Illumina Inc (ILMN) 2005 Q3 法說會逐字稿

[OPERATOR INSTRUCTIONS] This call is being recorded. And now for opening instructions, I would like to introduce Illumina's Director of Corporate and Marketing Communications, Bill Craumer. Please go ahead, sir.

Thanks, Richard. Good afternoon, everyone and welcome to Illumina's conference call. At the close of market today we released our financial results for the third quarter and nine-month period ended October 2, 2005. During this call we will review our results and provide financial guidance for the remainder of 2005, after which we will host a Q&A session. Presenting for Illumina today will be Jay Flatley, our President and Chief Executive Officer; and Christian Henry, our Vice President and Chief Financial Officer. The call is being recorded and the audio will be archived on our web site at Illumina.com.

During the call we will be providing guidance and discussing plans for future commercial activity. Our intent is for these forward-looking statements to be protected under the Private Securities Litigation Reform Act of 1995. Forward-looking statements are subject to risk and uncertainty. 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 up to date these statements. To better understand risk factors, we refer you to the document that Illumina files with the Securities and Exchange Commission including forms 10(K) and 10(Q). I will now turn the call over to Christian.

Thanks, Bill. Good afternoon everyone. And thanks for joining us today. I hope that you've all had a chance to review our financial results that were released after the close of market about an hour ago. I would first like to walk you through our operating results for the quarter, follow with an update on our financial guidance for the remainder of the year and then I will turn the call over to Jay for an update on the Company's progress.

For the third quarter we were pleased to record or 17th consecutive quarter of revenue growth with total record revenues of $19.5 million. This represents year-over-year growth of 44% and sequential quarterly growth of more than 23%. On a year-over-year basis, Q3 product and services related revenue grew 44% to $19.0 million, as a result of strong demand across all of our product lines. In fact, revenues increased in all product categories during the quarter, compared to the third quarter of 2004.

During the quarter, we shipped a record 19 BeadStations; in contrast to last quarter, none of these systems were leased. However, this quarter we have implemented a commercial leasing program which will allow us, in most cases to take the full revenue for lease systems. We also shipped and installed our 11th BeadLab during the quarter at the National Cancer Institute. This brings our combined install base of BeadStation and BeadLab to 102.

Services revenue increased 154% for the quarter to $2.7 million, compared to $1.1 million in Q3 of last year. Services revenue includes revenue from our genotyping services business as well as revenue from instrument maintenance contracts. Growth in services revenue was due primarily to the increase in the size and number of genotyping services contracts. During the quarter, we completed 28 contracts related to studies on a number of organisms including human, chicken, cow, pig and tomato.

Research revenue was $0.5 million for the quarter and is related to revenue generated under government grants. We expect research revenue to fluctuate from quarter to quarter based on the timing of R&D activities under these grants.

Cost of products and services revenue was $6.6 million for the quarter, compared to $3.5 million in Q3 of 2004. This represents a products and services gross margin of 65.3%, compared to 73.3% in Q3 of 2004. The decline in gross margin during the quarter was related to our product mix and the start-up activities associated with the Invitrogen collaboration. In Q3 2005, our product mix was more heavily weighted toward instrumentation, which has lower gross margins than Q3 2004.

Additionally, during the quarter we began shipping the first products under our collaboration with Invitrogen. The products shipped during the quarter were products that Illumina had historically sold on its own. However, under the new arrangement with Invitrogen these products are now sold under the collaboration and therefore the gross margin is split according to the collaboration. This had a negative impact on product and services gross margin during the third quarter.

In the fourth quarter, we will ramp up shipments of products that Illumina which has not historically sold which are the larger, higher value portion of the market. As this ramp occurs, we expect the revenue in the Oligo business to increase and the margin on the Oligo business to return to expected levels.

Research and development expenses increased by $1.7 million in the quarter to $7.1 million, compared to $5.4 million in Q3 of 2004. The increase in R&D spending was largely driven by the inclusion of Cyvera's research and development expenses in 2005.

