Socket Mobile Inc (SCKT) 2017 Q4 法說會逐字稿

Welcome to the Q4 and 2017 management conference call. My name is Adrian, and I will be your operator for today's call. (Operator Instructions)

Please note this conference is being recorded. I'll now turn the call over to David Dunlap. David, you may begin.

Thank you, Adrian. Good afternoon, everyone, and welcome to Socket Mobile's Fourth Quarter and 2017 Management Conference Call to review results for its fourth quarter and year ended December 31, 2017.

Presenting today from Socket Mobile are Kevin Mills, President and CEO; James Lopez, Vice President of Marketing, Sales and Developers; and Dave Dunlap, Chief Financial Officer.

Socket Mobile distributed its earnings release over the wire service earlier today. The release has also been posted on Socket Mobile's website at socketmobile.com. In addition, a replay of today's call can be accessed on Socket's website by selecting About Us, Investor Relations/Conference Call/Events. A transcript of this call will also be posted on Socket's website within a few days.

Before we begin, I'd like to remind everyone that this conference call may contain forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and section 21E of the Securities and Exchange Act of 1934, as amended.

Such forward-looking statements include, but are not limited to, statements regarding mobile data collection products, including details on timing, distribution and market acceptance of products and statements predicting trends of sales and market conditions and opportunities in the markets in which Socket Mobile sells its products. Such statements involve risks and uncertainties, and actual results could differ materially from the results anticipated in such forward-looking statements.

As a result of a number of factors, including, but not limited to: the risk that manufacture of Socket Mobile's products may be delayed or not rolled out as predicted due to technological, market or financial factors, including the availability of product components and necessary working capital; the risk that market acceptance and sales opportunities may not happen as anticipated; the risk that Socket Mobile's application partners and current distribution channels may choose not to distribute the products or may not be successful in doing so; the risk that acceptance of Socket Mobile products in vertical application markets may not happen as anticipated; as well as other risks in described Socket's' most recent Form 10-K and 10-Q reports filed with the Securities and Exchange Commission. Socket does not take -- undertake any obligation to update any such forward-looking statements.

And now with that said, I'd like to turn over the call to Socket's President and CEO, Kevin Mills. Kevin?

Thanks, Dave. Good afternoon, everyone, and thank you for joining us today.

We reported another solid revenue year with annual revenues of $21.3 million, a 2% increase over 2016 revenues of $20.8 million. Our Cordless data-capture revenue grew by 14%, from $17.9 million to $20.4 million. In 2017, Cordless data-capture products represented 96% of total revenue as opposed to 86% in 2016. The overall growth rate of 14% of our data-capture products was very much in line with the expectations we outlined at the beginning of 2017. In addition, we expect this underlying growth trend to continue in 2018.

In 2017, our expenses were $8.9 million, up from $7.8 million in 2016. The increase in expenses was primarily driven by increased research and development spending, which increased by about 20%. We also saw improved margins, reporting a 53.5% gross margin, up from 50.5% in 2016. In both 2016 and 2017, our EBITDA was $3.2 million. Our income and earnings per share in both years were impacted by changes in taxation that Dave will discuss later in the call.

Overall, 2017 was a solid year from a financial point of view. However, 2017 was a much more significant year as regard to progress towards our longer-term goals, some of which I'd like to highlight.

In 2017, we completely revamped our products, completing our DuraScan products and completing the development of our new SocketScan product family, which we are launching this quarter. Both of these product lines are new and improved versions of our existing products, which have been completely redesigned, retooled and updated for volume production and lower longer-term costs.

We completed most of the work of our -- on our new and improved software development kit, or SDK, which we call Capture which will enable us to attract more developers. The Capture SDK has been restructured to fit better into the many software development environments used by our developers and to better support the ever-changing world of mobile computing.

Our Capture SDK also supports our contactless reader/writer product, the D600. This is very important, as it enables our developer partners to add contactless reading capabilities to their applications with a minimum level of effort. And while we fully understand that contactless reading is not very pervasive yet, we believe that will change over the coming years, as Apple and Android devices add contactless reading capabilities to all their devices.

As we have mentioned many times, no one is going to write applications for hardware that doesn't exist. But in addition to the hardware, the developers also need the software development tools. Adding the contactless capabilities not only expands Socket Mobile's technology reach and transforms us from a scanning company to a data-capture company, but also makes us the first protocol for application developers who want to use this new capability to create new and compelling applications.

