What is the benefit of 2 drive thru lanes at a fast food restaurant?
The following photo shows the drive thru lane at a fast food restaurant. The lane splits in two to allow 2 drivers to place their order at the same time.
The following photo shows an overhead view of the restaurant:
As you can see, the two lanes converge back into one lane before the payment window. Therefore the payment and receipt of food is still done serially, which would seem to negate any benefit from orders being placed in parallel.
In addition, there are downsides to having more than one drive thru lane. A driver in one lane might not pay any attention to the other lane, which would increase the risk of an accident. Also, due to orders being placed simultaneously, the person at the payment window repeats the order to each driver to confirm that the orders/drivers are in sync. On more than one occasion I have had the wrong order repeated to me.
So what is the benefit of having two drive thru lanes?
Your question, "_What is the benefit of 2 drive thru lanes at a fast food restaurant?_" logically lead to the question, "What is the benefit of more than one queue inside a fast food restaurant? The benefits are the same.
I'm sorry that you've had such a bad experience: "On more than one occasion I have had the wrong order repeated to me." I've never had that happen. (And I've visited drive-thru lanes quite a lot.) Actually, I've seen local staff handle this well. For instance, if two people finish ordering at the same time, one staff member would delay a bit longer before responding with the order-recap/total, thereby controlling which car advanced first, so that the natural but avoidable race would be far less likely to occur.
Placing the order is the part that takes the most time, as there are variables. The payment and collection phase aren't variable - you're told a price, you hand over the currency and you're given what you ordered.
@Makyen, a pretty obvious explanation is that ordering takes longer than paying. The longest process is parallelized.
@PaulDraper, Yes, that is what I was implicitly attempting to call attention to with my comment. As Rob answered, shortly after I made that comment (so I did not answer), it's all about the time-efficiency of the entire *process* of: customer arrives➞(stuff happens)➞customer departs satisfied after giving money and receiving product. Ultimately it is about maximizing profit. This is done by carefully studying the bottlenecks in the process and applying additional resources, including parallelizing, where it can speed things up. (cont')
(cont')The actual order taking process is one of the major things that can be parallelized and doing so provides significant benefit. This is the same reason why you have multiple cash registers inside. The significant differences between those two is that A) to have two queues for cars takes up *dramatically* more room than an additional walk-in queue. Space taken by every portion of the process is a very significant cost/benefit trade-off when designing the site. B) The drive-up process can be split between order taking and payment efficiently and without significantly annoying the customer.
Maybe not obvious, but you'd need two different people inside taking the orders. Just one person to take the cash from both lanes, and one person to hand out food from both lanes.
I am in the fast food business having owned five units for over 30 years but don't have a unit with two lanes.
The reason for two lanes is a matter of timing from McDonald's research into this that I read years ago. (I am not with McDonalds.) It all comes down to timing and shaving 10 or 20 seconds off the time you wait in line is worth having. The slowest part in this queue is ordering. Not everyone is slow but the average is. And if one line gets bogged down by one customer, half of the other customers are in another line and continue to flow, hopefully.
In addition, it makes the line look shorter. So it's partially psychological but combined with increase feedthrough it makes the customer experience better. It's better to be third in line of two lines than sixth in line of one line.
Of course, sometimes it's a matter of parking space and two lines prevents cars from spilling out into the entrance from the street.
This is how I look at things. First customer comes in and gets waited on right away. Everything is great. Second customer comes in but he's happy cause he's next! But the third customer in thinks, "Aw, man, how long is this going to take?"
EDIT: The queuing theory, and all similar ideas, are very much applied to these things. Fast food operations are small factories that produce packages of meals no matter what anyone wants you to believe. Just like a plant, timing of operations is studied from before the restaurant opens to when a customer walks in the door to the time it closes.
This flow is just as important as anything else. I know how long it should take for a single customer at the head of the line to get through to checkout. I know how long it should take for the customer at the end of the line, too. If it's taking longer than those max times, my people make adjustments like taking problem customers out of that flow.