Selling, General and Administrative expenses were $7.3 million in the quarter, compared to $7.5 million in the third quarter of 2004. Unusually high legal expenses in Q3 of 2004 offset our growth in sales and marketing expenses this year, resulting in expenses that were essentially flat on a year-to-year comparison. Excluding these legal expenses, SG&A expense grew approximately 27% over 2004 levels. The growth in our sales and marketing expense is primarily related to the expansion of commercial activities required to support our revenue growth.

Finally for the third quarter, we reported a net loss of $1.4 million or $0.03 per basic and diluted share, compared to a net loss of $2 million or $0.05 per basic and diluted share in the third quarter of 2004.

Moving on to the balance sheet, we ended the third quarter with $50.2 million in cash, compared to $54.6 million at the end of the second quarter. Primary uses of cash included the continued expansion of our manufacturing capacity and the funding of our working capital requirements, partially offset by the proceeds received from the stock option exercises.

Accounts receivable day sales outstanding were 81 days for the quarter, which is slightly above our target DSO of 75 days. This was due largely to some extended Japanese receivables. We expect DSO to continue to fluctuate from quarter to quarter based on geographic mix of revenues.

I would like to now shift to a discussion of our financial guidance for the remainder of the year. We expect total revenue for 2005 to fall in the range of $72 to $74 million, which is consistent with the guidance provided in our last quarterly release. This represents top line growth of approximately 44% compared to 2004.

Full year earnings per share on a GAAP basis is expected to range between a loss of $0.56 to $0.53 per basic and diluted share. Excluding the effect of the one time charge for inprocess research and development recorded in connection with the Cyvera acquisition, we expect full year earnings to range between a loss of $0.16 and $0.13 per basic and diluted share. These figures are consistent with our previous guidance.

For the full year research and development expenses are expected to be in the range of $28 to $29 million, and selling, general and administrative expenses are expected to range from $27 to $29 million. These two are consistent with our prior guidance.

For the fourth quarter, we expect total revenue to range between $21 and $23 million. Net loss per share should range between a loss of $0.03 per basic and diluted share to break even. I will now turn the call over to Jay for additional comments before we begin the Q&A.

Thanks, Christian. And good afternoon, everyone. We are extremely pleased with our progress in all areas this quarter. We achieved record revenues, continued to grow our installed base, expanded our manufacturing capacity, launched key new products, as well as commenced shipping of product under our collaboration with Invitrogen.

Our order rates continued to be very strong during the quarter which resulted in another record ending backlog. As Christian mentioned, we shipped 19 BeadStations and one BeadLab. This puts our installed base of BeadStations and BeadLabs over 100 units, which means we have more than doubled our installed base in the first nine months of this year.

During the quarter, we continued to make progress toward expanding our array manufacturing capacity. As I indicated last quarter, we are aggressively working to more than triple our array manufacturing capacity; which, when completed earlier next year, will support our anticipated demand through the end of 2006. This tripling of capacity will be completed with a capital investment of less than $5 million, highlighting the cost-effectiveness of our manufacturing technology.

Additionally, in September, we successfully implemented 24/7 manufacturing on arrays which alone yielded a capacity increase of more than 40%. This capacity will be put to good use, as we are experiencing a very robust market, particularly for whole-genome genotyping. We're seeing a dramatic increase in the number and size of projects being undertaken in this segment, reinforcing the scientific value of performing large-scale disease association studies.

We began shipment of our Human-1 Genotyping BeadChip last quarter as our first entry into this exciting market. Our chip contains 109,000 SNPs, approximately 80% of which are selected to be in or near genes. These SNPs are widely agreed to offer the highest likelihood of being medically relevant. The data from our customers in the field have been nothing short of astounding, reflecting the underlying quality of our design and manufacturing technology. Yields on these chips in manufacturing have dramatically exceeded our expectations to the point where we will be delighted if they remain exactly where they are.

