Greetings. Welcome to the Fastenal 2019 Annual and 4th Quarter Earnings Results Conference Call. At this time, all participants are in a listen only mode. A brief question and answer session will follow the formal presentation. Please note this conference is being recorded.
At this time, I'll turn the conference over to your host, Ellen Stoltz, Assistant Controller. Ellen, you may now begin.
Welcome to the Fastenal Company 2019 Annual and 4th Quarter Earnings Conference Call. This call will be hosted by Dan Thornis, our President and Chief Executive Officer and Holden Lewis, our Chief Financial Officer. The call will last for up to 1 hour and we'll start with a general overview of our annual and quarterly results and operations with the remainder of the time being open for questions and answers. Today's conference call is a proprietary Fastenal presentation and is being recorded by Fastenal. No recording, reproduction, transmission or distribution of today's call is permitted without Fastenal's consent.
This call is being audio simulcast on the Internet via the Fastenal Investor Relations homepage, investor. Fastenal.com. A replay of the webcast will be available on the website until March 1, 2020, at midnight Central Time. As a reminder, today's conference call may include statements regarding the company's future plans and prospects. These statements are based on our current expectations and we undertake no duty to update them.
It is important to note that the company's actual results may differ materially from those anticipated. Factors that could cause actual results to differ from anticipated results are contained in the company's latest earnings release and periodic filings with the Securities and Exchange Commission, and we encourage you to review those factors carefully. I would now like to turn the call over to Mr. Dan Fournis.
You, Ellen, and thank you everybody for joining us for our Q4 earnings call. Last, we'll be filing our annual report here in the early part of first half of February. And so last weekend, I was drafting the letter to shareholders and employees for inclusion in that annual report. And one of the observations contained in that letter is a comparison of 20192015, because when I was when I'm writing that letter, I typically will go back and reread a few letters. And one reason you do that is so you're not too repetitive.
Another reason to do that to see what observations jumped out at the time. Back in 2015, we started out the year with an PMI index of 52.6%. That's the average of the 1st 3 months of the year as we disclosed back then. I'm not sure if there's been adjustments to the PMI since then. We ended the year just shy of 49%, I believe it was 48.6%.
And in 2019, we started the year a little bit higher. It was at 55.4% in the Q1. In Q4, we were at 47.9%. So and we've been sub 50 since August. If I look at it using an internal benchmark or an internal yardstick, and I looked at our top 100 customers, and this group represents about 25% of our revenue historically.
In the Q1 of 2015, about 72% of those 100 customers were growing. By the Q4, that number was 49%. And given our sizable relationship with each of these customers, the change in trends in a window like that is usually more about their underlying business than FastNOG gaining or losing market share with that customer. If I look at that same statistic in 2019, we started the year at 81%. 57%.
Again, it's looking at the percentage of our top 100 customers that are growing with us. And I think that's a great barometer on the marketplace. If I yesterday in our Board meeting, one of our directors asked, hey, if you're looking at Q4 and the question was about Q4, my comments here, second goal was on 2019 in general. What would you give for a grade on the quarter? And without giving it a second of thought, I looked at them and I said, you know what, I'd give it a B.
And the biggest reason for the B is we did not leverage our earnings. In other words, our operating income growth was less than our sales growth. And I just can't give an A to that performance. So I agree to the B. However, if I look at our team, leaders throughout Fastenal and the execution of the team throughout the organization, I think given the subdued activity as indicated externally in the PMI index as well as internally in our top 40 or top 100.
I believe the blue team executed well in the Q4. December, as you all know from reading our release, December was a tough month, we grew 1%. The midweek holiday over Christmas was not our friend. The January and been doing this, I've been at Fast On Offer 24 years. And that means I have done just shy of 100 earnings calls.
And there's one thing we don't do and we don't talk about current month and current month. I'm going to break that rule. January started with a midweek holiday. That Thursday Friday, 2nd, 3rd, was not our friend and we really haven't recovered from that. And I expect at this point, January will feel a lot like December.
I don't know if that means it will be 1%, But I think it's going to feel like December did. Our sales team is adamantly opposed to that. They see reasons why we'll do better and time will tell. I genuinely hope they prove my expectation wrong at this juncture in the month, but I'm just sharing that. This has absolutely no bearing on my thoughts for February or for 2020 in general.
Time will tell what the economy is going to provide for us and what headwinds are created or tailwinds are created. But December ended with a midweek holiday and January started with 1. So I want to make that observation. Flipping over to Holden's flip book. The I've touched on that second bullet where really we talk about December weakness and the impact of the holiday timing.
Despite everything that's going on and my comment about my belief in the Blue team and being impressed by their executing ability, we leveraged our operating costs in the Q4. We continue to focus on controlling our costs, not as a means just for the expense, but to put us in a position where we can continue to invest in our growth drivers because we still have incredible opportunities for growth. We have a relatively small market share and I truly believe we have a better supply chain model for our customers. As we talked about last quarter, the moderation and challenges of inflation and tariffs and our ability to manage through it has stabilized the price cost. Holden violated a rule last quarter and that he talked about his expectations for gross profit.