And while our long sales cycle doesn't enable us to impact revenue in the short-term, our Capture SDK is enabling our software partners to build the application that sets us up for continued growth in the coming years.

I would now like to turn the call over to James Lopez, who will provide an update on Q4 and provide an overview of the market. James?

Thank you, Kevin. In Q4, Socket saw a return of seasonality, resulting from our retail market, which represents 80% of our sales and is largely driven by our point of sale application partners. Seasonality is typical but was masked by better-than-average deployments in Q4 last year and is something we continue to expect in Q1 and Q4 each year.

Late Q4 and early Q1 typically generate less first-time scanner customers for Socket, as retail hardware deployment slows in the period between mid-November to mid-February. Retail-related scanner sales overall grew at 14% for the year, and while it represented 80% of our business in 2017, I'm happy to report that this is down from the roughly 90% level it represented in 2016.

This shift is largely the result of the success of our DuraScan durable scanners in gaining traction with our developers and consumers in commercial services, industrial manufacturing, transportation and logistics and healthcare markets. We expect this trend to continue in 2018, as we see the pipeline of new developers that joined us in the past year address the needs of these markets.

2017 overall continued to demonstrate healthy growth in our developer community and was our best year ever for new developer-generated applications, leveraging the Socket Mobile SDK with integrated support for our scanners. At present, we track over 700 applications and counting on the Apple iOS App Store, with Socket scanning support designed in.

Looking forward to 2018, in Q1, Socket will be releasing our all-new SocketScan 700 Series scanners. The SocketScan 700 Series was redesigned from the inside out to service a global scan platform built from the needs of our worldwide developer consumer, our customer and our end-consumer.

Our SocketScan series will feature redesigned ergonomics, intuitive performance indicators, better serviceability and extended battery life, which in some cases can more than double what was achieved with our previous generation. We expect our SocketScan 700 Series to organically replace our classic 7 Series products throughout the year.

Our new SocketScan series, and our DuraScan series will also address the growing demand we see across all markets for more cost-effective 2D scanners to serve as universal scanning solution, capable of reading a variety of 1D and 2D barcodes from any surface such as labels, printouts and smartphone or tablet screens.

We see this particularly coming from point of sale applications seeking to use more and more 2D barcodes to support various applications such as Loyalty. To respond to this growing need, Socket has just released our DuraScan D700 scanner, and we'll soon release our SocketScan F700 scanner, creating new entry points for 2D scanners with the same quality and performance expected and worthy of the Socket Mobile name.

Lastly, we see more and more need to innovate beyond hardware in 2018 to support mobile scanning in all its forms, and as a result, we'll soon be releasing our all new Socket Mobile Capture SDK. The Capture SDK and our DuraScan and SocketScan scanners were designed hand in hand to offer easier application development and integration for our development partners.

The Socket Mobile Capture SDK will be released in Q1. The first application built on the Capture SDK, our own scanner Companion App for Android devices, joins our previously released IOS Companion App and is available now for download.

We expect to deliver frequent updates to the Capture SDK and our Companion applications for iOS and Android throughout the year, as we further respond to the needs of our developers and consumers, as they seek to setup, configure, integrate, control and better utilize their Socket Mobile scanners. Now, I'd like to hand things over to Dave.

Thank you, James. Our revenue in 2017 included cordless barcode scanning revenue of $20.4 million, an increase of 14.1% over the previous year and 96% of total revenue for the year, compared to cordless barcode scanning revenue of $17.9 million or 86% of total revenue in 2016.

Our annual sales volume in units increased from 78,600 units in 2016 to 93,400 units in 2017. Our margins increased to 53.5% of revenue in 2017, up from 50.2% in 2016, resulting from increases on our sales volume and product mix, component cost reductions and from manufacturing efficiencies.

Revenue growth and margin improvements allowed us to fund planned higher operating expenses, in particular: higher engineering development costs and expenses for which you are now seeing the benefits in new and upgraded products, including our new durable 2D barcode scanner, model D740, at a much reduced selling price; our model D600 near-field communication reader/writer; additional DuraCase-supported smartphones for 1-handed barcode scanning; our soon to be released upgrades of our complete family of standard barcode scanning products; and new and improved software tools to make it even easier for our registered developer community to integrate our barcode scanning software into their applications.