We are well into the development of the next chip in the whole-genome genotyping family, the HumanHap-1BeadChip. This array is expected to contain approximately 250,000 SNPs designed with so-called Tag SNP Content derived from the Hap Map project. The use of Tag SNPs will enable researchers to analyze the human genome comprehensively and efficiently by examining between 250,000 and 500,000Tag SNPs, rather than looking at the entire collection of roughly 10 million human SNPs. The context for the HumanHap-1 was selected and designed in collaboration with leading geneticists and Hat Map collaborators.

All of the arrays in our Whole-Genome Genotyping portfolio use our breakthrough Infinium Assay. This assay is extremely easy to use, offers full automation and allows virtually any SNP in the genome to be accessed. In combination, we expect our BeadChip technology and our Infinium Assay to provide the market with industry-leading quality and performance delivering on the promise of the HumanHap Map project.

During the quarter we continue to make great progress in developing and launching innovative new products. We laughed first multi-sample mouse BeadChips for whole-genome expression analysis, enabling researchers to interrogate six mouse samples on one array at price points that are unprecedented. In the fourth quarter, we will launch a similar product for rat expression analysis. Together with our human expression products, we believe that these three organisms address about 90% of the market requirement for expression. Our growing portfolio of whole-genome expression products offers the highest level of performance and the lowest price points in the market.

We also introduced in the quarter a new multi-sample genotyping BeadChip configuration that uses our Infinium Assay to genotype between 10,000 and 15,000 SNP per sample. This customized product will enable researchers to perform large-scale customized genotyping with terrific economics. The first customer for our multi-sample chip is the Wellcome Trust Sanger Institute. We shipped a significant number of chips to them during the third quarter for studying the impact of the key class of amino acid-changing SNPs on disease suspectibility.

Next I would like highlight some of the important agreements we completed since our last conference call. In September, we announced that the Max-Planck Institute of Psychiatry has decided to use Illumina's Whole-Genome Genotyping arrays to undertake one of the largest studies ever by the Institute. The study will focus on identifying the genetic variations that predispose individuals to specific disorders, such as depression and anxiety, as well as understanding how genetic variation affects the response to psychotropic drugs. They will conduct these studies using a combination of our Sentrix Human-1 BeadChip as well as our forthcoming HumanHap-1 BeadChip.

A good example of disease association research enabled by our Whole-Genome Genotyping technology is the agreement we recently announced with PARC, which stands for Pharmacogenomics And Risk of Cardiovascular Disease, one of several disease genotyping project teams sponsored by the NIH. PARC investigators have launched a groundbreaking study into identifying and determining the effects of SNP-based genotypes on individual response to statins, a widely used class of cholesterol-lowering drugs such as Lipitor. At the end of the study, the project scientists planned to have generated information that will enable health care providers to genotype individual patients and determine which statin drug at which dosage rate will deliver the most favorable cardiovascular outcome at the lowest risk. The PARC study will initially use Human-1 and HumanHap-1 BeadChips for whole-genome association analysis.

Subsequent phases will systematically evaluate hapotides to identify the most informative DNA variations and those that have causal association with cardiovascular disease. Over 10,000 samples will be tested in the latter phases of this project, which will employee our Infinium and GoldenGate protocols with custom panels of SNP markers. Studies like this will lead to the development of improved therapeutic approaches and better diagnostics, while laying the groundwork for a more personalized approach to healthcare that recognizes differential response to drugs and enables more successful clinical outcomes. The PARC study design is an ideal model of how large-scale disease association studies will be conducted in a post Hap Map world.

Also in September, we announced an important genotyping services agreement with GlaxoSmithKline. Under this multi-year, multi-million dollar arrangement, GSK can use our genotyping services group to conduct genetic studies with thousands of samples using both our GoldenGate custom genotyping and our Infinium Assay for whole-genome genotyping. This umbrella agreement enables GSK researchers to rapidly define and execute new studies, as only a simple statement of work is required to initiate a project. This will facilitate rapid progression from study to study, as we will have the ability to avoid the lengthy legal and pricing negotiations on each project.

