75% of Proposals Aren't Viewed - What are your stats? - I want 2 drive a product/platform change

[gbuechler]

My background on the platform: 6yrs, 3mo on upwork

$20K+ total earnings (lifetime on platform)

2023 (3.5 mo, 136 Proposals) Stats:

VIEW RATE

All proposals (views/sent): 26%

Organic proposals(organic views/organic sent): 24.7%

Boosted proposals (boosted views/boosted sent): 33%

INTERVIEW RATE

All Views (all interviews/all viewed): 42%

Organic Interviews (organic interviews/organic sent): 40.7%

Boosted Interviews (boosted interviews/boosted sent): 44.4% 

Three quarters of my proposals are NOT viewed; and boosting proposals only increases the view rate to a paltry 33%. I find this to be a flaw of the upwork process/business model as we are spending our connections (which cost us monthly subscriptions and additional purchases). This is literally throwing money away and is a bad customer experience. 

So, I have an idea for improvement and would like other freelancers' stats to see if other's data support the idea. Please be transparent; What are your viewed proposal stats? Viewed, Organic vs Boosted?   

Thank you - gb

[d9fc6d8d]

65%-80% of my proposals get viewed, depending on the week. I don't boost. I had a somewhat hard time a few months ago, so I really worked on my proposals and now they work just fine.

[gbuechler]

Ze Eduardo - Thanks for the response.

The content does't drive the opening and reading of the bid. Your content (letter, answers, etc., ) are after the fact of where I see the flaw is. 

[2779bde0]

Hi, are you able to provide any details. I have been trying different things and still maybe have a 25% view rate.

It's tricky when you only have a very short period of time to get a client's attention. Thank you.

[2779bde0]

This is for the community. I wanted to share some recent insight on the whole concept of writing job proposals.

As a freelancer, I recently had the need to hire a another freelancer on one project, to proofread some copy I wrote, before turning it over to a copywriting client.

Holy Cow.

It was jaw dropping to see the responses I got, and gave me great perspective as to what clients see when they open up the Pandora's box of proposals.

I had clearly put, at the end of the job post, that I wanted to see the first and last names of a specific, former, US president, as the first two words of the job proposal. Only about half of the respondants did so.

Sorry if I'm babbling. Many went on and on about their accomplishments and credentials, which is great to a point, but tell me what you can do for me today.

That experience taught me to get to the point, fast, and tell the client that I understood what they were asking for, and how I am qualified to do a great job for them. I try to keep the length of my proposals under 150 words.

Anyone else have a similar experience?

Thanks.

[kelly_e]

Martin—Good stuff! SO many questions, since I haven't been on the hiring end on UW for several years.

One: Could you see the first 220 or so characters without "opening" a proposal? (So that you could decide which to open based on a preview.)

Two: How many proposals did you get in total, and does UW give you any sort of email notif as proposals come in? This I care about because if there are fifty in an hour, but I'm working that day and I don't see it until the next day and then I send a proposal, I'm always wondering whether a client assumes those 50 are all there will be and never looks again (because UW doesn't tell them more have come in).

Three: Did boosting make any difference to you in which proposals you chose to open, and were the four boosted proposals (if you opened all four) of generally higher or lower quality than the rest?

Four +++: Did you scroll through all pages of proposals? Did you simply choose to open the first few, or did you choose randomly from all the pages of proposals you got? Did you wait until most seemed to have gotten there to try looking through, or did you check in a few times? Did it make any difference to you, in your choice, whether a freelancer jumped on the listing and got there (relatively) early, or were you still interested if a proposal came in a day or two later?

Five: Was there any *real* method to how proposals appeared to be sorted in their presentation to you—I know UW claims to have an absolutely fabulous way of showing you your "best" matches first, but many freelancers suspect this is currently heavily weighted toward new-to-UW and away from long-timer-users—did you feel this was the case, or something else?

Six: How many of the proposals (that you opened, or if you could see previews, that you skimmed over) would you guess were written by bots? Could you tell for sure? Obviously people who couldn't follow the instruction you placed at the end might have been, but also a lot of folks here just seem to manually spam job listers, as well, so that might not be conclusive.

