Suggestions for The Bidding Feature

[chefperryp]

I've been thinking about how the bidding system works (and doesn't), and I'd like to get your .02 on a couple of ideas...

Personally, I think that the current system is lenient to the point of it being ridiculous and counterproductive.  

I don't have a problem with bidding, per se, but the way we're doing it is illogical and unsustainable, both for artists and clients. People are bidding 20, 40, 60 (or many more) connects right up front for jobs that barely pay enough to cover the cost of those connects.

Frankly, I have a hard time believing that this is being done by the "best and the brightest" freelancers that Upwork has to offer (no offense, lol).

The idea that simply throwing more connects at a proposal in any way communicates the quality or professionalism that the customer can expect from the bidder is ridiculous.

Clients aren't getting a "better" pool of prospects, there simply getting the ones who are desperate, or just don't care.

In poker, we call it "buying the pot."

The only thing that actually benefits both the clients AND  the freelancers are the work history, experience, and reviews of the bidder under consideration. That’s the only way to even attempt to guarantee that they’re getting the best person for the job.

(And before you say that the client “still sees all of the applications”…BS. If being in the “Top 3” didn’t dramatically increase the odds for those bidders, then the whole system would be pointless, wouldn’t it? Statistically, the client is going to focus on those spots, and thier implied "higher value".)

That being the case, why not run it more like a "limit" poker game?

There’s an "ante" (the mandatory amount of connects required to even apply for the job) that everyone has to pay, followed by rounds of "upping the ante", with a maximum amount that can be raised each round, until bidding stops, or someone is hired.

So, for example, if the highest bid is 12 connects, the maximum allowed raise would be 3 connects over that (25% of the "pot"), or 15 connects.

This could:

1. Encourage more serious freelancers to bid on jobs instead of immediately being put off by a huge upfront bid they have to beat.

2. Upwork still gets its new "rake" (you guys aren't fooling anybody, lol!)

3. Bidders who are truly interested in the job are more likely to come back and "raise" their bid, while impulse bidders get a chance to cool down and think about it. This is likely to decrease the number of less qualified bidders.

Maybe only freelancers with a minimum success rating can bid for the top spot? 

Individual pots might not get as high as they do now, at least not as quickly, but I think a higher percentage of freelancers will actually use the system if it seems more logical, and less risky.

Thoughts?

[feed_my_eyes]

If you think that the bids are crazy now, they would go through the roof if everyone was allowed to keep coming back to raise the stakes even higher. And the top three positions would be changing continually, which would water down any benefits and probably annoy the hell out of clients. No, just no.

[25005175]

I've simply been in favor of returning to the blind bidding that we had earlier this year. I am intrigued by your idea. But I think that there is a more important element to change, which is on the client side.

After some recent discussions on how much the boosted slots get in the way / clutter up the interface, I think that the solution would be to create a popup tab for the boosted slots, so that the client can simply exit out of the popup to return to the interface. The boosts remain, but as a tab. It's like showing up to work and seeing three adverts next to the front door. They get your attention for the moment, but then you can walk past them. If you are curious later, you can always step back outside.

[williamtcooper]

Hey Perry, Upwork's bidding process works very well. Those that bid high connects are only going to win if they are appropriate. I have won jobs bidding and not bidding. If it's a great match bid for the top 3, otherwise pass and go for the long list. I like the option to bid for work that I know I have a high probablity of winning if seen. Thanks!

[w62n]

the whole "bid" thing is completely unethical. it reduces upwork to basically a "pay to play" situation. 

[martina_plaschka]

It is neither ethical nor unethical. It is how upwork choses to run their business. It's a take it or leave it situation. 

[wlyonsatl]

Perry P.,

No bidding system will satisfy everyone, but how do you know, "Statistically, the client is going to focus on those spots, and their implied 'higher value'" because they are in the special TOP 3 boosted bids list? If Upwork's existing algorithms for ranking based on a freelancer's skillset were working to the benefit of clients, then the TOP3 boosted bids should have little additional little value to clients whose projects have any complexity or need for above average experience and skills from the winning freelancer.

The only party that knows whether boosting helps boosting freelancers win projects is Upwork. And Upwork ain't sayin', although the Boost Your Proposal page does claim, without providing any details, "When you’re able to be first in line for projects you’re excited about, you’re more likely to land the work you want." So, Upwork, how much more likely? Does boosting double the chance of posting a winning bid? Or is the chance of success 10% higher with a boosted bid? Does the freelancer’s JSS matter? Does project value matter? (NOTE: Upwork does give some numbers here: https://support.upwork.com/hc/en-us/articles/11983621573395. But any recent graduate in statistics could apply common analytical tools to put these results in perspective. Raw numbers are only minimally, if at all, meaningful.)