I've always said that, in my restaurants, if something goes seriously wrong at any point in our morning setup, it takes us a full 24 hours to recover. The customer won't always notice but we do.
+1 Because this is the only answer (till now) mentioning that maybe the true reason is not UX, or the number of customers served, but just optimal use of the available space.
I voted for Robb since experience wins over theory. Never the less I have some comments: 1. there are TWO UI aspects: 1. the people in the queue and 2. the people serving the queue. This would increase the customers satisfaction, possibly increase the workers satisfaction (the happier the customer is the nicer they are to deal with), but may make the work a bit harder for the worker.
@iheggie I'd say that the experience confirms the theory. Queueing theory is a whole area of mathematics and it tells you all kinds of useful things. For example, it's better to have a single queue shared by all the servers than it is to have a separate queue for each server.
@david, the funny thing is I doubt the person who originally did this had a correct theoretical model. Most people would blindly apply the "more is better" theory than actually work up a proper model of the system. What I suspect is most likely is some ideas guy convinced the staid accountant style guy to give it a limited go. The accountant measured the profit and customers served per hour stats and decided to roll it out across more stores. Meanwhile the ideas guy probably had moved on to the next idea. I doubt anyone had a correct detailed understanding of the mechanics of why it worked.
@david - Queuing theory doesn't state that a single queue shared by all servers is always optimum! Instead it gives you the intellectual model and maths to determine which is best. Where the time to take the order exceeds the time to take the money or to serve the food then overall efficiency will increase with multiple queues at that point. It also allows rudimentary prioritisation by time taken to order +incentive to order quickly to get ahead. Your web request doesn't wait for someone else's DVD to download. Likewise you don't form a long single queue to get into the football field.
@iheggie "Queuing theory doesn't state that a single queue shared by all servers is always optimum!" Sorry -- I didn't mean to suggest it does, but it can be hard to express complex ideas in short comments.
@david - yes, it forced me to do many revisions, and still came back and told me off for taking too long to edit lol. I suspect you also like the theory behind these things like I do - just I have learnt that isn't common. A good book for a laugh and ideas is "The Inmates Are Running the Asylum: Why High Tech Products Drive Us Crazy and How to Restore the Sanity" (I am definitely a "homologicus" - I got invited to the cockpit on a Aussie - USA flight (pre 9/11))
Long lines spilling out into the road is definitely one of the big factors driving this. In Phoenix a downtown McDonald's was torn down and rebuilt to accommodate the traffic issues. There are news articles about it online.
Great answer. I was actually wondering just the other day how they guarantee the meals are "delivered" in the same order that they were ordered (my local Maccers has one lane/queue, but two order points in that lane). I couldn't decide whether there was a potential problem there (that had been solved with clever algorithms or maybe just reg plate scanning) or whether I was imagining it and the problem solved itself lol
It's better for some situations like a quicker user gets through the line faster. Example: a car full of kids takes 3 minutes to order. A single person who knows what they want takes 30 seconds to order. By having two lanes, that single person has the ability to bypass the car full of kids. They get their food faster with minimal effect on the car full of kids.
This is like the benefit of having multiple processors in your computer. The greatest % increase is when you go from 1 to 2
If taking an order takes *2t* and paying and handing out an order each takes *t* then the time saved for each customer is *~t*.
@PhillipW Because that works so well for "Carpool Lane" and "10 Items or Less?" ;)
Note: just like David Meister mentions in his great answer, this is based on Theory of Constraints
Let's say we have 2 rows of 4 cars and it takes 3 minutes to each car to make an order, plus 1 minute to pay. Since they're taking orders in parallel, it will take 12 minutes to each row to go through, plus 8 minutes for all cars to pay.
That totals 20 minutes.
Now, think about the same scenario, only with just one drive-thru:
it will take 32 minutes.
Same case, with just one drive thru. The first 3-4 cars will just wait patiently. The other cars will grow impatient, and maybe even try to get out and go to another restaurant, causing trouble to the cars behind or getting stuck in the middle. Needless to say the amount of frustration will skyrocket. If you want more drama, try to picture this with 3 kids yelling and you'll get the perfect nightmare (and obviously, this user will never get back, granted).