We were able to maintain operating profitability as measured by income before tax of $2.3 million in 2017, compared to income before tax of $2.4 million in 2016 and as measured by EBITDA of $3.2 million in both years, or about 15% of total revenue.

Our balance sheet continued to strengthen. Subordinated convertible notes and accrued interest of $1.2 million matured and converted to equity in September. In combination with profitable operating results for the year, our end of year key balance sheet measurements year-over-year all increased, including: cash from $1.3 million to $3.4 million; working capital from $1.9 million to $6.5 million; and our current ratio, current assets divided by current liabilities, from 1.45 (sic - see press release - 1.37) million to 3.85 (sic - see press release - 3.81).

As Kevin and James have noted, our sales in both 2016 and 2017 continue to be driven by retail point of sale applications, and the success of the registered developers serving the underserved small retailer with their point of sale application software.

Sales of our products into the retail markets tend to be seasonal and are lower between mid-November and mid-February, while retailers focus on holiday sales. 88.5% of customer purchases of our cordless barcode scanning products in both 2016 and 2017 were made in small quantities online through our worldwide reseller network that includes Amazon as our largest retailer.

The other 11.5% of cordless barcode scanning sales were comprised of larger deployments by businesses that totaled $2.1 million in 2016 and $2.3 million in 2017. There's no particularly seasonal pattern to these deployments, and the deployment levels can vary widely quarter-to-quarter. They are additive to our revenue and make a difference to our quarterly results.

Our fourth quarter results only included $100,000 of deployments whereas our fourth quarter in 2016 included deployments of $1 million. As a result, our fourth quarter revenues and operating results were lower and at levels that produce breakeven operating results as measured by income before tax of $68,000 and by EBITDA results of $256,000.

As we address 2018, we expect deployments to be a positive factor in our future revenue growth, not only in mobile point of sale but also by applications in nonmobile point of sale markets.

Income tax accounting generated adjustments to our fourth quarter operating results in both 2016 and 2017. In the fourth quarter of 2016, we set up $9.8 million of deferred tax assets, the largest asset being the future tax benefit from the application of net operating loss carryforwards to shelter future federal and state income taxes.

At the end of 2016, with our 3 years of profitable operating results and positive outlook for the future, we met the GAAP standard for deferred tax assets as realizability being more likely than not. The effect of the change was to increase our 2016 net income by $9.7 million from $2.4 million before tax to $12.1 million.

During 2017, we recorded income taxes of $1 million, of which over $900,000 was net operating loss carryforwards applied as deferred taxes and requiring no cash. The corporate tax reduction legislation passed by Congress in December of 2017 reduces our effective tax rate from above 40% in 2017 to about 28% in 2018 and beyond.

We revalued lowered our remaining net operating loss carryforward benefits in the fourth quarter of 2017 with a charge of $2.5 million, and will record future income tax starting in 2018 at the lower effective tax rate.

Our remaining net operating loss carryforwards at December 31, 2017, are expected to shelter future taxable income of approximately $21 million for federal taxes and $11 million for state taxes, allowing us to keep the cash that would otherwise have been paid in taxes relating to that income.

I'd also like to briefly comment on the tender offer we filed on February 2, 2018, to purchase 1,250,000 common shares or about 18% of shares outstanding, at a price not less than $3.75 and not more than $4.25 per share. Tendering information has been mailed to all stockholders of record as of -- on January 31, 2018. The offer expires at the close of business on Friday, March 9, 2018, unless it were to be extended. All tendered shares accepted by the company will be paid for in cash at the highest tendered price accepted.

We also reported on February 2, 2018, in our Form 8-K a new bank loan of $4 million that will be drawn around March 1, 2018, as part of the proceeds payable to stockholders tendering their shares. The loan has a repayment period of 48 months. Our Board of Directors approved the offer.

As we stated in the tender offer documents, we believe that the offer is a prudent use of our financial resources and presents an appropriate balance between meeting the needs of our business and delivering value to the stockholders. The offer expresses our confidence in the company's business, our market position and the long-term growth potential of our industry.

We believe the offer is an appropriate mechanism to return capital to our stockholders that seek liquidity under current market conditions. No recommendation will be made by management, any member of the Board of Directors, D.F. King & Co., the information agent for the offer, or American Stock Transfer & Trust Company, the depository for the offer. And no members of management or the Board of Directors will participate in the offer.