But we do know 1.3 million projects were completed through the first 11 1/2 months of 2022. If the average boosted number of bids used to win those projects was 10, on an annual basis that would result in $1.95 million (1.3 million x 10 x $0.15)  in additional payments made by freelancers to make those boosted bids. That's a big number. And is actually even bigger because it doesn't count the boosted bids by the second and third place freelancers who didn't win the project or by other freelancers who "received an eligible interaction from the client while boosted." even if the boosting freelancer does not end up in the project's final TOP 3 boosted bids list. Using my assumptions, Upwork's annual revenue and pretax profit would be increased at least $2.0 million through the boosted bidding process. Maybe more. Maybe much more.

Every little bit counts for a company struggling to finally report an operating profit to its shareholders. The numbers for boosted bid counts could all be larger or smaller than I have assumed here, but until Upwork provides some detailed, statistically-significant information about how boosting helps freelancers and/or clients, I assume the main driver for Upwork use this boosted bidding system is its own additional revenue (which has minimal ongoing cost to Upwork and goes straight to Upwork's bottom line profit). 

On simple projects where Upwork has a huge number of freelancers whose skills are sufficient to complete the work, I expect boosting makes it easier for the client to save time identifying a freelancer from among the TOP 3 boosted bidders. Ironically, this means that freelancers who do the simplest, lowest paid work are going to pay the most for bidding relative to the value of the projects they bid on.

But I see no reason to think that smart clients with complex projects (the kind of clients I want) are going to base their hiring decision on who is willing to pay Upwork the most in order to be chosen, ignoring every proposing freelancer’s skills, experience, JSS score, etc., etc.. 

I would happily see factual information from Upwork that shows my assumptions here are wrong (or right).

[25005175]

NOTE: Upwork does give some numbers here: https://support.upwork.com/hc/en-us/articles/11983621573395. But any recent graduate in statistics could apply common analytical tools to put these results in perspective.

---

I did that analysis here. TL;DR - the effect is paltry. Over 1000 bids, I would expect to get 7.4 more conversions (8.6 if I landed the top spot each time).

---

[wlyonsatl]

We're on the same page, Jonathan.

But you are trying to analyze the un-analyzable.

Without access to a good deal more detail on each of 1,000s of projects posted during a given period (the past three months?), even the best statistician would have no idea whether the "boosted" proposals had more (or less) success because they were boosted or for other reasons (such as winning and losing freelancers' JSS, total "perceived" value of project for winning freelancer, total number of proposals submitted on each project, etc., etc.).

[25005175]

It was very hypothetical, yes. There were strict assumptions that I made, outlined in the post. You are correct that there are other variables involved, that could increase or decrease the effect - but I think the effect would very rarely reach more than +10 conversions out of 1000 bids.

ETA - deleted a word for clarity

[kbadeau]

I've only boosted twice, both times to get my name at the top of a list of 20-50 people who had already applied for the job. I won one and lost one. My winning bid was 14 connects. It was super worth it to get to the top of a stack of proposals that were probably low ball and frustrating for the client. But we can talk about it all day every day... what we really need to know is what do the clients think about it?

[wlyonsatl]

Kelly B.,

I think understanding the benefits of boosting need to be explained/clearly measured for both freelancers (who may be paying Upwork at least an additional $2 million dollars per year for boosting) and clients (who see three freelancers highlighted who are willing to pay Upwork compared to how Upwork's long-standing proposal-ranking algorithm helped clients find "the best" freelancers).

For example, if boosting substantially increases the chance of a proposal winning a project regardless of the boosting freelancer's JSS, that would be worth knowing for both freelancers and clients.

[deec6e54]

Good idea, but you're directing your question to the "party"

that sells, and generates cashflow, from selling the "Connects."

The CEOs heard a proposal on the topic, believed it to be a

good plan to help reach sales goals, and now we have "Connects"

to help us win the bid! Kinda resembles penny auctions like Quibids

or Dealdash, right? Now I'm going to go "design" a proposal that, at

my houry rate, cost more to create than the job pays. ;j)

[fbd54129]

What I think is bidding is taking us no where. The thing that matters is work history, reviews and the profile on upwork. Bidding should be removed and rather upwork can come up with a algorithm like google where they rank the ones on some pre defined fair basis. This will not only give the new talent chance to grow but also the old champions in the race will have their piece of cake. The game should be fair for all . Just bidding for 20-30 and upto 60 connects these days is not something good in my opinion.

[yofazza]

They don't profit for years and they're looking  for a way to change it.

'Fair for all' is not at the top of their agenda when they have 12 million freelancers (where most of them won't generate fees) roam freely only to put burden on their expensive servers.

upto 60 connects

You kidding? Some people spend hundreds of dollars for a single bid. 😁