In September we also installed a BeadLab at the National Cancer Institute. The BeadLab, which is our turn-key production scale genetic analysis laboratory, will be used initially for research into the genetics of breast and prostate cancer.

As we look forward into the product pipeline, we are focusing on exciting additional applications that will have significance in particular disease areas such as cancer. During the fourth quarter, we expect to launch our first products that will examine Loss of Heterozygosity or LOH for short, in DNA copy number. These new products will add to our growing suite of novel solutions for understanding the genetic basis of cancer while expanding the overall array market with new applications.

Another field where we are supporting a growing number of customers is in neurology, which includes autoimmune diseases such as Multiple Scleorsis, or MS. MS has been studied extensively and much of the recent analysis for the MS Consortium has used our GoldenGate linkage and custom SNP products. The October issue of Nature Genetics includes two articles which describe the discovery of genes that confer suspectibility to MS, both of which used Illumina array and assay solutions.

As customers complete research using Illumina's technology, we are seeing a very nice ramp-up of publication activity. This reflects the great data quality and high-value of Illumina products.

Finally, I would like briefly update you on the progress we've made related to the Invitrogen collaboration and with respect to the development of the Cyvera platform. As Christian indicated earlier in his comments, we began shipping our first collaboration products under the Invitrogen collaboration early in Q3. The products shipped to date under the collaboration have been primarily plate Oligos which addresses the smaller portion of overall market. We've now completed installation of the infrastructure required to both produce and distributes Oligos in tubes, which is the larger market and the part that Illumina has not historically addressed. True production is now ramping up and we expect to have all collaboration products in manufacturing by mid fourth quarter, which we expect to both grow revenues and expand gross margins during the fourth quarter.

I'm pleased to report that the development of the Cyvera platform is on track for a launch in the second half of next year. As a reminder, the Cyvera platform complements our Bead array technology by addressing the low to mid multiplex markets, such as the emerging opportunity in molecular diagnostic. The platform will provide important advantages to the diagnostic market in cost, throughput and flexibility.

In closing, we are very pleased with the progress we are making in our ability to execute to plan. And we are particularly happy with very high data quality that our customers are reporting and the growing strength of the Illumina brand around the world.

I think we are now ready to take questions. Operator, could you please explain to our listeners the procedure for the Q&A session?

[OPERATOR INSTRUCTIONS] With will take our first question from Edward Tenthoff with Piper Jaffray.

Great. Thank you very much. Just two quick questions on the array side. Firstly, can you hear me okay, Jay?

Sure.

Okay. Hi, how are you?

Good.

First thing on the genotyping side, how have sales/orders/customers been impacted by competitive launches? And can you also tell us how many customers you are currently shipping, give or take just give us an idea, expression rates, too, at this point.

I guess on the first part of your question -- last quarter we talked about two significant deals that we had lost to competition. And most of the deals that we see in the marketplace continue to be extremely competitive. However, we feel like we are doing extremely well in head-to-head competition out there in the marketplace. But we are really very pleased with what we've been able to produce in the orders that we've been able to book.

With respect to the number of customers in expression, we gave an update on that number during last quarter's call but we don't currently plan to that on a quarterly basis. So we will do that intermittently. I can say that we continue to see good reception for our expression products. We launched the mouse product during this quarter and as soon as we launched that we instantly had some very substantial orders. And as we fill that portfolio out with the rat product in the fourth quarter, we think our portfolio now is in place, that anybody would be able to switch over to our platform and obviously that's our goal.

Real quickly, you had mentioned that you are now leasing the systems differently. Can you mention how that is and can you give us an idea what total system revenues were in the quarter?

Ed, this is Christian, how are you doing?

Hey, Christian, how are you?