Would LOVE to hear how you experienced the current state of things.

[2779bde0]

Hi Kelly. Thanks for the questions. Rather than restate them all, I'll list the answers.

1. Yes, I could see the first 220 characters. I was looking for something that engaged me, more than the "I am highly qualified to do your job" opener. Show me. How are you highly qualified?

2. About 15 proposals. At the top of the list were the six-figure mega-earners, that UW recommended.

3. No. Boosting didn't make a difference.

4. Yes, I scrolled through all the pages, knocking out all the ones that didn't have the keywords that I had asked for in the post. Then went back and reviewed what was left. I watched bids come in for a couple of days before making a decision.

5. If there was a method, I didn't notice.

6. Can bots write job proposals? A lot of them looked Copy Paste.

Circlig back to the original topic of this thread, most of the proposals people submit aren't being read. What also is unnerving to me is the number of jobs where there are no interviews, and no hires.

Or is the big picture that these same clients who post jobs here, then disappear, are also going to other platforms, and then go "shopping" for the right freelancer from the entire pool of candidates?

I have been here for almost four months, and have sent 130 + proposals, many in a specfice niche. The ones in the beginning were weak, as was my profile. Both much better now. It took almost 30 proposals to get the first job. Where do I need to be to get interviews, or even views? I imagine it's somewhere in those first two sentences.

You would think that experienced freelancers here would have cultivated their long-term clients, provididng the well from which those freelancers drink, so they wouldn't have to send any, or fewer proposals no make the big bucks, and the less experienced ones here would compete for what is left.

What do you think?

[25005175]

---

Martin S wrote:

6. Can bots write job proposals? A lot of them looked Copy Paste.

---

Yes. I'm not a programmer (I know some VBA), but it is not difficult to utilize the Upwork API to auto-submit proposals as jobs are posted. I feel pretty confident that the API isn't even required. Essentially it is just website scraping followed by a submission macro. I doubt one would even need to know any of the code for the proposal submission pages/forms, because I think that part of the script can be created simply by recording button clicks from a normal submission.

[the-right-writer]

2. About 15 proposals. At the top of the list were the six-figure mega-earners, that UW recommended.

This is the problem. Clients see the top three places and believe Upwork is recommending the freelancers, when it is paid advertising by the freelancers. It has nothing to do with skills, ability, or knowing what they are doing. In fact, the ones who boost the most, all the time, probably have zero skills and can't do any job.

[2779bde0]

Agreed. People who think quantity is more important than quality are boosting like there's no tomorrow. I used to go down that road.

There has to be something said for people who take pride in their work. If all that matters is price and quantity, than these low-ballers can bid themselves into oblivion with the clients who will accept that kind of work.

[d13bb621]

Great insight and advice.

[affadda7]

Hi, I'm new to the platform but have some stats for 3 months:
217 offers sent O - 53%, B - 47%
44 views 50/50%
15 interviews conducted 53/47%
5 hires 60/40%.
That's not very good. I have a Plus account too.
Respectfully Anatoly 😒

[gbuechler]

Hi Anatolii - Thank for the data!

80% unopened is exactly my point, and that your boosted proposals are getting the same open rate means that boosting has zero affect on your business other than a higher opportunity cost. 

[affadda7]

100%, I sometimes think it's a waste of time and money for the boosts, I just need to be more clear in my cover letter writing and react as quickly as possible to fresh requests that fall into my professional space

[25005175]

Proposal stats - I'll show you mine if you show me yours 

[gbuechler]

OOOO, that is a GOOD link! Thank you!

[tlsanders]

I'm trying to imagine what you think this has to do with Upwork's business model. Do you imagine that clients and employers in other settings read all of the dozens or hundreds of applications they receive? 

Re the question you asked, 28.5% of my organic proposals and 50% of my boosted proposals have been opened in 2023. But, my sample is much smaller than yours--I've sent 17 proposals this year.