Same situation. Now you come with your car and see a long line of cars that gets to the road. Being smart, you say to yourself: is the time I will spend here worth it? Quite probably not, so you move out to another place.
Consider the cases mentioned in my first point. Now let's say you need 10 cars to pay your employees, and you have only one employee per station (taking orders and cashier). For the purpose of this example, you work at full capacity 8 hours.
So, with 2 drive lanes you have 3 employees, therefore you need 30 cars to pay the costs. As I mentioned before, it will take a total of 4 minutes to each car. But since they go in parallel, it will take 75 minutes to pay the costs, with an average of 2.5 minutes per car. Then you'll have 405 minutes of remaining time, or 162 cars (
405 minutes / 2.5 minutes = 162 cars)
One lane case: you need 2 employees, hence 20 cars. It will take them 100 minutes (20 x 4 minutes) and leave 380 minutes of remaining time. Now
380/4=92so you saved a bit in employees wages, but your income will fall drastically, almost to half.
Let's say you've a busy restaurant. Then, oh catastrophe, something happens. Let's say the ordering computer goes kaput. Or your employee fails to show up. So you lose sales until you fix the issue. By having another lane, you will prevent the chances of losses by (literally) 50%
It makes all the sense, it's all benefit with barely any downfall
A point on "By having another lane, you will prevent the chances of losses by (literally) 50%": actually you _double_ your chance of any loss, because there is twice as much stuff to go wrong; but we reduce the cost of any single failure significantly, and better than half the probability of the system falling over completely (this all assuming the event of either lane closing is independent from the other). It is interesting to read about the number of engines on aircraft (commercial and military), and how these considerations effect such choices.
You're assuming there are multiple order-takers, so that orders can be taken simultaneously, but I have never seen this. In fact, I have never seen more than one booth in operation and that booth both takes orders over the speaker and takes payments at the window, further reducing the effect of parallelisation
Your additions are wrong – the payment can happen in parallel to the ordering of the next customer, so it is 15 minutes versus 25 minutes.
@PaŭloEbermann I'm providing a simple and easy to see/explain model, not a real-life model. Of course that could (and will) happen, but for the answer's sake, I simplified this to improve visualization of data. Yes, I could also use regression and statistical models, but... would they help anyone? I don't think so
Reduce time and/or reduce queue before the process starts.
The process goes:
- client orders
- restaurant prepares the order
- restaurant serves the order
If the restaurant places two lanes then two clients can start the process at the same time or at least make two queues. I guess seeing a queue of cars might make potential clients go away, so if there are two lanes for the first step you reduce the amount of clients that haven't started the process to half (and move the queue to after the order).
It is not the same to wait because one already ordered, than wait to order, because in this last case the client might simply leave.
About reducing the time, as others have stated, the first step is the one that doesn't depend on the restaurant and might take more or less time depending on the client.
In the case where only one person is taking orders, the time for each client would be the same as having one lane (no time reduced). However, there would be two short queues instead of one long (at the start).
Since the two lanes are BEFORE the payment window, it doesn't really benefit the "restaurant" at all to have the 2 lanes. I could place an order, land at a cue, then bail before I've paid and AFTER I've executed the order. The only way a place like this would benefit from the 2 lanes would be if the user paid upon ordering because only then are they obligated to wait.
@coblr well, you are right in that, the client can simply leave at any point before payment. But if the client already ordered, I believe it is more probable he will stay than if he didn't order yet.
You missed the, "client pays," step, which in theory happens in parallel to preparation. @coblr I think you're assuming that the restaurant isn't willing to risk the loss of preparing food that the client never pays for. I doubt the risk is all that high; most customers pay and get their food. Fast food restaurants already budget in some lost food for things like mistakes in preparation anyway.
@coblr: Actually, there's generally quite a substantial benefit to the restaurant, though it's not in forcing people to pay. The benefit they get is more people getting in line due to the *appearance* of the line being shorter. Most people going through drive-through could 1) skip eating right now entirely, or 2) move on to a different restaurant very easily. The longer the lines look, the less likely it is that other people will get in line at all.