Good. On leasing, basically, historically the Company had leased them on -- using our own balance sheet and therefore, we would have to take the revenue over time as the lease payments were made. Now we've worked out a relationship with a third party where we can point the customer to the third party. They will handle the lease and we can take a sale and a -- and all of the revenue right at the point-of-sale or in the same quarter. In this third quarter we didn't have any leases come through. So it didn't really have an impact this quarter but we put the capability in place.

Great.

With respect to total instrument revenue we haven't historically broken that out.

Okay. Great. Thank you.

Thanks, Ed.

Your next question is from Aaron Geist with Robert W. Baird.

Good afternoon. Congratulations on a very nice quarter.

Thanks.

You're welcome. If you could spend a bit of time trying to give us a little bit of magnitude in terms of the gross margin hit from the Invitrogen and instrument sales it would be helpful for us on a go-forward basis, in that you talk about gross margin returning to historical levels or expected levels. Thank you.

Aaron, this is Christian again. We don't really want to break out the differences, but if you look at the historical gross margin that we've had in the last few quarters, those are kind of the -- those are kind of the rates we expect to return to as the Invitrogen collaboration completes its ramp-up. Of course it's going to be highly dependent on the particular mix in any given quarter. One of the drivers in the third quarter was the fact that we sold a BeadLab system which was a higher dollar value, which immediately skewed the mix towards instrumentation.

To follow up on that, we sort of consistently said that we think our long-term gross margins are going to be in the mid to high 60s and obviously it's our goal to push that toward the higher end of that range. And if you look at the difference between the high-end of the range and where we wound up this quarter, there were those two factors that influenced that difference and they were about approximately the same order of magnitude in terms of effect.

Thank you. Can you give us a little bit more color on the capacity utilization of the BeadLabs currently in the marketplace? You've been very successful in placing BeadLabs and BeadStations. Do we still have a lot of traction in terms of utilization on a go-forward basis?

We do. We continue to see very large consumable throughput on our BeadLabs. As we've mentioned many times before, there tends to be a pretty wide spread between the lowest consumable users and the highest consumable user and sometimes that shifts around with time. So last year one of our largest BeadLab users was outside the U.S. and this year our largest BeadLab user will probably be inside the U.S. and that's both because of a decrease in one of the BeadLabs and a substantial increase on the one in the U.S. The mix shifts around a little bit depending upon how grant funding gets allocated, what particular projects they are working on, but in general we are very pleased with the overall consumable rates that are being used on the BeadLabs.

Two last very quick questions. First, when do you anticipate giving 2006 guidance. And lastly, you had a little bit of a spike in current assets. Are you building inventory and can you disclose what specifically that inventory is for, if it is in fact inventory?

Sure. With respect to the guidance for 2006 we will give full year 2006 guidance on our fourth quarter conference call which will be -- which isn't schedule yet, but it will be early February. With respect to the balance sheet and current assets the AR line grew a little bit more than we wanted it to because of some Japanese receivables and that drove current assets up. Inventory also did grow. And inventory grew really to support the magnitude of the business. So more raw materials, more work in process. Since we moved the factory to 24/7 on Bead arrays we've had to make sure that we have enough supply so that the factory never stops turning.

Sounds good. Thanks much.

If you recall, Aaron, the way we make our arrays and particularly the fixed content arrays, is we generate a single Bead pool that has all the content in it. So for the 100 K chip we developed and put into inventory a single Bead pool that has 100,000, really 109,000 Bead types in it. And that's a high-value Bead pool that can make many tens of thousands of chips. And that Bead pool resides in inventory and will be depleted over the over which we manufacture those arrays.

Thanks much.

We will take our next question from with [Zarack Gershad] with Pacific Growth Equities.

Hi, guys. Congratulations on the quarter.

Thank you.

Could you tell us, just refresh our memory as to -- does the told installed base number include BeadStations bundled into BeadLab?