[gbuechler]

Hi Tiffany - The flaw is the unread bids. Those cost freelancers connects (which have monetary value). I would like to see a model where unread bids are refunded. No matter how good a cover letter may be, if it is not read, upwork has not provided the value of that bid. The value created for the project owner is a curated platform where (hopefully) greater than 90% of the communications are spot-on. That benefit is not provided to the freelancer. Much of what we do just goes into the ether. 

[tlsanders]

What I was asking was how you saw unread bids as a flaw in the Upwork system when if you send a job application through Indeed or mail a resume to a company or send a message through a job board (or virtually any other channel), there's a decent chance no one will ever look at it. 

The value Upwork provides is to draw hundreds of thousands of clients to the site so you can connect with them in one place instead of sending out thousands of cold emails or visiting dozens of job boards or whatever. They spend millions of dollars each month bringing in those clients. 

Perhaps you're not old enough to have operated in the offline world, but back in the day when I had to send my resume in by mail, the US postal service never paid me back for my stamps if the client ignored me.

[yofazza]

From how I see it, it's like opening a store in a "quiet" online marketplace (correction on the next post below). But instead of having no buyers, there are buyers who don't see my stuff. Fees doesn't really matter.

I can't compare it to sending proposals directly (or selling directly), which I believe has a high chance of being seen. In this case there's 'no problem' if I don't get a response.

[tlsanders]

But buyers DO see your proposal (in your analogy, your store). They just make a choice as to whether or not your store is one of those they want to walk into and look around, based on what they can see from the outside. 

Is your belief that your proposals have a high chance of being seen in other contexts based on anything other than the fact that Upwork is the only channel telling you whether or not they were seen? 

[yofazza]

From what I understand, buyers don't see it. They don't even see 'my store' from 'the outside'.

I should've written: The marketplace is not quiet. It has buyers, but they don't see me exists.

I know you mentioned that as a client, you'll look into a lot of proposals. But there is also evidence that some clients are 'fed up' and 'leave' after clicking a few proposals. Saw one complained in the Client forum, and also a funny thing where a forum thread is filled with annoying-type of proposals. I'm not a client but from that forum thread I can understand if a client leave after reading just 5 proposals.

My understanding: my proposal previews are not visible to the client unless the algo put them on top, or If they're boosted, or if the client is patient enough to take time to look.

--

I asked whether the 'views' numbers are correct, moments after the feature is released. If - for example - viewing email notification might prevent potential client to click-open the proposal. Mod explained that the view stats could be considered correct.

Even if it's not correct, I still experienced the 'less interviews and hires', compared with before the changes (the search algo change, the boost, and a few others), which I also mentioned that I can understand all of them if I see it from Upwork's POV.

[tlsanders]

You're definitely right that many clients only open a fraction of the proposals they receive. But, I would be surprised to learn that any serious client or competent business person would make that selection based on which proposals landed at the top of the page. It is extremely easy and takes only a minute or two to skim through the "storefront" information, which helps the client decide which proposals are most promising before deciding which to open.

[yofazza]

Just like for the freelancers, I a couple times suggested that better 'education' is needed for the clients:

• So they don't get 'snared' by copy-pasted long proposals that doesn't even address their job description.
• So they can implement a good/safe method in hiring a freelancer where profile and stars alone are not sufficient to guarantee a good result, and also maybe...
• So they can 'be patient a little' and at least quickly skim through the previews of all proposals, so they don't rant in the client forum later after having a useless app for $10,000+.

Many of them do things that we might consider as 'bad practices'.

On the freelancer's side, sending long-copy-pasted proposal that doesn't even address the job description, can get you clients. Complying to the client's 'weird' tests like writing 'blue elephant' in the proposal does not necessarily mean - like most people said - that the client is bad. It's actually a great way to know that the FL have read the job description (and filter-out most proposals I'm certain). Bidding higher than the client's unbelievably low budget doesn't always work, when you look at the client's history and all of their previous work are actually done in that unbelievably low budget (for whatever quality, but I only see 5 stars on both parties).

The definition of 'good or bad practices' are unclear these days.

[tlsanders]

I would agree that the idea of general good and bad practices is silly. I have won jobs on Upwork with three-sentence proposals and gone on to earn many thousands of dollars with the client. I have won clients on Upwork with 4-5 paragraph proposals and gone on to lucrative long-term relationships. The only true "best practice" is to assess the posting and respond based on what you've learned from it (about the client, not just the gig). 