We are also restricted under SEC rules from commenting on the tender offer, beyond what is stated in the tender offer documents. All questions may be directed to our information agent, D.F. King.

Now I'd like to turn the call back to Kevin.

Thank you, Dave. In summary, our data-capture revenue growth continues to be strong, and we feel very positive about the future. Our application-driven business model continues to be validated by the strong results in the U.S., and we have seen very solid increases from online resellers like Amazon, which grew by 28% last year. We are also seeing stronger run rate business in Europe and Japan.

Our products will soon be available in China. Yes, we know, we have been predicting this for a while now, while we underestimated the level of paperwork and approval involved, most of which has now been resolved.

We believe 2018 will be a big year for Socket Mobile, as we launch newer versions of our products with attractive price points and new features which we feel will both solidify our market leadership position and enhance our growth and profitability.

With that said, I would now like to turn the call back to the operator for your questions. Operator?

(Operator Instructions) And our first question comes from Will Hamilton from Manatuck Hill.

Just wondering on the enterprise sales front. So you mentioned $1 million a year ago versus $400,000 this quarter. What was the full year number for that?

So, just to clarify, we had $1 million in 2016, and $100,000 in 2017.

In fourth quarter.

In fourth quarter. And in both years, we had $2.1 million for the year of 2016 and $2.3 million for the year of 2017.

Okay. So for the full year, it was up year-over-year just for this particular quarter?

Correct.

Okay, that's helpful. And then, what's the pipeline like for that right now for '18? Timing can always be iffy, I know, but just give a sense of the pipeline?

I think that we would see overall a higher number during 2018 than the $2.3 million we recorded in 2017. But I think as you correctly point out, our ability to predict the timing is very limited. So therefore, we don't. But overall, we do have a lot of projects that would encourage us to believe that we will continuously grow, both in the deployment as well as the run rate business.

Okay. Was it a little bit higher than normal in Q1 of '17, just to understand?

Yes, I think that if you look at the last year, maybe, in golfing terms we kind of did birdie, birdie, par, bogey, right? But overall, I think we did okay. But I think the first half we had stronger deployments than in the second half, yes. And by a reasonable amount.

Okay. So just in terms of our expectations for the first half of this year, okay.

Right.

And then collectively, Kevin, if you look for the year, you still would think that mid-teens type growth like you delivered for the cordless scanner revenue for '17 is achievable again in '18?

Yes, absolutely. I think that the underlying trend of around 14%, 15% is achievable, and that's our target for the year, yes.

(Operator Instructions) And our next question comes from Brandon Beylo from CIC Wealth Management.

So I just had a quick question. Your gross estimates of about 14% to 15%, just to kind of piggyback off of that, is that factoring in any sort of growth estimates for your retail sales in China? I know that China is probably a pretty big question mark in terms of how much sales you're going to generate there, but does that go into that estimate? Or is it just a flat sales report for China?

Well, it's an overall estimate, right? It would include some sales in China. However, based on our business model, we currently don't have many Chinese software application providers recommending the Socket product because the product isn't available to their end users.

Before we can really attract the developers we want in China, we have to have product availability first. So -- and generally speaking, from product availability to when the first apps really become available that are going to drive sales would be a minimum of 12 months. So a lot of what we are doing will impact '19 and '20 as opposed to '18.

And our next question comes from Brandon Swanson.

Do you mean Steve? I just had a couple of questions. One was on revenue by geography. Last quarter, the third quarter '17, we had about 80%, U.S.; 13%, Europe; and 6%, Asia/Rest of World. What did the trend look like for the fourth quarter? Is it similar? Are we growing more overseas?

Yes. The answer is, yes, I believe. And this is -- well, for the whole year, we had 76% -- 78% was domestic, and we had 22% international. We would expect that trend to continue, probably with 2% or 3% coming off the U.S. and adding to international. The international business, particularly in Europe and Japan, seems to have found its peace. It's growing faster, but it's a small base. So to answer your question, yes, international is stronger right now than the U.S.

Okay. And then a follow-on to that, with James being based in your California office, how are we supporting the Rest of World from a developer perspective when they need assistance or guidance or day-to-day help because we got time zone issues? Is there a James in every region? Or how are you actually helping to develop the growth in the Rest of the World?