Yes, it does. The individual BeadStation count we tend to site separately and we have 11 BeadLabs, some of which have more than one BeadStation.

And the BeadLabs sold in the quarter, was that of the 1,000 or 2,000 variety?

It was a very large configuration -- so 2,000.

Great. Thank you.

Our next question is from Eric Schmidt with SG Cowen.

Good afternoon. Nice quarter. Can you comment, Jay or Christian, on the make-up of the 19 BeadStations, whether they are the 500 GX variety or what percent might be that variety?

Yes, if you look both in this quarter and historically we tend to sell most of the systems at the high-end configuration. I don't have a number exactly in front of me but I would guesstimate that at least 80% of the installed base is GX's at the moment.

And that wasn't different in the quarter?

Pardon me?

That wasn't materially different in Q3, either?

That's right, that's right.

It sounds like from your comments, Jay, that the Invitrogen collaboration, at least in Q3, didn't really impact top line revenues. Is that correct?

I would say it wasn't a material impact. Our Oligo business continued to grow in the quarter. But there wasn't a significant impact in top line revenues and that's because we weren't selling into the largest part of the true business during the quarter. That's about to commence. We've got all the pieces in place and we are just doing the mechanics of the switch-over now.

So you kind of took a hit to margins with no benefit to the top line in Q3 -- or little benefit to the top line?

That's correct.

And then last question is on the 100 K chips. Just kind of looking for a little bit more color on what --you said there's been outstanding reception and feedback on the Human-1 BeadChip. But can you talk about the numbers of customers that have bought into the chip? Or the numbers of chips sold? Or competitively -- I know it's a very dynamic space, whether any capacity constraints that your competitors may have been having have impacted that roll-out?Or any kind of visibility around that chip at all?

Yes, I think it's a little early for us to start disclosing specific numbers of chips or customers. I can say that we've competed head-to-head on quite a number of deals and we've done quite well, we think overall, in those competitive arenas. We are now close enough to the launch of the 250 K that we have the ability to package together the sale of the 100 K and the 250 K, which you've seen in a couple of deals that we already announced, both the Max-Planck deal and also in the PARC deal. And you will see more of that, particularly as we get closer and closer to the 250 K launch. So we are quite pleased with the overall reception to the chip.

I also can say that the data quality portion of this has been just incredible. We've exceeded our specifications on the chip in terms of accuracy and call rate. And it's a real credit to both our development team and manufacturing teams for being able to produce chips of such outstanding quality.

In terms of what may be going on with other people in the field, it's hard for us to know who has come to us as a result of any issues anybody else might be having. So it's a little inappropriate for me to comment specifically on that.

I appreciate that. Thanks a lot.

Our next question is from May-Kin Ho from Goldman Sachs.

Hi, Jay, you mentioned earlier that with the Invitrogen infrastructure that you are now gearing up for the two Oligos that might increase your gross margin, going forward?

Yes.

What kind of increase are you thinking about? And also you mentioned that there will be a growth in revenues.

Right.

Can you give us some idea about what you are thinking there, too?

Sure. If you recall how the structure of the Invitrogen partnership works, we've always planned as we entered into this collaboration, that our overall gross margins in Oligos would be lower than what they were when we were a stand-alone company in that business. But that almost all of that gross margin would be net margin. So what we expect is that, as we get all the products through the switch-over if will you, that we would expect about a 200 basis point increase from where we are now in our Oligo business from where we were, say, in the third quarter. And the margins in the third quarter were materially down from what we experienced on our own. So I would say overall, if you were to look at the corporate gross margins we reported this quarter you could add about 200 basis points to that number, based on our current mix assumptions going forward, as we fully integrate the Oligo business.

And when do you expect it to become completely integrated?

We think it will happen around the middle of this quarter. We are really just now in the mechanics of the switch-over. We have all the instrumentation fully qualified. All the infrastructure in place. All the IT systems up and running.

And going forward, will economy of scale affect this as well? So let's say in '06, same time this year will be better -- ?