Personally, I think clients with secret words are bad clients. That's because I know that any client who has a bit of business sense can tell in a few seconds whether or not a proposal was written by someone who read and understood the job posting. The only reasons to use a code word to screen would be incompetence (which makes for a very bad working relationship) or extreme laziness. It might work (though it also drives away a lot of good freelancers), but it's a sloppy cheat versus using your own business sense.

[gbuechler]

Let me rephrase - flawed service model

Freelancers who pay both a service fee and a cost to bid are not being serviced well.

As to your other comment about how long I've been in the business... I have over 35 years in my line of work (Corporate, Consulting, and Agency recruiting). I am not so old to not remember mailing resumes, getting the first fax machine in our office to send to clients, and yes, even my first email address (which just happened to be with an ISP with a direct connection to DARPANet - my email address was just two letters back then! Yes, I AM really that old!

The point is that bids cost freelancers in addition to paying fees for hours/projects (double dipping). To have bids refunded would be a HUGE move forward for freelancers since we have ZERO ability to affect that. Simply refunding unread bids is an easy fix. 

As a way to prevent spamming bids - give the project posters the ability to flag bad bids. That is a simple and very effective tool. On the backside, ISP or MAC addresses can be used vet out spamming accounts. 

Finally - as a Corporate Recruiter I read EVERY SINGLE APPLICATION. I also require this of the recruiters who work for me. No excuses what so ever. If a candidate applies for a role, we have the responsibility as an employee (or agent) of the company to read it. 

[tlsanders]

I'm sure that you understand that as a corporate recruiter, your perceived responsibility to the end client is VERY different from a small business owner's own responsibility to manage his business efficiently and effectively. There is certainly no imperative (just for example) that a writer looking for someone to transcribe some interviews read 125 proposals. In fact, that would be an extremely foolish business practice, since the point of hiring someone to do the transcription is to save time.

[celgins]

I don't think the stats/analytics are worth much. There are so many parameters and conditions that factor into where and how often a profile is seen in relevant client searches, and where a proposal shows up when submitting a bid. Upwork's algorithms are machine learning programs that use marketplace data to predict future behavior. The AI-driven algorithm ranks proposals based on Upwork's proprietary data, and all of this is beyond the control of freelancers.

The only thing we can control is the quality of our cover letters and the strength of our profiles. Even if you write the best ever cover letter, it still doesn't mean it will be viewed and lead to an invite.

Oh, and my sample size is even smaller than Tiffany’s. I have sent three (3) proposals in 2023. I have never boosted a proposal and all three (100%) organic proposals were opened. I was interviewed for two and invited, but I declined both of them because I have been super busy with offline work projects. The third job was closed without a hire. So, my stats are mostly irrelevant.

[gbuechler]

Hi Clark - I think you actually are proving my point with " Even if you write the best ever cover letter, it still doesn't mean it will be viewed and lead to an invite." 

Unread bids are literally wasted time and money. I think that unread bids should be refunded, just like if you get outboosted. 

Likewise, I think the model needs tweaking where employers should have some incentive to read all submitted bids. 

Thank you for the post

[tlsanders]

What you don't seem to understand is that it's up to you to market yourself. When the client gets your proposal, they see your picture. They see the title on your profile. They see your JSS. They see information about your earnings and/or hours worked. And, they see the first two lines of your proposal. If all of that information doesn't entice the client to choose your proposal to be one they open, that has little to do with Upwork.

[celgins]

I've never seen a business or government model where the customer/client reimburses a service provider/contractor (freelancer) for the costs they incurred to submit a proposal. I don't like the Boosting feature, but the Connects feature is simply a cost of doing business.

I'm curious about your incentive suggestion. What type of incentive could Upwork provide to clients who open all submitted bids?

[yofazza]

Clark, there are a few people I've seen that has a high view ratio including new freelancers that has like 6 hires - not just views - in their first month. But your 100% views is the best I've seen recently.

