Editor's note: I wrote this about Grubhub's original pay model update, introduced in March, 2019. Grubhub tweaked their pay model again in late June, 2019.
Grubhub created quite a stir in the driver community by notifying drivers of a new pay model. They seem to be taking a page from Doordash as they are not providing details of how they will figure pay.
In the past, Grubhub paid a base fee plus a mileage fee. Under the new pay model “You will be paid a per order amount (“Delivery Pay”) + 100% of your tips for every order you deliver.”
Sounds like Doordash, right?
On May 11, the new pay model went into effect in Denver and some other markets. That allowed me to get a first hand look at what it would look like.
Here are some screenshots of offers.
Ignore the chicken scratches where I write in the straight line distance measurements (thank you Google Maps) – I copied in the pay summary for each delivery at the bottom and wrote in what the old pay model would have been.
This was a double order – on this one, pay was higher overall than it would have been under the old model, by 85 cents
This too paid slightly better. So far I've noticed $4.53 pop up a few times.
I work downtown a lot so typically have shorter distance deliveries. This one was a bit longer, so I was curious if I'd see much difference. This too came in slightly higher.
Speaking of distance, this was the longest drive of the day. 3.4 miles as the crow flies, of course a bit longer with real distance. It did pay a little bit less, making me think that maybe the distance component is a little lower paying.
This one came in a bit lower, though not significantly so. Short distance didn't seem to make a lot of difference, at least not here. There does seem to be a slightly lower starting point than the $3.50 – that shows up in the next one.
I had a couple that were $3.40 delivery pay – both on short distance deliveries. Interesting thing to note on this one is the $3.40 is less than the $3.50 initial fee under the old model.
I did not capture the screen shot for the offer on this one. This was by far the most significant difference in pay. Okay, I'm always happy to see a nice tip like this, right? The customer was 0.9 miles away as the crow flies which would have been 45 cents for distance. That would add to a $3.95 delivery fee. The new model paid $3.41 MORE.
Why the difference? I wonder if they anticipated a longer wait time. I waited about 20 minutes at the restaurant for this, being a larger order.
One other delivery makes me think that anticipated wait time is a factor.
I had to double and triple check this one when I saw the pricing. One paid more and the other paid less. This makes me wonder if time was a factor, because the one that paid significantly more wasn't due to be picked up until ten minutes later than the lower pay offer.
This was the lowest tip offer I had. It was interesting that this was another that had the $4.53 delivery pay – same as the delivery pay on the $3 tip portion of the double order from Mastrpiece.
So far, the new model is paying out slightly higher than the old model.
I did take some deliveries I wouldn't normally accept, just trying to see if I can pick up any patterns. So far I have a small sample, only a dozen or so deliveries, and it's hard to say for sure. But here are a few takes:
The overal difference was minimal. When taking out the one anomaly (the Rush Bowls trip) the total delivery fees were only about 3% above what I calculated the original model would pay.
I did wonder if there would be a Doordash like component where delivery pay would go up when tips were down. Grubhub seems to be testing that concept out in some markets. The three lowest tipped deliveries had tips of $2, $3, and $3. Two of those had a $4.53 delivery pay but the other three dollar tip order had $3.40. So why the difference there? Does time of day or busier market conditions fit into the difference?
Wait time seems to be the biggest variable. If they do pay extra based on wait time, I'm curious how they figure that other than estimated wait time. Every delivery paid exactly what I was offered. If pay were based on actual wait time, there would have been fluctuations.
It's hard to say how distance figures into the equation. If they are factoring in other things such as wait times, I would guess that the distance part of the calculation will go down a little.
I'm not sure that it makes a lot of difference how it looks now.
The bottom line is, Grubhub is removing their transparency in how they are calculating fares. Pay seems slightly better right now on average. However, one of reason I believe they are doing this is that they can slash fees later without drivers really knowing.
In their email they sent out about the new model, they made sure to include the following statement: “Each time you toggle on in the Grubhub for Drivers app, you are accepting the most recent payment terms that have been offered to you.”
This gives them room to evolve their pay model without creating an uproar over each change. I think they are implementing this so they can move to a model closer to Doordash's. But to make changes and adjustments, they have to get to a less transparent model so that they can make the changes more on the fly as market conditions change.
It's too early to tell, honestly. With a lot more data, maybe some of the trends can start showing up giving a better idea. The bottom line is some trips will pay a little more, some a little less.
In time I think that the normal pay model will inch down a little, freeing up money that Grubhub can use to supplement tips or to offer better incentives during peak times.
The timing of the new model going into effect in some markets is interesting as it comes when this article shows up in Fortune Magazine that Doordash has now moved ahead of Grubhub in earnings market share.
The bottom line is, Grubhub has to do something as they are tumbling fast. Uber Eats is not far behind. Whether this or any other model will be enough is not clear.
Time will tell.