So there's not exactly a James in every region, but there is definitely a regional sales manager who looks after opportunities and developers for Asia Pacific and Europe. And because Socket has some international routes, we have other employees internationally to help as well.

Are those some folks we hired in '17 as we saw the growth coming or have they been there all along?

We certainly hired someone in Asia. We based someone full time in Asia at, say, December '16, so for all of '17, and that has helped. And in Europe, we actually hired before that, maybe another year before that.

The other thing to point out is that most developers are more comfortable with e-mails than they are talking as a general rule. And that we do do email support based on our development team, and we have some people in the Asia region, we have some people on East Coast and we have some people here. So we do have a little more geographical coverage on a global basis, but a lot of the time it's e-mail driven.

Yes. Okay. All right. And then one last one. I noticed we put an announcement out that we were going to reduce price on this -- on one of the newer scanners that you had. And I'm just wondering how that price reduction is going to impact the margin going forward in '18.

So the margin expectations are the same. And it's not actually a reduced price, it's a new product that was designed to have a lower MSRP and a lower cost base.

Well, it looked to me like, when I went online, it was cheaper than what I was expecting it to be by about $100. And so I was wondering, are we going to -- are margins going down? Are we -- since we're now making more of these, we've got better costs on the components of it? Or how are we able to maintain the margin if we're reducing the price on it was what I was trying to wrap my head around.

So it's a newly designed product that has different design costs with similar margin expectations. And it was designed -- in this case, it was a 2D product. It was designed to offer a more affordable 2D option to the market than what we were able to design in prior years. So from the ground up, it was a redesigned product with different cost factors.

Okay. So it sounds like the cost of it is going to be lower, and you were able to pass that cost savings on to your customers while maintaining the margins on it. Is that correct?

That's exactly right. Exactly right.

Maybe I could add just one thing to that, Jim.

Sure.

A lot of our customers are using 1D, but they're first-time barcode-scanning customers. Historically, 1D has been -- or 2D has been twice as expensive as 1D, right? And for a lot of people, that's an awfully big step. Based on market feedback, a lot of people have told us that if the step was 30%, 40%, 50%, that they would buy the more expensive product to future-proof their solution.

And that really was the motivation, it's to hit a price point that makes it attractive so it becomes a universal scanner and people don't have to worry if they bought the 1D or the 2D, it just reads everything. And I think that will play well with our less technical and first-time customers, which is the reason behind that product.

And the next question comes from Matthew Galinko from Sidoti & Company.

First question is around the SDK. I'm wondering if that's available or in use today in China? Or whether you could say that whether your efforts to get products into the region is generating developer interest there in advance.

Okay. So yes, we have developers in China, right. But -- and we don't know where those developers' applications will go. So people hire third parties to write applications. And certainly, we see a lot of development going on in India and China and other places, but often it's for product deployments outside of China.

And so yes, right, we do have that development. And it's difficult even for those developers who are working for third parties to get product. So -- but we don't really have any, what we call, anchor customers in China. We have a number of international customers who are used by certain brands that have shops and other types of facilities in China that would like to use our scanners. And currently, that is difficult for them to deploy.

But to me this is part of an overall process. The local guys must be able to buy the product easily before they design it in, otherwise it's just a headache for them. They design it in, your end customer calls up and says, "Where do I get it?" And you go, "You can't get this." And you go, "Why did you design it in?"

So we have to break that cycle, and that's in the process of doing that. We feel that China is an important going forward strategy, but it's not going to significantly impact our sales in 2018.

Got it. Okay. I think the D600 made into the market, I think, what was it, Q3, Q4. I'm wondering, do you expect that to be pulled into any deployments in 2018? Or sort of how do you expect the demand curve to move on it?

I would say that you're going to -- well, we already see a lot of development activity happening around it. But like our other development cycles, there's going to be iterations and development happening around it. And a lot of the units I think this year are going to be for the purposes of vetting solutions that you're going to see in 2019, 2020.

But it's an important product because a lot of these solutions are very different than the barcode solutions we've used before. They're much more advanced in what they're trying to do with data capture. There's a lot of integration with devices that also read in your field and smart tags, rewritable tags in particular. So I think you're going to see a lot of developer activity around the sales this year and then more deployment activities in the following years.

