of different methods, and trying to understand which cancers are best detected with which assay methodology.

Ultimately, the goal here would be to create as general a panel as you can, but likely, this whole market will begin with more targeted panel types from us and from others, that have specific knowledge of the variants that are known in particular cancer types.

But you wouldn't use the existing NIPT test as sort of the default Liquid Biopsy.

That makes sense.

Appreciate the comments.

Thank you.

Our next question comes from Ross Muken with Evercore.

I'm going to stick on the topic of instrumentation or pushes and pulls.

We've hovered around $150 million or so a quarter for, I don't know, four or five quarters now.

You intimated that X Five, X Ten will be in that 20 to 30 placements, units, a quarter range.

As we think about what's going to break us out of this band to resume, obviously, at a huge step up from the $100 million or so, or $85 million to $100 million range you've been doing for some time, as we think about the next phase of the step up, is it really predicated on a new platform launch, or is it, to your comments before, maybe something that develops in pop sequencing that's outside.

You're privy today to be able to model, or does it come from one of the other smaller platforms?

I'm just trying to get a sense for where you see the biggest ability to see that trajectory change.

Russ, I think it's going to come from all the new application areas that we're working on in the market development and pop seq is really only one of those.

Average risk is another, and the whole VeriSeq NIPT platform, we think, will enable that, and that's starting to penetrate the market in Europe.

It will be a significant contributor in China here over the next year or two, in terms of NextSeq placements.

I think what's going on in the consumer markets could have some impact in sequencing.

I think over time, the whole consumer space will begin to migrate away from arrays and towards sequencing because of the additional power you get from knowing all the bases, rather than just the targeted bases.

What we've done in forensics and HLA, we're in the very early phases of adoption in those markets, and those markets are going to be reasonably widely distributed, meaning there's lots of places that will run those applications, and therefore, I think will support instrument placements as well.

If you look at the existing line up of hardware, I think the upward momentum in instruments will be driven by market development across new applications.

And maybe looking at operating leverage, so you've been growing, based on my math, on an organic basis, OpEx in the 20s and 30s, sorry on SG&A, and that started to slow, and you got some really good leverage in 4Q and 1Q, and it continued to slow this quarter.

How should we think about that line, because it looks like R&D has been more constant, 20%.

How should we think about the investment in the commercial organization, and how much leverage you can show there, and how we should think about that relative to the top line momentum?

Well, on a non-GAAP basis, our R&D and SG&A went up as a percentage of revenue this quarter from last quarter.

So we do have a lot of open reqs, we have a lot of important things to be working on, and so we're going to continue to add headcount in key areas in commercial, and in the R&D and research groups.

So we're not at the point right now where we're predicting additional expense leverage, and we think that's the right position for the Company to take.

Again, non-GAAP, we're at 37%-plus operating margin and we're pretty pleased with that.

As I mentioned in my remarks, we added 315 net heads, to punctuate Jay's point a little bit.

You'll see a full quarter effect of those additions in Q3, plus the new additions, the open reqs that Jay mentioned, starting to filter in.

Perfect.

Our next question comes from Dan Leonard with Leerink.

Can you comment on interest from new customers in the HiSeq 4000?

I believe you gave commentary on the benchtop market, but I don't think there was commentary on the high throughput market.

Clearly we have new customers in the HiSeq part of the market, as well, Dan, but it's much less as a percentage than we have on the desktop instrument version, just because of the capital cost.

And most people who have been in the sequencing market already own some version of a HiSeq, so they wouldn't be a new customer to us.

Okay.

And my follow-up, Jay, I'm trying to better understand the sequential decline in consumables revenue.

I hear you that arrays were weak, but just given the relative proportion of arrays versus sequencing, and all the commentary on sequencing, I would have thought sequencing on consumables would have been enough to sequentially offset array weakness.

Can you help me understand what I'm missing?

I can cover that.

Obviously, Q1 was very strong in sequencing in consumable growth.

Q2 continued to grow, but it was obviously offset by the array business declines, which brought the overall growth number down.