Gross margins?

Yes.

A little bit. Under the agreement there are some -- some fixed transfer prices, in terms of how the arrangement works. So to the extent that we do better in certain areas in our actual manufacturing costs, there's a chance of some improvement to the extent Invitrogen does better in their actual distribution costs, they have the opportunity to gain some in the distribution allocation.

Thank you.

[OPERATOR INSTRUCTIONS] We'll take another question from John Sullivan with Leerink Swann.

Hi, guys. A couple of quick questions. Have you given a date yet regarding the official launch of the 250 K. product?

We have not.

Okay. And secondly, regarding current chip manufacturing capacity -- this one went past me before I got it, I think -- you said you are currently operating 24/7, regarding current chip manufacturing capacity. Is that true?

We switched over right at the beginning of September to 24/7.

Does that imply that you are at relatively full capacity today for chip manufacturing?

Yes, but if you recall how our manufacturing process works, the incremental additions for capacity can be added in small chunks. And the constraint in our manufacturing process are the decode imaging systems and those are added one at a time as we need them. And so we are in the process of ramping up the build of those decode instruments and we plan, as I mentioned before, to about triple the number of those decode instruments that are available to us, if you compare where we were at the end of last quarter to where we think we will be in around March or April of next year.

Okay. So relative to September 30 you expect to be three X by March 30?

That's about right.

Okay. In the meantime, you are ramping up -- you are continuing to ramp up human and mouse expression arrays, you are ramping up a 100 K. SNP and you want to bring a rat expression to market. Should investors be concerned that you are pushing up against capacity -- short-term capacity ceiling?

Well, I can say that our factory is running pretty much all out but it's not at the point where we are seeing any problems in managing our business. And every week we are adding new decode capability. And so, every week our capacity is increasing. This is not a one time step function where we are running at this capacity, say until January 15th, and all of a sudden then we take a big step function. Our capacity additions are made in small increments. Every week our capacity is going up and we are managing that process consistent with the growth of the business.

Okay. Fair enough. Last question, did you say that you completed 28 unit-typing services contracts in the third quarter of '05?

That's right.

Can you just get some sense as to what the average genotyping service contract looks likes, it's term and the average dollars associated with one?

A typical term is about three months. Most of that time has to do with gathering the samples and doing the SNP selection with the customers. We actually -- often that actually happens before the three-month clock starts. And then typically it takes us -- we give a three-month commitment from the time the SNP selection is finished and the samples are received for us to do the manufacturing of the Oligo pool, the genotyping and then the data analysis. We typically deliver a bit earlier than three months. But I'd say the typical project is finished between two and three months.

The minimum service project that we take is about a $50,000 project. I would say the average project is in the several hundred thousand dollar range. And then we have them that are in excess of $1 million -- many of them are in excess of $1 million.

Okay. So there was an unusually large number of projects that were concluded in this quarter from that average that you just mentioned. Is that fair?

It's a little bit lumpy, but it's at an increasing rate. I don't recall at the moment the number we had last quarter but it was a substantial number last quarter as well.

I think we had 22 last quarter, if I remember correctly. I think the message here is that the services business is ramping up as the rest of our business is. And we are focused on just making sure that we execute and deliver the appropriate data quality. As you might appreciate, this is also an excellent way for us to get people used to the Illumina technology. And we've seen several of these service customers end up becoming BeadStation purchasers as well.

No doubt. Do you -- is there a percentage of capacity number, even a rough one that you would quote us now, regarding your ability to provide genotyping services?

It depends really on the contract, so it's really hard for to us predict.

Okay. Fair enough. Congratulations.

Thank you.

At this time we have no further questions. I would like to turn it back to the speakers for any additional or closing remarks.

Okay. I guess that's a wrap, Richard. Thanks, everybody, for joining us and we look forward to speaking with you next quarter.

This does conclude today's conference call. We thank you for your participation. Have a good day.