Got it. That's helpful. And I was hoping if you could just go back over -- you did talk a fair bit about upgrades of SDK. Can you just kind of bridge my understanding to how that -- how does the experience ultimately change for the developers? Does it allow them to move more quickly? Or is it just a more pleasant experience for them? And how does it -- how do you think it sets you compared to other scanner companies that are competitors?

Well, I mean, for us, a lot of the help with the developers is in abstracting the platforms that they're using with the data capture technology that we provide. And a lot of them aren't experts in data capture, near-field or in barcode. And so what we do is we enhance our SDK so it works better with the platform that they're trying to develop for, and it abstracts the technology that our scanners are built upon.

And so part of it is in how the SDK works with the operating system, part of it is in how the SDK offers more command-and-control over the scanner itself but provides it in a very developer-consumable form, whether it's a library or as forms of service or as sample code. It's all that middleware in between that developers need to kind of realize their desires around the application without having to become experts.

And maybe I could add, Matt. Historically, we had a hard time supporting many of the cross-platform type software development environment. There's a number of them like Xamarin, Cordova, et cetera, where somebody writes an application in C++, but it's compilable in both Android and Apple.

One of the things we've been able to do with the new SDK is support those type of cross-platform environments. Now in the case of Xamarin, there are 1 million Xamarin developers according to Microsoft when they bought them. Today, we really don't support them well. In fact, you could say we don't support them at all.

With the new SDK, we will be able to support that community, and they could still live in their Xamarin world with our tools, right? Our structure historically made it almost impossible. Our structure going forward makes this very possible. So we will be able to add the hooks and components required to allow the developers to stay within their own environment and still use our scanners, right?

And that is particularly as we have larger customers, they're not willing to write in native Apple and native Android, they want a more high-level language. And historically, we weren't good at supporting that. Going forward, we will be.

Yes, we built the new SDK to be very flexible like that.

Got it. All right. And so I guess you commented on this, but you'll sort of add the development environments that you're sort of compatible with over the course of the year, and I'd imagine into 2019 or do you think it's something that's completed in the early part of the year?

I think we're just going to continue to enhance. I mean as the market evolves, our plan is that now that the SDK has been written to be more flexible, we're going to distribute it more flexibly. So today, as an example, you have to download it from our own website. In the future, we'll deploy it through deployment outlets like CocoaPods and NuGet and Maven, depending on your platform and your preference.

As the operating systems evolve, we'll be ready to make adjustments to the SDK and deploy it through all these -- the deployment mechanism. So our goal is it's a new platform that helps us deliver the SDK better so that our developers can integrate it easier.

And the next question comes from Brian Swift from [Sutter] Security.

Can you hear me? I'm out in the desert. I dropped off a couple of times. If I ask you a question you've already covered, I apologize, and we can do it offline.

Could you comment a little bit on your -- on the competitive landscape, your market share over the last couple of years? I would think, with all the products you've introduced and the ones you've talked about that are coming out, that you -- that I would think maybe you would -- your goal would be go back to the -- your growth you had in '16 as opposed to matching the 15% in '17.

And then my second question relates to your strategy and taking [some time] the Dutch auction. You may have covered it, and I -- it was when I was dropped off. But if not, I'd like to hear a little color on the thought process behind that.

Well, Brian, last September, VDC actually issued a report on our market on mobile companion scanning. And from that report, we were able to glean that Socket has about a 36% worldwide share of companion scanners. That was kind of the first time we've had an independent party give us that information.

And our strongest market is companion scanning, but the mobile scanning market is broken up by companion scanning, sled and sleeve scanning and ring scanners. We don't participate in the ring scanner market any longer. But we definitely have a very strong position worldwide in companion scanning. I think in the U.S., in fact, we are approximately 50% of the companion scanning market.

The market overall is growing at about 7.2%, according to VDC, compared to the 14% growth that we've shown. So we're kind of outpacing that. And in the future, certainly, we expect the market to continue to grow at about 7.2%. We see ourselves performing a little better. And the swing element of our business is going to be around the deployments, how well the deployments come in and how well the pipeline of deployments kind of exits on a yearly basis.

And to answer your question on the tender offer, Brian, we're not allowed to answer your question on tender offer price, so...

I can talk about the mechanics a little bit.

I only have 1 bar anyway, I'm probably going to get bumped off.