I suspect he won't do that again, but what time will tell. But one observation I'd have for everybody listening to this call, don't get caught up in the minutiae. The real issue in Q4, our sales fell off a bit more than we expected and we have an incredibly large fixed cost trucking fleet that operates. And whether we're bringing 1 pallet or 3 fourths of a pallet product to a branch, We're running a truck. It's still a stop.
And that delevered more in the Q4 than we expected. To add a little to that pain, our freight revenue also weakened. That's a short term execution issue. If I were to attribute to anything, I think there's a bit of fatigue in the organization from all the tariffs and pricing and inflation energy that expelled in the summer and fall months. And I think it showed up in our Q4.
That's an execution issue on Fastall and that's on me. But don't get caught up in the minutiae of a gross profit, a 10th of a percent or 20 basis points. It's deleveraging of a freight network. And that happens always in the Q4. This year was a little bit more acute.
Flipping to the Page 4 of Holden's flip book. On sites, we signed 3 62 for the year. Our stated goal was 375 to 400, came in just shy of that. We ended the year with 11 14 active sites, about a 25% increase from end of last year. Sales growth with this group, and this is excluding transferred sales.
So this is looking at peer growth. So if we have a $30,000 customer that pulls all the branch and we go on-site, it's looking at dollars above $30,000 It was up in the low double digits. As you can appreciate, in our more mature Onsites, we did see some degradation and that's a sign of the economy. Going into 2020, our goal remains 370 to 400. We're really excited about this growth driver in our business.
Total in market locations were 3,228 versus 3,121 at the end of last year. We closed or converted 36 locations, traditional branches. We also closed 26 on sites in the 4th quarter. Now that probably raises a few questions in any of your eyes. Here's a couple of things to always keep in mind while fast and all.
Some organizations prefer only showing pretty pictures. Some organizations only prefer talking about good things. We talk about the real. We talk about the things that change. We address today's issues today.
One of the things the team at Fastenal has learned to live with over the last 5 years is every year I share with them what I consider my top ten suggestions on life, things that I learned from friends and family as a kid, and that includes where I grew up, family was also your neighbors. It includes teachers and coaches I was blessed to have in my life. It also includes the people you associate with in your adulthood. It starts with your spouse, extends to your kids, but it also extends to friends and associates you come across. Number 5 on that top ten list is make great decisions.
Share the reason why and start today. If we have an on-site location where that customer's business has downsized, maybe they've closed that facility, maybe the economics don't work and we decided to take that relationship and move it back into the branch, that's not a sign of failure. That's a sign of wisdom. That's a sign of saying how are we allocating our resources and what's the best resource allocation for our customer, for our next customer, for our people and for our business. And I consider those decisions to close an on-site that doesn't meet the requirements to be a good decision provided we don't think it's going to recover in a reasonable timeframe.
Exiting an on-site doesn't mean you don't come back, doesn't mean you lost customer relationship, it just means the economics for being on-site aren't holding true right now. You take a step back. I think that's a great decision and we will make those decisions every day. From a vending perspective, we signed 5,144 devices in the 4th quarter, essentially in line with our in 21,857 for the year, essentially in line with our goal of 22,000 devices. Our installed base ended up at 89,937, an increase of about 11% from last year and our product sales through those devices were up in the low double digits.
In the order of avoiding confusion, earlier in the year we celebrated vending machine number 100,000. This 89,937 includes machines that are principally producing product sales. We have another 15,000 devices out there that are leased out to customers for check-in, check out purposes, just to clarify that. Finally, e commerce, our sales grew at 25% Q4 to Q4. For the full year, we were up 32% and this includes a 35% increase with our national comp customers.
The way we think about e commerce, it's about making our business a little bit more efficient every day because as our supply chain partnership with our customer grows, most of our business activity is coming from vending, from bin stocks, where the customer really isn't ordering new product and we keep stepping deeper and deeper into the business. Today, those two pieces are about 30% of our revenue. I believe in the future that number will more than double as a percentage of our business as we come more in trends in our customers business. This is really a reflection of stuff that customers order outside of that supply chain window, the non recurring stuff. With that, I'm going to turn it over to Holden.
Great. Thanks, Dan. Good morning. Let's just jump right into Slide number 5. Total sales were up 3.7% in the 4th quarter, which included a particularly poor December growth of just 1%.
December was affected by holiday timing and extended customer shutdowns, but even setting these aside, business activity was generally soft throughout the quarter. This was reflected as well in the macro statistics with the leading U. S. Purchasing Managers Index averaging a contractionary 47 point 9 in the Q4 and printing a 47.2 in December. U.
S. Industrial production swung to a decline of 1.1% in October November. Manufacturing was up 5.1%. Heavy equipment lagged at 1.4% growth with weakness as well in oil and gas, metals and transportation. Construction was up 3.1% in the 4th quarter, consistent with the 3rd quarter.
Larger construction customers again outgrew smaller ones, though we have seen a moderation in the rate of contraction in those smaller customers. In the Q4 of 2019, we sustained the pricing that was implemented and realized in the Q3 of 2019 as a means of offsetting general inflation and tariffs. Clearly, things could change, but as we enter 2020, weaker end market demand and a more encouraging tone around trade policy has reduced inflationary pressure. From a product standpoint, non fasteners continued to lead but did decelerate with 5.1% growth in the 3rd quarter, while our more cyclical fastener line was up 1.8%. From a customer standpoint, national accounts were up 8.2% in the quarter with 57 of our top 100 accounts growing.