As I mentioned on the call, we've seen eight quarters now of sequencing growth, and sequencing consumable growth in excess of 30%, so we're very pleased with the Q2 result.

I mentioned that both HiSeq X and NextSeq were slightly above their ranges, and the other instruments were within their ranges.

So very comfortable with that outlook.

Okay.

Thank you.

Our next question comes from Bill Quirk with Piper Jaffray.

Want to thank you for the comment around the clinical and the translational instrument trend.

Looks like that ticked up here nicely in the quarter.

Can you help us think about maybe the bigger picture here?

I guess what I'm trying to balance is, we had this very rapid adoption curve going on in the clinic, frequently around oncology, as well as some other applications, but at the same time it looks to us like there's a little bit of inconsistent reimbursement trends in the States, and so I'd love to hear some longer-term thoughts there.

And then also, we're seeing some really nice encouraging interest over in Europe, as well, and any color there would be great.

Thanks.

Okay.

Yes, I mean, reimbursement's classically challenged for customers in the lab business.

Part of the work we're doing in the OncoPanels is to deal expressly with that problem, and why we're working to hard to get IVD products approved, so that we can have standardization around the products, which makes the reimbursement easier, because you have the clinical evidence and you have an FDA-approved product.

So our whole focus is to push these products toward IVDs, that we think will accelerate the overall reimbursement.

With respect to Europe, I'd say we had a good quarter in Europe.

It was not as good as the Americas.

The Americas really, really had a great second quarter.

So we're super strong in the Americas, good in EMEA and not good in APAC, is the overall profile of the geographies.

Got it.

And then just a quick one for Francis regarding the ERP system.

You mentioned that you're going to live here in about a week or so.

Is there any -- I certainly recognize you've been hiring a lot of people so you have additional OpEx.

But is there any potential decline or associated reduced spend associated with that now, that you have it implemented, or I should say will have it implemented next week?

Sure.

Actually Marc's driving our ERP implementation, so maybe I'll turn it over to you, Marc.

I wouldn't expect any immediate reduction in spend, other than what's in capital in general, because the project-related expenses tend to hit there.

We still need a significant team to continue to manage through the initial period, which we call hyper care, and on an ongoing basis.

And then thinking about longer term, one of the reasons for doing this project is it enables to scale and grow at a rate that is lower than revenue growth, and give us some leverage, and we would expect to see some benefits from a system like this over time, over a longer period of time.

But we're going to be in this mode of implementing follow-on spaces of ERP for the foreseeable future.

Got it.

Thanks.

Our next question comes from the line of Steve Beuchaw with Morgan Stanley.

I'll start on the X. I'll reiterate the thanks for the 20 to 30 guidance.

That certainly gives us more clarity there.

Though maybe just a follow-up to Isaac's question.

I wonder, given how specific HLI has been about their plans to ramp up capacity this year?

Have you incorporated any HLI capacity expansion into that 20 to 30?

Well, we've incorporated an overall forecast that averages our probabilities across all the customers we know of.

If the comments that have come out of HLI actually come to fruition, that's going to be an upside to the number, if that happens in 2015.

And then on NIPT, you made a comment here in the Q&A about thinking about a move into average risk.

I think the last time you commented on that process, you had a view that in the US, that move might be more likely for a late 2016.

Am I sensing that maybe you're a little bit more optimistic about the timing there, given some, perhaps more open minded language from ACOG here recently?

I guess our view hasn't changed on when we think ACOG or the guidelines will change, with respect to average risk.

But we are seeing the tests being increasingly used in average risk patients, largely on a self-pay basis.

And I think in Europe that will happen, particularly with the launch of the VeriSeq NIPT.

I think labs have a more efficient way to do NIPT testing, which will make the self-pay market more accessible.

So I think you'll see the leakage rate into average risk increase.

And specifically, we view the ACOG commentary as a positive step in that direction because they move -- they removed the recommendation that you don't use NIPT for average risk screening, and so we feel, incrementally, we're seeing steps towards that time frame.