@coblr: How do you bail from a drive-through line? Unless there's nobody behind you, you're boxed in on all four sides, so you have little choice but to keep moving forward. Granted, you could just refuse to pay, but why would you after waiting that whole time?
@FighterJet, you bail by just driving away. I haven't been to many drive thrus that have a wall or curb stopping you from leaving, but I guess it does happen so, fair enough question. Even in the pic in the OP, you just let the person in front move ahead a little (finally!!) then leave. No, it doesn't happen often, but I've been in a situation where I was on the fence about getting fast food anyway, and just decided that after waiting for 15 mins, it was a sign. I do see the point of a shorter line. In&Out always has LOOONG lines, so whenever I see one with a short line, I'll go just for that.
@FighterJet I've seen a lot of lines that are blocked off once you pass the speaker. Most have some kind of opening that lets you leave after passing the speaker. This makes sense; if they don't have what you wanted and you never placed an order, it's better for you to get out of everyone's way. You *could* abuse this by getting out of line after ordering.
I think another way to look at it is by an analogy:
In the men's restroom count the following:
Just because you see 8 urinal stalls does not mean one also has to provide 8 doors, correct?
The underlying point is that you need to provide more parallelism at the slower steps. Assuming, and I think it is a good assumption, that ordering takes more time than paying it makes perfect sense to have more ordering-windows than paying-windows.
You see this same feature in tons of, at first glance unrelated, user experiences:
- Lanes fan out at toll payment points on the highway. You may have a 2 lane tollway but 8 parallel toll booth lanes.
- You may have 10 self check in kiosks but a single baggage drop point
- You may have a dozen manual check in counters and weighing scales shared between two agents but only a single conveyor finally leads to the baggage sorting room.
the payment and receipt of food is still done serially, which would seem to negate any benefit from orders being placed in parallel.
There is no reason to assume this is true.
The answer, from a UX perspective has to be:
The benefit for the user is getting their food faster because the restaurant can process more meals per hour, leading to shorter queues.
There are two things to consider here to understand why the basic premise of this question is probably wrong.
Simply moving from a concurrent step to a serial step does NOT negate the benefits of the earlier paralellism.
Let's look at an analogous example from web development. We want to fetch data from two different servers (parallel processes) and then use that data to render text into a web page from top to bottom (serial process).
If it takes 3 seconds to hit each server and 1ms to render the response from a server we can obviously either take 6 seconds or 3 seconds to finish the task depending on our choice of parallel or serial for the first step ONLY.
The second step's parallelism changes the overall situation by 1ms, making it completely irrelevant.
You see the same situation at the fast food restaurant.
Time to take an order, create and package the food ~= 5 minutes.
Time to take money and hand food through a window ~= 30 seconds.
Provided that the internal team has the facilities to prepare at least 2 orders at once, the result is:
Step 1 in parallel = 5.5 minutes total
Step 1 in serial = 10.5 minutes in total
The parallelism of step 2 doesn't negate anything here.
A system with a series of sequential steps is constrained ONLY by the slowest step.
This is called theory of constraints. It has a wikipedia article too.
The summary is that in a system with a chain (or even multiple converging or diverging chains) of processes that need to be completed, the system moves only as quickly as the slowest step.
Attempts to optimise any step other than the slowest step have no positive impact and can even have a negative impact on overall productivity.
Imagine a traffic jam at peak hour where many lanes (A) converge into fewer lanes (B) and then diverge back to many (C). Something like this:
It should be clear that adding additional lanes at:
A - will make traffic worse by increasing congestion ahead of the bottleneck (incidentally, this is why we use words like "bottleneck" to talk about a limiting factor in a process)
B - will increase throughput of the system
C - will have no positive or negative impact on the traffic
Your example of a fast food restaurant (or literally any other system in the world) has just one severely limiting factor at any point in time. Nothing else is worth optimising.