Non national account sales actually contracted in the period with just 54% of our branches growing in the Q4. November December can be difficult months from which to draw conclusions about the future and that is certainly the case this year. However, business conditions clearly remain sluggish and the feedback from our regional leadership remains cautious on the early part of 2020. It remains our intention to take advantage of this environment to continue to invest in our growth drivers even as others pull back. Now to Slide 6.
Our gross margin was 46.9 percent in the 4th quarter, down 80 basis points versus last year. This annual decline is primarily a function of product and customer The 30 basis points decline sequentially was in line with normal seasonality, though we were a bit shy of our expectations that gross margin would be at least 47%. This is primarily attributable to freight, as Dan commented on. While the cost of maintaining our captive truck fleet is fairly stable quarter to quarter, our ability to charge for freight to partly offset those costs can vary depending on market conditions. In the Q4 of 2019, our freight sales declined 7.5 percent, which resulted in less absorption of our fleet costs than anticipated.
With our pricing sticking from the 3rd to the 4th quarter and with weakening demand causing inflationary pressures to moderate, price cost was not a meaningful variable one way or the other affecting gross margin in the Q4 of 2019. Our operating margin was 18.7 percent in the 4th quarter, down 30 basis points year over year. We believe the Blue Team did a good job controlling expenses as SG and A as a percent of sales of 28.2% was better by 60 basis points and a record low for a 4th quarter. However, as we have commented before, given our intention to continue to invest in growth drivers even as conditions slow, it will be difficult to defend the operating margin when growth eases into the low single digits. Our inability to leverage occupancy costs in the Q4 of 2019 are a perfect example of this, with current growth not being sufficient to produce leverage over the vending investments that drive our market share gains.
Looking at the other pieces of SG and A, we achieved 20 to 30 basis points of leverage over employee related costs, which were up 2.4%. This was largely due to reduced FTE growth of 1.4% combined with lower performance based compensation in the period. In the current environment, we would expect further low single digit growth going forward. We realized 30 to 40 basis points of leverage over general corporate costs with lower bad debt expense in the absence of certain legal and structural expenses incurred in the Q4 of 2018, offsetting higher IT spending in the Q4 of 2019. Putting it all together, we reported Q3 2019 EPS of $0.31 up 5.4 percent from $0.29 in the Q4 of 2018.
Turning to Slide 7, looking at cash flow, we generated $252,000,000 in operating cash in the 4th quarter or 141 percent of net income. Higher earnings combined with lower working capital needs produced the cash flow. Full year cash conversion was 107%. Net capital spending in 2019 was $240,000,000 exceeding our targeted range of $195,000,000 to $225,000,000 The sources of higher capital spending in 2019, vending trucks and hub capacity were planned. The overshoot related to our addressing a number of facilities in 2019 as a means of managing a sharp increase in our volumes over the past 3 years, 19 as a means of managing a sharp increase in our volumes over the past 3 years and a couple of those facilities simply costing more to complete than we anticipated.
In light of weaker macro conditions, we expect net capital spending of $180,000,000 to $205,000,000 in 2020. We paid nearly $500,000,000 in dividends in 2019 and increased our dividend for the Q1 of 2020 by 13.6%. We finished the quarter with debt at 11.5 percent of total capital, below last year's 17.8%. Inventories were up 6.9% in the Q4 of 2019. Nearly 3 quarters of this growth related to inventory to support on sites.
Hub inventory growth slowed sharply, a function of both deliberate efforts to reduce inventory as well as weaker demand, while field inventories were slightly down. Setting aside the impact of demand, we believe there remains further opportunity to improve our inventory days in 2020. Accounts receivable grew 3.9% in the Q4 of 2019, the slowest rate of increase in more than 3 years, which largely reflects slowing demand. Mix and customer behavior will likely continue to influence AR days in 20 with our expectations being flat to slightly higher days. We believe these factors will continue to produce good operating and free cash flow in 2020.
That's all for our formal presentation. So with that, operator, we'll take questions.
Thank you. At this time, we'll be conducting a question and answer session. Thank you. And our first question is coming from the line Ryan Merkel with William Blair. Please proceed with your question.
Hey, thanks. Good morning, guys.
Good morning.
Good morning.
So first off, in terms of price capture, you said it was consistent with last quarter. So I just want to confirm, does this mean pricing was up 90 to 120 basis points? And then sort of a related question, I recall that you thought you'd get some incremental price into 4th quarter. Did this not materialize? And why was that, if that's the case?
No. So what I would say is, if the impact of price in the Q4 is probably in the neighborhood of 70 to 100 basis points, which would have backed off a little bit from where Q3 was. But again, much as we saw much of the year, that reflected having to grow above some increases from last year. If I look at the overall price level in our business in Q4, it was slightly up from where we were in the Q3. So we look at the overall pricing activity from Q3.