Got it.

Very clear.

Thanks, everyone.

Our next question comes from the line of Zarak Khurshid from Wedbush Securities.

How important or meaningful was Berry following the Chinese FDA approval in driving your NextSeq backlog and NextSeq sales in the quarter?

We did get some orders from Berry.

It wasn't some gigantic single order placement.

I think they'll be spread out over time.

We did get some units from them but nothing that in and of itself moved the needle in a big way.

Sounds good, Jay.

A follow-up on NeoPrep.

How are things shaping up with the pipeline there?

Any thoughts on attachment rates to HiSeq currently, and as we think out into the future?

Thanks.

Yes, the incoming order rate is pretty good on NeoPrep.

We're just -- we've been anxiously awaiting the time when we felt we could fully ship this product, and we're now there.

As we mentioned in the script, we worked really hard on this, and it's complex to get this product super robust, because we do tens of thousands of operations in a single run, and if you look at the statistics on this, if any one of those operations fail, a sample doesn't work.

And so it's required some really great engineering to get this to where we've gotten it now, and we're feeling really comfortable about it with the first kit, we're close to releasing the second kit, and then we've got a line up of kits behind that.

So we've made really fundamental improvements to the technology, and we understand it really, really well.

So I think you're going to begin to see the shipment rate increase, and the shipments in Q2 were really pretty insignificant, because we were still dealing with a handful of test customers, to make sure we got them very happy with the platform and the results, success rate results on the instrument.

Wonderful.

Thank you.

Our next question comes from the line of Jeff Elliott with Baird.

First one from me, Jay.

On SeqLab, do you have any color you could share on where that offering could bring customer utilization on the X Ten, really in light of your comments about the concentration of utilization amongst a handful of customers.

As a follow-up, do you have any plans right now to expand the X Ten to other sample types or beyond whole genome sequencing?

I'll make a comment on that, and maybe Fran Francis can add some comments about the product itself.

Our hope certainly is that having SeqLab in place allows us to accelerate pretty substantially the rate of customers coming up the curve on a HiSeq X Ten.

As we plotted the consumable rates, you heard us say in the script that it's very biased, the consumer was very biased toward a few customers at the high end.

And then it's -- we've used the word bimodal, but really if you look at the plot, it just trickles down in a relatively linear fashion to customers at the low end, particularly geographies who are not using it as much as they should.

So our goal is to just move all those customers up, and I think having software architecture around the boxes that simplifies all the lab management challenges that our customers have is going to push that up.

We're not ready to quote a rate on how fast that will go, or what it will do, but given the fact we're already running above our guidance in terms of consumables on the X, to the extent we can move that curve at all is going to be very, very helpful to us.

Francis, any comments on the product itself?

Sure.

One of the things we heard consistently from customers around the X is that they were viewing the IT and informatics challenges as daunting, and as, frankly, taking them time to actually get the lab up and running, and so the idea with SeqLab is to provide them this integrated solution that will help them in number of ways.

One, it will speed up how quickly they get the lab up and running, and then two, it will enable them to be higher-velocity sequencing labs.

We have data, for example, when we look at our BaseSpace users, the data we have suggests that today about 50% of our customers have connected to BaseSpace, and the data shows that customers who are connected to BaseSpace, on average, are higher-velocity customers than other customers.

And so we see that in higher pull-through.

So the idea with SeqLab is to enable our high throughput customers to get that same benefit.

There was a second part to your question that I didn't get to.

Remind me what that was?

Yes, opening up the X Ten to other sample types?

Or -- yes.

Right.

Yes.

Well, we continue to evaluate that.

We, in our planning sessions, debate every quarter what the future of the X will look like, in terms of whether we stick with X Five, in terms of how the units are deployed, and whether we open other applications or other organisms.

We don't have any decisions we're ready to share today about that, but we continue to evaluate it, and at some point it's likely, we'll make some changes on the X platform.

But nothing to report today.

Thanks.

Our next question comes from the line of Dane Leone with BTIG.