I think I also had a high view ratio before. Before the stats existed, before the boost thing. Although I don't always send proposals, I can get a job here in 1-2 weeks when I look, when I start sending proposals.

I usually get the job, once a client responds to my proposal. I don't send too many proposals, and I never use all of my monthly connects. I believe my view ratio was high.

It started to change on mid 2022, around the time of the boosting introduced. I started to need to spend all connects every month.

When the stats feature was released I saw that I often had 0 views.

I get no hires for months, although I look, although all connects are spent. So I do experience the "trending" complains.

I noticed a difference around the end of January when I got two jobs, after I played around with my "properties" (modifying my profiles, skills, etc). But that's it. I'm back into getting no views right now. I wanted to do some more tests but Upwork increased the required connects which basically will limit my proposals to 1-2 per month. I also wanted to wait until I get that new "which proposals are seen" feature before I do more tests.

This two hires are from that end of January:

---

Greg: As you might've noticed, your and Anatolii's vew ratio are not something unusual recently. Mine is 17% which is worse (Although more data will increase accuracy, it's still stats. I understand about business expenses, I just have my own reasons for not buying connects, or doing boosts).

---

I sent a proposal a few days ago and I know from the stats it wasn't seen. This is the job description:

"I need someone to fix outdated WordPress inside new version. The site is hosted on GoDaddy. This is probably a quick fix but I don't know how to do this myself."

It's vague, but I can tell most likely the client's website is broken by an ugly PHP notice from a deprecated plugin function in a custom theme. Updating WP and the plugin. doesn't fix it. Here's the start of my proposal, the "clickbait":

"Is it some outdated function in your custom theme? Can you explain more?" (followed by more explanations and my 'provable' WP credentials)

But it didn't work, the client didn't click it. Most of my proposals are like that. It used to work. What could prevent the client from clicking IF they see the clickbait? Is it my picture? 😶

[celgins]

But your 100% views is the best I've seen recently.

Although I don't always send proposals, I can get a job here in 1-2 weeks when I look, when I start sending proposals.

Radia, I'm glad you said this because it might sound arrogant when I say it. 😂 I'm not very active on the site right now because I'm too busy with offline projects. But when I do start sending proposals again, I am completely confident that, if viewed, I will be invited for the job. Like you, I almost always get the job, once a client responds to my proposal. I'm 100% certain I would have gotten the two jobs I declined because of the interactions I had with the clients.

However, a sample size of three proposals isn't great. I only send proposals to jobs where I'm 100% certain I can provide a high-quality solution for the client. I briefly explain in my cover letters why my solution works, and why it is likely better for them, and the clients usually agree. But if I begin sending 5 proposals a day, my stats will look similar your recent experiences with low to 0 views and fewer hires.

But it didn't work, the client didn't click it. Most of my proposals are like that. It used to work. What could prevent the client from clicking IF they see the clickbait? Is it my picture?

It could depend on how the client is viewing your proposals. But I think the algorithm adjustments over the past year (along with other factors) has affected how and where proposals are ranked.

[kelly_e]

Yep. It's well beyond beastly.

Since the beginning of March, when I decided to start an Excel spreadsheet to slice-and-dice the data a little better, my overall view rate is 25.68%; my boosted-view rate is 40%. (I tried boosting selected proposals for a couple of weeks to see if it had any ROI—IMO, no, it doesn't.) My *legit* contact rate—not counting people who reached out but turned out to be scammers—is 1%.

Though I have gotten a new job in this brief time period, it was from a proposal I sent before March... so nothing new from this 1% contact rate. Nothing. When so few clients view your proposal, what can you expect?

One of the ways I sliced it up was to look at daily rates—for about a month I was seeing significant differences between proposals sent on certain days of the week, but that has completely overturned so it may have been an illusion.

And WORST of all—that's the overall rates, but the rates started out higher in the early weeks of the database. My view rate last week was just 14%, with the day of the week which had been performing even better than the boost rate, instead turning in a ZERO view rate.

I'm horrified to look at the spreadsheet and realize it's been 296 proposals in that time. Which is well over twice what I put out in the entirety of 2022, when I did not have anything like this difficulty in getting views, legit contacts, and at least some decent jobs.