And for the most part, I would characterize Q4 as living up with our expectations as it relates to the just
yet? We aren't seeing price pressure materialize. Now,
just yet? We aren't seeing price pressure materialize. Now what we're commenting on has more to do with inflation. Remember, what we've said is we've had 2 things affecting us over the past 12 months. 1 was the generalized inflation.
The second was the tariffs. And frankly, the generalized inflation was a much more significant impact on us than the tariffs were. What we are seeing is with some of the activity around tariffs of late and with the overall sort of slower demand level, some of those inflationary pressures that have been a part of our business for the past 12 months to 18 months have begun to moderate. But that shouldn't be read to suggest that we haven't slight color to that. First off, there's always price pressure.
That's not a constant in life and it's intense now. If I but I make that comment again any quarter. So one element that does exist in the Q4, that does exist in the Q1 is keeping the facts straight and the confusion at a minimum. And what I mean by that is, we've talked in prior calls and we've talked internally continuously about the tariff and antidumping duties that are out in the marketplace. You have the 232 antidumping as it relates to steel and aluminum that was put in place.
And then you have the 301 tariffs. So the 301 tariffs and I get an update on this every 2 weeks. And even that's not enough in many ways, because there's so many pieces that are flying around. I suspect if I'm in a branch in Fastenal, I'm having a customer talking to me in December and saying, Jeez, the tariffs were just eliminated. Now the List 4B that never went into effect that was had been postponed was canceled on December 13, but that number went into effect and that was largely consumer goods.
So that impact of that is less in our world. Then the next comment might be well, list, I thought it went from 15 to 7.5. Well, it was just announced that List 4A is going from 15 to 7.5 and it's taking through that piece. For us, the bigger one was frankly List 3. Nothing's changed on List 3.
So our biggest challenge internally is to make sure that an individual in a branch, a branch manager, a district manager, a regional leader, a national accounts person is as familiar with these pieces as what I just laid out. Because when you're familiar with the facts, you can have a discussion. If you're not, it's very difficult to have a discussion about it. So that pressure is there primarily because of all the noise and confusion that can come into the marketplace and a lot of the misinformation that can float around. But the fact is 4B never went into effect.
4A, it was just announced that it's dropping. And but List 3 is where the dollars are, at least for our marketplace. List 1 and List 2 are less meaningful.
Yes, that's helpful color. Yes, I think we're all sort of interested to see how that materializes if tariffs are removed, what happens with pricing and such, but I guess TBD on that. And then on the On-site, you mentioned the closings and you mentioned the economics. I'm curious, is the reason for the worst economics due to the slowing economy or is this more of the local team didn't assess the opportunity correctly and this is something you can fix?
I think it's typically economic. I'm sure there's examples of ones that we decided either to pull it back because it didn't materialize as what we thought. Keep in mind, and I've used this example before, I remember traveling with a district manager several years ago and he had an on-site that when we were driving to the on-site, he said, yeah, and this individual knows as much about on-site as anybody in our company. He's based in Wisconsin. And he said, I'm taking you to this one and I'm not really sure if it was a good idea or a bad idea, but here's why I did it.
And he talked about going on-site. He talked about what the relationship was and he talked about where he thought he can get it to. And he thought he can get it to 75,000 a month, which is a small on-site for us. But he had a real frank discussion with the customer that, that, I'm only going to be able to staff it this many days a week, but it gave him the freedom to move the branch a little bit further north and it was a better location for the branch. There he was looking at and saying, if it doesn't work out, I pull that back into the branch.
As it turned out, he got it to 75 and a couple of months later, he sent me a message and said, hey, we picked up some OEM parts and we were to get it over 100. That change occurred because the customer loved what we were doing and they figured out ways to buy more from us. That doesn't always happen. And so sometimes you might have one where you're going in on an on-site and you think you can get it to X and you find out that you can only get halfway there, that 30 got to 80, but you can't get to 130 and you pull it back. Don't consider that a failure.
We took a $30,000 relationship grew to 80,000. We're having a frank discussion with the customer. It's easier for us to service it from the branch. I consider that we're engaging with that customer and it's win. And if that's 1 of the 26, I'm fine with that.
I'll add a little color to that as well. I mean, I think looking at the causes of the closures this quarter, there were really no differences versus the ones that we covered with you last quarter, right? So it's all sort of the same reasons. And we have to remember that in those, as Dan alluded to, some of those are simply taking business that we had at On-site and moving it to a branch. So we didn't see anything different in sort of the character of those closures.
And I think you have to think about the larger installed base. Once you get to a certain level, you do review that base for underperformers. And again, as Dan is saying, I think that's a healthy transition. But I think you should also think a little bit about the culture of the business. I mean, if you think back to vending, I think we started that initiative in 2,009, but it really began to get legs in 2011.
And in 2016, we sort of went through and we looked at all the machines we put in and said should this machine really be here, right? And that year, we had a higher than normal sort of removal rate than we had seen in previous years as we sort of evaluated the installed base. And we're in 5th year of this on-site push. And I think that our culture in some respects is to sort of find these growth drivers, run fast, grab ground and clean up a little bit later. And at the end of the day, what you're seeing right now is we built a good installed base.
We're evaluating what we have. The churn has gone up a little bit. It's down from Q3. And frankly, I think we're expecting a lower rate of churn in 2020. We'll see how that plays out.