All right. But we're -- obviously, D.F. King can provide whatever detailed information you need, but we're prohibited from answering questions on the tender offer other than explaining the mechanics, which we're happy to do at any time.

(Operator Instructions) And the next question comes from [George Melis].

On the...

Could you speak up a little bit, please?

Yes, sure. I'm just trying to understand the seasonality a little bit better. Last year, in the fourth quarter, there was this very small uptick in the flow business from the September quarter. And then sequentially, the business was up pretty much every quarter. In this December quarter, there was a really pronounced drop from September. And so it's -- that seasonality didn't exist a year ago, and now it seems very pronounced. Can you sort of explain a little bit what happened?

Sure. So first of all, the seasonality did exist a year ago but was largely masked because we had $1 million in deployment in the same quarter, right? So it looks like it didn't exist but, in fact, it did.

But the flow business was 4.1 in September and 4.2 in December, so you didn't have that seasonality for the run rate business.

Well...

Some retail deployment in there, too.

Yes, there -- okay, so just so we're clear, we actually see October and November as being quite strong months with the [drop off] in December and January and a return to normal business in February. The reason behind this is that very few people deploy new point-of-sale systems during the Christmas period.

And then, in addition, in the January period, most people are closing out the year and evaluating what they're going to do for the next year, plus shows like NRF and other retail-centric shows happen in early January, so people get to see what's new and exciting.

So our business basically, I would say, shuts down from around the 30th of -- well, we'll say from Thanksgiving through early February, okay? That's the slowest period. Now it's only slow for people who are deploying mobile point-of-sale. Obviously, we have some customers who are adding scanners. We have customers that are not in this business. But this we have seen historically.

If we get deals in Q4, which is often what happens, then it masks that. That didn't happen in 2017, it did happen in 2016. So the masking made it look like we didn't have as much seasonality as, in fact, we did.

And I think as James pointed out, maybe just to add, historically, we were 90% point-of-sale centric. We have, with the addition of our other products, reduced that to about 80%. Now the 80% is still a higher number than the 90% but -- in terms of actual units, but our dependency on the seasonality will fade as we become -- as we spread out the business.

George, there was 1 other factor in the first half of 2016 which was also -- and as well, and that carried forward through the year and that's -- we were still in the final stages of phasing out our handheld computer business.

So we had $2.2 million of computer business in 2016. We discontinued it as of the end of June. And then -- but we still saw some residual activity as well over the next couple of quarters. So all of those helped fill in the seasonality factor in 2016.

Okay, okay. And then one quick last question. In '18, so point-of-sale declined from 90% to 80%. So of course, it means the rest grew. What kind of mix do you expect in '18? Do you -- first of all, do you expect the point-of-sale on a dollar basis to keep growing? Or is that flattening out? And what are the major other verticals that are driving the growth?

So first, we do not expect point-of-sale to flatten out, we expect to continued growth in the point-of-sale business. And I'll let James answer what other areas we expect to see.

Yes. I mean what we're seeing is the other areas which, for us, until we built our DuraScan product, were living off of our nondurables and really needing a durable. There we're actually growing faster, right? So as Kevin said it earlier, they're -- it's a more modest number. So we're starting from a low base, but we're seeing a lot of strength as our DuraScan product gains a reputation as a reliable product.

And those markets are primarily things like commercial services, which includes that things like event management, service companies, applications or things like order management, utilities monitoring, equipment management; industrial and manufacturing, those are things like quality control, auditing, process control, transportation and logistics that has a lot to do with track and trace, picking and put away or order fulfillment; and health care, which is primarily around health care process control.

Those are markets where our durable product -- it was really designed for those spaces. They have IP requirements, they have concrete floors and durability requirements. And until we had that DuraScan product, we weren't able to satisfy them as well. And we're seeing the DuraScan product starting to do that.

Okay. So DuraScan would be driving a big chunk of your growth in '18?

Yes, it's driving it directly and indirectly. Directly because, obviously, it fits the needs. But as Socket fits the needs, the nondurables also begin to have an opportunity in those markets, and we're seeing a lift there as well. So those markets, in general, are performing much better than they were historically.

And we have no further questions at this time. I'll turn the call back over for final remarks.

Thank you, operator. We would just like to thank everyone for participating in today's call and to wish you all a good afternoon. Thank you.

Thank you, ladies and gentlemen. This concludes today's conference. Thank you for participating, and you may now disconnect.