Thank you for taking the questions.

So I just wanted to revisit the consumable question, Q on Q, because it seems to be a topic of concern.

Could you maybe clarify some of the projects that you have going on with the HiSeq X customers to help them ramp up, and how you see that playing out?

We've spoken to people in the industry, saying yes, getting samples can be difficult, and it's going to take time.

Just maybe any color there in terms of how you see that higher utilization, higher capacity genome center really ramping up over the next couple quarters?

Well, I'd say on the X Tens, X Fives and X Tens, the classical high throughput sequencing labs are the ones that are at the high end of the user group that we have today.

So they already knew how to do sequencing.

They had plenty of samples.

Many of them had samples in the freezer that they just couldn't fully sequence before, because of price points.

So those customers tend to already be at the high end of our curve.

The ones we're working on are the ones that don't have the LIMS systems in place, don't have the analytics in place to analyze the data, and so some of them will turn on the X system, and all of a sudden they've got 5,000 genomes, but they don't know how to analyze the genomes, and so they slow down a little bit until they figure that out.

The SeqLab helps that a lot.

In some places, we can help with samples, and some places we can't.

We just have to let them deal with their sample acquisition challenges.

Some of that has to do sometimes with consent.

Some of it has to do with whether they're trying to access clinical pipelines and regulatory issues around access to clinical pipelines and samples.

But in some cases, we can help customers get access to projects if they're doing things more in the services business.

And we do do that pretty routinely.

And then you have, at the very low end of our utilization curve those customers where the purchase was almost more of a vanity buy, and we try to not do many of those, in fact, we push back on quite a number of customers who said we want to be in this game, and there was really no clear thing they were going to do with the X, and so we haven't sold to them.

But there's still a few at the low end that are not utilizing the system as we expect they should.

Okay.

Great.

Thank you.

Our next question comes from the line of Tim Evans from Wells Fargo Securities.

Thank you.

It seems like there might be a little bit of a disconnect, if we frame things in terms of your TAM, which you've laid out.

Arguably, the commentary around some of these markets has been that uptake has been maybe faster than we've expected, and yet your revenue growth for the year has been -- your expectations haven't really gone up in tandem.

Can you comment about whether you still feel comfortable around some of the key assumptions that went into building some of the bigger pieces of that TAM model?

Yes, I think we feel quite comfortable with that.

And in fact, there's things that have happened since we talked about the TAM that not only make us comfortable with the numbers we gave, but probably would cause us if we were to revise them today, to do that in the upward direction.

In our markets, in general, the way the markets develop is that you see a gradual adoption rate which is -- which extends over some relative period of time, a year, a couple years, depending upon the market segment.

And then the market hits an inflection point, and that inflection point can be driven by factors such as reimbursement, by the publication of clinical studies.

In the case of NIPT it became a marketplace that exploded because the clinical data came out, and replaced an invasive test.

And so in oncology, we're going to see that inflection point happen.

Is it going to happen a year from now or two years from now?

It's hard to predict.

But there will be an inflection point in the oncology market, where that market will explode, and that would be around perhaps the time when in practice guidelines it becomes standard of care to sequence every tumor when it's biopsied, and that will cause the market to go very, very quickly.

In the pop seq market, you'll see that inflection point once you begin to see the existing pop seq projects begin to demonstrate the fact that sequencing whole human genomes actually has an impact on the quality of care, and the economics of healthcare.

And once you start seeing that, that market will explode as well, in our view.

In the consumer market I think there's factors around that as well that we could go into about what's going to kick the consumer market into high gear, but I think there's some potential things that might happen here over the next year or two that could do that as well.

That's very helpful.

Thank you.

I will now turn the call over to Rebecca Chambers for closing remarks.

Thank you, operator.

As a reminder, a replay of this call will be available as a webcast in the investor section of our website, as well as through the dial-in instructions contained in today's earnings release.

Thank you for joining us today.

This concludes our call and we look forward to our next update, following the close of the third fiscal quarter.