But in the end, what you wind up getting, much as you did with vending is, a great business where we're ahead of the field in terms of being able to implement it and it contributes to gaining market share. And we're looking at this very much the same way. The other thing to your point about opportunity, this past quarter, the on-site team actually sampled 400 locations. I'm not sure which 400 those are, but they wanted to sort of get to the same question that you're asking. And these were decently sized 400 Onsites in terms of the average, but they found that in the products that we currently deal in, there's still more than 2x the revenue potential than we already have in those Onsites.
And again, these aren't smaller Onsites. So the potential is we don't see any diminishment in the potential for this business. And I think what you're seeing us transition to, I would say, is historically consistent with how we execute our growth drivers and leading have a great business.
Our next question is from the line of Robert Barry with Buckingham Research. Please proceed with your questions.
Hey, guys. Good morning.
Good morning. Good morning.
Could you say or could you how much the freight was a headwind to gross margin? It probably impacted us by 10 to 20 basis points relative to the expectations.
And you think it now that just goes away or that persists as long as the sales growth is low single digits?
Well, I think as long as sales growth is low single digits, there's going to continue to be some pressure there, for sure. And that's not new and that's not necessarily unexpected. I think Dan also talked about, we just have some focus to bring to that, right? We spent a lot of effort in the last 12 months on pricing discipline on the product side of our business. And the degree to which you've seen us not really have to talk about price cost is a direct result of the success of that, right?
We've been able to mitigate some of those inflationary pressures. We need to bring some of that discipline to the freight side of it as well. And that's going to be a focus and a goal for 2020. And the degree to which it is or is not a drag will depend on our level of success. But again, we always have to put this into perspective.
We're talking about 10 to 20 basis points. Now we want those 10 to 20 basis points, right? But we're not talking about 100 basis points of risk here, 150 basis points. And we're also talking about something which is related to an asset, that being our captive fleet, which is a huge competitive advantage for us. So there's things that we have to do better.
It's focusing on that will be something we do in 2020. We do have to address the market. That doesn't make it easier, but there's some things that we can do from a self help standpoint. And that's probably how I'd characterize the situation with freight. Rob, the only element I'd add to that is part of it is about year over year growth, but part of it is about sequential.
So if you think about where we were in October to November December, the delever that's occurs because you have this infrastructure to support that and it falls off. History has shown where January is relative to October. And depending on what the start of January will be in that normal landscape will still come into play. But once you emerge from that slowdown that you get around the holidays and you get further away from that and the dollars climb back, then that delever that occurs from our distribution, our trucking freight, our trucking fleet that dissipates because we not only want that 10 to 20 basis points back, message to our team is here's when we expect to get it back. And then to Dan's point, don't forget seasonality as well, right.
Again, the crux of the issue is that you have relatively stable costs in your fleet and seasonalities. You get into Q2 and Q3, you get a seasonal uptick as well just naturally. So Dan's right. Part of this reflects simply the seasonal nature exacerbated by the cyclicality. So that is an element of it, but we're going to work on executing it better as well.
Got it. I guess bigger picture, where do you see price cost going from here? It sounds like if the non tariff inflation was actually the bigger piece and we're seeing commodity deflation. I don't know how do you think about the puts and takes because I think you said the tariff inflation will peak in 2Q, but if you're seeing some meaningful commodity deflation like steel, I mean, could you see price cost, I don't know, even be modestly positive? You see any line of sight to that?
Or how should we think about it tracking?
Hey, Rob, I'll make a deal with you. If you can tell me what tariffs are going to do in the next 6 months, I'll tell you what I think our reaction is going to be. I've had numerous times over the last year where I have a discussion on Friday. And by Monday what we just talked about had completely changed and the discussion became irrelevant. So I'm going to respectfully decline to answer that only because I don't know sometimes from week to week and month to month what's going to happen with tariffs.
And if it makes it the economy, I can look at trends and look at prior years in Fast Stone, I could at least give an informed thought on tariffs, we really can't. Yes. All I know is we focus every day on shedding the best light we can for our customers on building the best supply chain for our customers, and where appropriate, moving sources of supply. And that we've done quite aggressively in the last 10 to 15 months. And the only thing that I would probably add to that is, over the past 6 to 9 months, the pricing teams, the way they've been sort of structured, the work they've done to sort of surface a lot of our challenges, the tools that they've deployed to allow us to react to events, we're in a far better position today than we were a year ago to manage whatever happens, right?
So to the extent that tariffs go up or go down, we'll be able to adjust to that in the way that we promised our have to see we'll have to see how that plays out. But I think that it really is it's the other side of the same coin, right? I mean, many of our national account contracts have terms in it that are linked to specific steel price indices or what have you. And so those conversations get had. In the end, the question that we have to deal with our customers is what's the right action to again be able to neutralize the impact on gross margin.
We're not trying to take advantage of our customers. But at the same time, we're just looking to deploy the tools that we developed in the last few months, last couple of quarters just to essentially neutralize the impact and that would be the intention. But again, we'll have to see how things play out as the year goes on.
Got it.