Keeping in mind: I don't apply to anything except jobs I can excel at, both because they're a good fit and because I have the experience for the job; I make my proposals custom to the client's needs and keep the first few lines highly relevant and juicy, to enourage curiosity, positivity, and engagement; and other than obviously having another year of experience to offer, there isn't anything different about ME than there was in previous years on and off Upwork, when my success rate made something like sense.

.

Now, is there a platform change that Upwork could make, in this cautionary tale?

Probably. I'd certainly bet there's a change or ten that they made that *led* to this cautionary tale.

Are they *interested* in keeping highly experienced professionals happy by making such changes, so we can throw more fees at them while we're actually *working* instead of prospecting?

Signs point to NO.

[25005175]

Since the beginning of March, when I decided to start an Excel spreadsheet to slice-and-dice the data a little better, my overall view rate is 25.68%; my boosted-view rate is 40%. (I tried boosting selected proposals for a couple of weeks to see if it had any ROI—IMO, no, it doesn't.) My *legit* contact rate—not counting people who reached out but turned out to be scammers—is 1%.

---

Your viewed stats are excellent, considering the shotgun approach that you are using (256 propoals in under 2 months) AND your service set, which is high competition (even more so now that several of your services can be done by AI and other generative programs).

The more I learn about the difficulties some groups of Freelancers are having and the current/trending state of AI/generative software, the more I'm convinced that William T C is correct about the danger of AI to many freelancers. I am fortunate that generative AI is presently not a threat to my field (3D engineering design). While there are generative tools, they are expensive and in most cases are strictly for reference, because their outputs cannot be produced by anything except a very, very few 5-axis CNC machine setups and some 3D printing systems. As things stand, I've actually sent very few proposals in the past 2 months (under 10? not sure) and I've received several direct invitations, a couple of which have turned into hires. I don't pretend that it is because I am the best in my field (as my rates show), but I am certain that is because of my field that I get these invitations.

[kelly_e]

I'm sure it seems shotgun because the number is large, and I don't mean to get hung up on your use of one term—but it's careful watching and curating my feed. If I don't *have* a new UW job, then I consider my job to be *locating* one. Or leaving the platform if this is only a waste of time... hence my spreadsheet & internal data analysis. A whole lot of jobs are put up on UW daily in marketing, design, and business strategy, so even at that number, as I said above, I'm very selective. With my feed tuned to avoid lowballers, terms I know won't be productive for me, certain fields, etc., I still *don't* apply to well more than 95% of what I've selected to look at.

So if you think my open rate is "excellent," it's because I make sure the client's needs and my expertise are an excellent match before sending a word. It's dismal—takes many opens to get a contact, and at least a few contacts to get a new client, just as it does when I work with clients off UW, just as it always has in business—but if that's what passes for excellent, then Greg's concerns are even more valid.

I still preferred having more work coming at me and doing a lot less applying, as before this year, and not wondering much about my stats because my UW results were steady and increasing—but I put these stats out here to show that it's no fluke, Greg's feeling of not getting the needed views to get to YES—it's real.

Since others were discussing not having much in the way of stats to confirm the issue, I wanted to back his sense up, with a larger sample size and someone who is definitely *not* shotgun in my approach.

And no, fwiw, AI can't do what I do at all. What AI can do, I don't bother to send proposals for.

[gbuechler]

I'm with you - I don't apply to roles that are lowballs (I call that a race to the bottom), and I only apply to roles that I am fully qualified for. The lack of even opening bids is bonkers in my mind. 

[tlsanders]

Fully qualified is a very low bar. There will likely be hundreds or thousands of people here who are fully qualified. 

I only send a proposal if I am virtually certain that I will be the best possible option for the client, and the only real variables will be whether they're willing to pay my rates and whether my timeline works for them. If I'm not sure that I'm going to be among the top 2-3 candidates, I won't send a proposal.

[gbuechler]

Kelly E - THANK YOU! Your experience is exactly my point. The organic/boosted data REALLY helps in my thought process. g

[buypreciousmetal]

Now how do you get so much information on your analytic?