Think of the following stages of throughput in real space/time terms of burgers per minute per square meter, after all, this is how they pay rent (I'm just guessing rough numbers):
- Collecting orders (~ 2 mins) from a drive thru lane (~20 m2) = 0.025 b/min/m2
- Preparing orders (~ 3 mins) from a burger grill (~1 m2) = 0.3 b/min/m2
- Delivering orders (~ 0.5 min) from a kiosk (~2 m2) = 1 b/min/m2
Now also consider that the nothing can be done before you place the order yet preparing and delivering orders can be done in parallel with taking new orders.
It should be clear why:
- A very large amount of physical space shown in the OPs aerial photo is dedicated to taking orders, relative to other tasks
- Taking orders is handled in parallel (it is the step that needs to be optimised)
- Every other step (before and after) can be safely handled in serial (trading throughput/latency for less physical space requirements) as their throughput does not effect the throughput of the system overall
It also hints as to why perhaps having 2 lanes for orders makes sense but not 3, or 10 or 50 - this would be out of proportion for the size of the kitchen and other available facilities required to process burgers.
One potential avenue for followup, would be whether this restaurant is really optimised for consistent user experience (latency) or just total revenue per square meter (throughput). As far as sending items concurrently through parallel/sequential processes goes, they are not necessarily the same thing. It's quite possible to optimise overall throughput in such a way that individual items may suffer additional latency (like when the OPs order is forgotten at the kiosk).
This is a pure UX question. I did similar research for bank teller lanes in college. I will answer your question with one word - perception.
When you drive up and see 5 cars ahead of you, you might be like "F that I will try somewhere else." They know once you order you will hardly ever leave. So they just want you ordering and don't want to lose your business. Your question has two lanes in the example... Sonic took this to the max as some of their locations have 40+ ordering points. It is all about perception as you are still in line. They just want you to think that the line is as short as possible until you order.
I am not saying the other answers are wrong. They make valid points but they are all secondary to this point. As you can see below, you can drive up and order right away at Sonic and you will be the first person in line ;).
I already said what you said, but I fail to see why all other reasons are secondary to yours. Besides, perception isn't created in a vacuum, but it's the resultant of all the stimuli the user receives... namely, everything you consider secondary
@Devin - No you didn't - at least not in a coherent way. You are talking about queue times and how long it gets someone through. I am talking about the perception of time a customer feels like they will be waited on. In the two lane example we did numerous studies and unless there is one staff member per lane the time savings is often negligible as there will be an increased rate of mishaps. Now Sonic has taken this further because they make it seem like you are waited on right away AND the consumer has no ability to calculate length ... cont.
cont... because the consumer has no idea who has and hasn't ordered. In the two lane example it is pretty easy to calculate how many cars there are ahead of you... and 3-5 mins per car... so you can get to it. But places like McDonald's basically think that their average customers will not go through these calculations and if they do they would have already spent time in line and will not give up on a sunken cost.
Pure speculation but it makes sense.
Payment and receiving food are very quick. What takes the most time in any food place, including fast food, is food cooking/preparation. If they can get multiple peoples orders they can get all the food preparation out of the way and simply trade food for money at the window.
Actually as @Josh Carr mentioned in his answer, The most lengthy process could be the ordering of the food. My instinct has always been that with fast food most of the food is already prepared and is cooked very quickly
@yitzih it's certainly possible ordering could take longer, it will depend entirely on the order. I could just drive up and say I want a #10 and be done ordering in seconds or I could 5 kids in the backseat and take all day. At the same time they could shove some chicken in a bucket in the prep process or prepare two dozen tacos. The main point is that the window isn't going to be the bottleneck so it makes sense to have multiple lanes.
Having worked in a McDonalds I'd say the main advantage of 2 windows is so that the orders are placed and they can start being made. It just means more orders can be placed faster. "the person at the payment window repeats the order to each driver to confirm that the orders/drivers are in sync", I know in England, the staff taking the orders get a video feed of the customer, this automatically takes a photo when the first option is placed on the order. The person at the payment window then gets a face next to the order to confirm who placed it. This doesn't always work because the pictures aren't always perfect but the majority of the time it works well.