Okay. Hey, just lastly, could you say how December was tracking before the last Hey, Rob. Yes. Okay. I'll take you through.
No problem.
Yes. No problem. Thanks guys.
You bet.
Our next question is from the line of David Manthey with Baird. Please proceed with your question.
Hey, guys. Good morning. Good morning, David. Over the past couple of years here, your gross margins have pretty consistently been down 100 basis points year over year and some of that secular or some cyclical. And just as we look ahead to 2020, if the decline is less than that, what would be the factors you would see as driving a better outcome?
Yes. So I think there's 2 things to think about next year in terms of big sections. And assuming that the price cost dynamic remains neutral, as it has been the last couple of quarters. And that's obviously going to depend on anything that occurs that we have to react to, but we don't know what that looks like. Based on what we know today, I think that with the slower small customer business where we have fairly good margins and expectations for growth on the Onsites, I think you're looking at mix probably being a 50, 60, 70 basis points drag next year.
Now that could change depending on the relative growth between the small customer and the large customer, but that seems like a decent range to me. I think that we also have to think just about the cadence of the price trend from 2019, right? I think in the first half, we did not do a great job addressing inflation and addressing tariffs. Actually, we did a better job on tariffs than we did inflation. But the in the second half, I think we did a very good job on that.
And I think that success will carry over into the first half of twenty twenty. So you might have some easier comps in the first half of twenty twenty before that sort of normalizes in the back half. And I think those are the 2 big things to think about when you're thinking about how to model our gross margin next year. And then, yes, I think those are the 2 big things to think about at this point. What I'd add is if we're able to shallow that out, It's some things that we're doing, specifically inside the organization related to some of the products.
For example, we're challenging and this is ongoing. We're always challenging our folks from the standpoint if I think of our vending platform. So that's highly repetitive transactions. How do we channel as much of the spend to a select group of preferred branded suppliers and our own exclusive brands, our private labels. From the standpoint of it, it's not just about the cost of the product, it's about the cost of the whole supply chain because it's product we have on the shelf.
If it's on the shelf, our distribution centers operate more efficiently. Our trucking network and our freight costs are more efficient. Our local labor costs are more efficient. And the first two of those three pieces impact our gross margin. The third obviously is operating expenses, the branch labor.
But it allows us to have a product offering in that machine or in our ongoing supply chain offering where we can offer a better price to the customer because our cost structure is inherently lower and that typically helps our gross margin. So we're sharing some of the benefit with the customer. So it's a win win scenario and we're driving more spend to our preferred suppliers.
Great. Thanks guys.
Thank you.
Our next question is from the line of Nigel Coe with Wolfe Research. Please proceed with your question.
Thanks. Good morning, guys. Great color on the gross margins, Holden. Just thinking about you normally give forward quarterly guidance. We normally see gross margins in 1Q pretty flat.
It feels like Q over Q, and it sounds like that holds true in 1Q 'twenty. Any comments on that?
Well, a couple of things. First, I would dispute the idea that I normally give forward guidance. Dan commented on how we rarely talk about the current month and the current month because it's hard for us to know what's going to play out until it's all said and done. But sometimes things are moving in a direction where you deserve a little bit of guidance. And for the last two quarters, I've given you a little bit of forward guidance on the next quarter because we had a lot of moving pieces as it related to the inflation, the tariffs, the success around pricing, etcetera.
And when your Q2 gross margin is down, I think it was 180 basis points. I think you all needed a little bit of guidance about what we were seeing and expecting. And so I provided a little bit of that color. But the idea that we usually provide that color, I would disagree with and I'm not going get into that game now that frankly, I think most of these moving pieces are kind of stabilizing out. So I'm going to remove myself from that game until a future period where there's a lot of volatility that you need a little bit of guidance on.
But you asked about the seasonality. At this point, if I assume that the trends in the back half of twenty nineteen are sort of the ones that will carry over into 2020 with sort of a neutral price cost dynamic and then our usual mix piece and all that, I think that judge what you think the seasonality is going to look like and I think that's a reasonably that's a reasonable way to approach the cadence of the quarters.
Okay. Got it. And I didn't mean to whiz him out, but thanks for the color, Holden. Just look, my question is around the attrition on the on sites. I'm just now that we've got to a base of over 1,000, some level of attrition is normal and should be expected.
So I'm just curious if you can think about what you consider to be as a CEO, CFO, a normal rate of attrition going forward. Is 2% per quarter the right number? Is it lower than that? But maybe also if you could just clarify the freight issue? It sounds like that's I understand the seasonality and the absorption, but that just sounds like it's more of a price and obviously we're seeing LTL and TL rates obviously down.
So I'm just curious if we should be tracking price of freight volumes as a proxy for what's going on, on that side of your business. But my real question is on the attrition on the on sites.
I'll touch on the on-site attrition and the second one might count as a third. So I'll let Hailey answer that. We'll take through that one, Nigel. If I think of I'm going to answer it first in the context of vending and then I'll transition to On-site. So if I go back a few years ago, Holden mentioned in 2016 the number of vending machines we're moving.
It actually started before that. So in 2012, our vending really took off. And we pulled out if you look at the number we had at the end of 2012 and how many we pulled out in 2013, it was 20 some percent of the devices we pulled out. Because we've gone from a handful of district managers doing vending to a big chunk of district managers doing vending and a big chunk of our branches doing vending and we didn't really have it dialed in where they should go and because it was a new industry we're creating. And that 20 some number was 20 some percent the next year.
And then it dropped I think to 17. I don't have the stats in front of me, but then it dropped to 16. We believe long term that number is going to settle in around 10%. Right now, we're at about 14. So 14% of the installed base, that 89,000 we cited a few minutes ago, I would expect, would get removed based on history in 2020.
And our goal is can we get that down to 10? If I looked at on-site historically, this is back when we had a couple of 100 of them. I used to figure we'd close about 10 a year. That's about 5%. I think if Chris Van Dallen were sitting here and Chris leads our team that really drives and owns our on-site model, the team that is involved with evaluate our on sites.
I think the number he typically thinks it is about 6%. So how that plays out quarter to quarter is anybody's guess. But it's probably in that neighborhood. History has said it's going to be it's probably 5. A person who is more informed than me thinks it's going to be 6.
In the interest Nigel of not encouraging people to ask one question with 12 parts. We'll address your second question when we talk after the fact.
Okay. Thanks, Ed.
Thanks.
Our next question is from the line of Hamzah Mazari with Jefferies. Please proceed with your question.
Good morning. Thank you very much. My first question is good morning. Dan, a number of years ago, you had something called pathway to profit and 23% op margin was sort of the goal. So maybe do you view that as achievable still and any thoughts on that target as you look long term?
Yes. So that was pretty close to that one back in the day. I'll touch on that. In fact, I talked about pathway to profit in this year's letter to shareholders. So there's a little bit more insight you can gain from that in early February.
But the pathway to profit was about our branch network and about the fact that as our average branch revenue went from $70,000 to $80,000 I'm talking about the revenue per month, so $70,000 a month to $80,000 a month to $90,000 to $100,000 and beyond. As that number went up, we had a whole bunch of branches out there that were doing $120,000 $150,000 $200,000 a month And we already know what the profitability is of that group. So what we talked about in 2007 when we introduced Tapway to Profit was, we're opening fewer branches today, which means the dilution factor is going to go away and the average revenue per month is going to keep climbing. And as it's climbing, the inherent profitability of our branch based model will shine through. We had a whole bunch of branches that were deep into the 20s, But a branch that does $50,000 a month doesn't have that kind of operating margin.
A branch that does $70,000 or $80,000 a month doesn't have that kind of operating margin. So that 23 is still true for our branch network, but branches represent about 70% of our business today and the other 30 is On-site. For On-site that number is down in the upper teens from the standpoint of where the profitability is. There's still a pathway to profit element of it. When we go from having 200 Onsites at the end of 2014 to having just over 1100 on sites today, that's a massive takeoff.
And so our revenue per on-site actually dropped because we opened so many new ones. And so we've been living through that for the last 4 years of that deleverage of our on-site business because our average revenue dropped. We're approaching now an inflection point where our steady state of how many we're adding is going to kind of neutralize. And our revenue per on-site when we get later into 2020 and into 2020 one is actually going to start inching up. And so that will still come into play.
On a previous call, I believe I cited the fact that I think our branch network, over time that branch network moves ever closer to 200,000 a month. Today it's in the I forget the exact number we cited, but it's in the 100 and 30, 135 neighborhood I believe. But as that moves up, I would expect that profitability piece to improve. As the mix changes, I see a path to an operating margin where you have a fifty-fifty business of a branch and on-site in that low 20s. I've cited the number at 22%.
Time will tell if I'm right or if I'm full of it. But the number you cite, Hamzah, was really about the branch network, which is a subset of our business.
Got it. Very helpful. My follow-up question, I'll turn it over is, on freight, how much of a cost advantage is your captive fleet? And do you view that cost advantage as off offsetting execution risk? Just any broad high level thoughts there?
Thank you.
Look, well, the cost advantage is a constant, so it doesn't offset execution risk. Execution risk stands on its own. But the cost advantage, we've always felt that running on a highly utilized trucking network, where you have products going and by highly utilized, I mean there's products going both directions. The number we've often cited, if you trust what it cost to move it on our trucking network versus shipping industrial products small parcel. And shipping a relatively low value product small parcel is incredibly expensive.
And that's where the e commerce world lives. And that's why they keep trying to build different types of networks to change that dynamic. We've often cited the relationship of 10 to 1, that it's about 90% cheaper. But again, we're moving a relatively low product on a captive trucking network. A third of our products, so if I look at our Fraster products, there's a big chunk of that product category that has secondary operations.
So it isn't just that we're moving it from supplier to the branch to the customer or from the supplier to the customer with the on-site. We might be sending that faster product out for a secondary operation. It's going to go get heat treated. It's going to go get plated. What it means in our faster industry and the reason we built the captive network to start with is many of our competitors because they're using a lot of LTL shippers and other parties don't have that advantage.
And so their freight costs in many cases can become equal to, if not greater than their product cost. And so one of the reasons we enjoy the gross margins we do in fasteners and the reason we built the trucking network is we've just built a better means to move product around. But execution risk, they're disconnected in the standpoint of what you're doing and what are your habits and behaviors every day. But I think it's fair to say, Hamzah, that the value that affords us because of our captive truck fleet is well in excess of a 1 quarter 10 or 20 basis point lower result than we expected. We get huge dividends from the ownership of our captive fleets.
Just given the dividends we raised there.
Great. Well, thank you very much.
Thanks.
The next question comes from the line of Josh Pokrzywinski with Morgan Stanley. Please proceed with your question.
Hey, good morning guys.
Good morning.
Dan, if you might
mind commenting on how we should think about mix kind of historical recoveries. Dan, you led off talking about maybe the analog to 2015 that you saw. If I were to look at similar points in time or other periods where you think it feels like today, what is the mix on the way out? Is it that the on sites recover faster because they're captive? Is it that the kind of the smaller customers or non national accounts where there's higher mix recover faster?
Just trying to calibrate how we should think about mix as markets eventually recover?
Yes. So if I think about my comparison to 2015 was more about just a trend in what we're seeing in the numbers. The other observation I'd make looking at 2015 is, in 2019, we started at a higher point and we finished at a higher point. And I really attribute that to the success we've enjoyed taking market share at a faster clip and the backlog we have in energy. I mean, in the last 3 years, we've signed either renewals or new contracts for about 1500.
That's an incredible tear that we've been on. And that builds a certain level of just market share gains you get. And so when I look at our relative outperformance in 2019 versus 2015 in a similar PMI index scenario, it's what the team has done to engage with our customers. And we have a massive footprint. So to the extent that the percentage of our top 100 snaps back and it moves into the 60s and it moves into the 70s because of there's a bit of a lift in the economy, that piece will snap back pretty fast and our large account Our accounts with a contract relationship would grow disproportionately faster because they've contracted disproportionately faster.
And but and that's just a reflection of recovery in the underlying market. So I would expect that to play out. The speed at which is going to be dependent on what the market is doing and maybe holding it as more insight. Yes. I mean, I think this is an area that I think can be over thought many times.
I mean, the fact is whether it's an on-site, whether it's non fasteners, fasteners, most of our businesses, essentially all of our businesses are cyclical. And so when you see an upswing, it's fun talking about an upswing by the way, I hope it happens. But if you see an upswing, what happens is you see the growth rate at particularly our older on sites get better, just as you see the growth rate in our smaller customers get better, just as you see the growth rate in our fasteners and non fasteners get better. Now does non fastener growth coming out typically maybe outpace I'm sorry, does fastener growth coming out outpace non fastener growth? Probably, just like it outpaces it going in.
But if you look at Page 5 and you kind of look at the non fastener and fastener lines, there's not a difference in how they move, right? They really move in the same direction. The gap may narrow or widen depending on the cycle. But as long as we're successful with Onsites, as long as we're successful with vending, I don't think you're going to get a period where mix works in your favor, frankly. And that's just how the math works out.
This is always where I always feel a need as well, however, to point out that if mix is going to continue to work against us, if we're successful or when we're successful with our on sites and vending and our growth drivers, you have to remember that we get good leverage over SG and A. One piece of perspective I might offer you is, if you think about our gross margin Q4 2016 to Q4 2019, I think it's down something like almost 300 basis points. If you think about our 4th quarter operating margin over that same period of time, it's down 60. There is inherent leverage in Onsites. There is inherent leverage in the pathway to profit that we asked about earlier.
And so I wouldn't expect that mix is going to start pushing our margins up because I expect we'll be really successful with our growth drivers. I wouldn't conclude from that that our operating margins are at risk because there's inherent leverage ability in the growth drivers at the operating expense line as well. Say, we're coming up on the oh, say, Josh, I'm going to we're coming up on the hour. So I'm going to bring the call to a close. I'll just close with one thought and that is and I often bring a personal touch into this and I apologize for those people that are rolling their eyes right now.
But last weekend, I had the opportunity to go to a funeral, a funeral for a very dear individual. Her name was Lillian Peterson. She was known as Till to family and friends. For me, she was both in a way, while she was a neighbor, she was also my godmother, and I consider that family. She was 88 years old.
She's married to her husband for 67 years. She and her husband had 6 children. As you can appreciate, a multiple of that in grandchildren and great grandchildren. She was my Sunday school teacher. She was a giver of great advice and affection.
She was the writer of many letters. Over the years, I received many letters from her and as did many others. Said simply, she was a second mom. I alluded to earlier the top bunch of things that I didn't know, It dawned on me that she was an incredible influencer on 3 of them. Number 8 on that list is trust people.
By exposure, Bob Kirtland put a little adder on that and that is incubate and cherish ideas. Number 9 on that list is cherish, embrace the special words. Say please and thank you. Always say you're welcome. Be willing to say you're sorry and show respect for all.
And finally, enjoy something outdoors in all four seasons. She loved to ride bikes, she loved to go for walks and she loved to cross country ski in the winter. And if you look foolish enough to live in Wisconsin or Minnesota, you better do all you better enjoy all 4 seasons. I'll close on that note. Thanks, everybody.
Thanks, everyone.
Thank you. This concludes today's conference. You may disconnect your lines at this time. Thank you for your participation.