Fiverr Community Forum

Current state of things, Frank's POV

Hey everyone!

As promised I will try to create a post, in an effort to shed some light over the issues that most of us seem to be facing ever since we got hit with the wonderful twofer of Covid-19 AND the analytics system breaking down.

I will share what little insights I have over how SERP may or may not work, but I will avoid diving into too much detail as I have in no way figured things out. (Nor is that my intent)

OK so to kick things off, let me just state the obvious:

Everything I share here is based on my personal observation, and I am not making any claims that I have a hack or I am certified to tell people what to do.

Same, same…but different
I have been on the forum since 2013 and please do keep in mind that seller’s number 1 gripe is their ranking and how unfair the system is.

At first everyone wanted to be on the front page, as that meant you got dozens of deliveries per day, until you passed out trying to facilitate all orders.

Then as the marketplace evolved, everyone started complaining about their gig’s rank in search results.

So while this issue is constantly coming up as it is our nature to complain and looking for something to blame when sales are low, I noticed that every time the marketplace shifts the number of people complaining surges.

I think that every time Fiverr makes a change in how the algorithm processes things, a lot of sellers -even senior ones with a proven track record- start getting these weird blackouts, as I call them.

It’s when you get a feeling that someone turned off the faucet that was getting you leads/sales. Small periods in time when it feels like it’s the end of your career on Fiverr, as no one is making a peep. And to make things worse, you see other sellers in your category that have orders in queue.

Contrary to public belief (a lot of people think that since I am a PRO and TRS level seller, everything is coming up roses) I too am hit every time something changes.

Here’s my sales analytics for the past 30 days, let me know if you can spot the blackouts:

Keep things in perspective
If our gig is on the top of search results, we feel like it’s only fair, as we view our selves as the best seller that ever was, and our gig will always be superior to those of our competitors.

When our gigs rank lower on search results, especially when newer gigs or gigs with poor reviews rank on top of ours, then surely:

-things must be broken
-Fiverr is out to get us
-the system is rigged
-someone is hacking their way to success

You know… All that good stuff…

So what is up?
Ok so this is what I keep saying, and people refuse to listen to:

Fiverr’s algorithm isn’t broken. It’s a really complex combo of software and machine learning, designed to do one thing, and one thing only: make Fiverr money, quickly, and with as little risk possible.

A lot of sellers take things personally. Fiverr is unfair, CS agents are refusing to help, the system doesn’t care about us.

“If my gig doesn’t make money, how will Fiverr profit?”
I see a lot of sellers who used to be on a hot streak present that argument.

Well, there are thousands of sellers like you, and millions of gigs just like yours, so if you don’t make them money, someone else will.

Things may seem irrational, because they are. This is essentially a robot programmed to do that one thing. If it seems illogical to you that you used to be at the top of the search results and now you are not, I assure you there is definitely some logic behind it, it’s just not one that we can comprehend with what we know.

OK now it’s time for the good stuff!
That being said, I do feel as there’s an issue here. In fact there is a symphony of a lot of tiny issues that have cooked up quite a storm:

•We are living in abnormal times, where a pandemic is shocking our financial systems with new threats each week

•There has been a surge in new sellers, dropping in on an already saturated marketplace

•Fiverr definitely messed something up when they started implementing their promoted gigs feature (which seems to have cause all gig analytics to just stop working altogether)

All of the above, are acting as an accelerator that makes the algorithm move quicker.

Time to put your tinfoil caps on!

As I keep saying there are several metrics that the algorithm uses in order to come up with the optimal SERP to streamline Fiverr’s profit.

I can start naming them, but to be honest it’s mostly guesswork, but we all know there are several factors in play here. Here are the ones I think Fiverr’s search engine is mostly looking at:

-your gig’s SEO (it’s used to define your relevance)
-your gig’s potential value (it’s used to better match your offering with buyer’s behavior)
-your gig’s performance (this is on two levels: turnaround efficiency and secret reviews
-your profile’s “status” (your ability to close, meaning convert leads into sales)

That’s just the tip of the iceberg, as there are a bunch of variables we don’t know, and will never be able to find out.

I think the best way to explain each point is by discussion/answering questions, so I will be active on this thread to talk with any sellers that want to exchange experiences and shed some light into this.

(I also have a theory that Fiverr no longer pushes for big-ticket items, as they pose a greater risk, with a longer turnaround time. So in my case that was truly worrisome, as I shifted my core business to cater bigger projects/clients back in 2017.)

It’s all about relevance

As I have mentioned before on several occasions, Fiverr search results are more dynamic than we think. Good gigs don’t go to the top and stay there. Getting to the top of the search results is not a reward. It’s not something that happens exclusively to good sellers with great gigs.

Fiverr serves -what it thinks- are the most relevant results to buyers’ queries. That’s it.

That being said, based on my personal experience, after the initial shock of Covid-19 plus Fiverr’s screw up, I have come to one conclusion that has been useful to me.

We are now in a phase where Fiverr sellers are going to have a hard time staying relevant.

The marketplace is -and always will be- trying to correct itself. So staying relevant means we now need to be reactive. We as sellers need to react to what’s happening, based on our personal goals, financial needs and the data we have on our hands.

So in my case, here’s when I reacted to the marketplace:

So this is the same graph as before, but I now have marked when I started re-acting to what the marketplace was telling me: “you are getting more and more irrelevant by the minute”.

Mind you the 30 days before this graph had even more blackouts in them.

So what did I do?

I started closing faster.
Meaning I started taking projects I would normally reject, or in some cases I reduced frictions that usually occur due to budget constraint or deadline’s overlapping.

This is me reacting to the situation, by compromising at my discretion in order to get some orders in, which would signify Fiverr’s algo that I am still able to perform.

I tweaked SEO, only on dead gigs
Once again a reminder that I am part of the “success” program, meaning that I can change the entire description of my gig, plus all tags and my gig will remain visible in search results, no downtime for manual review. So tweaking stuff for me meant that I wouldn’t have to wait 48 hours to see if my gig started appearing again or not.

Even then I only made minor modifications, and only on gigs that appeared to be completely dead.

I shifted my focus when delivering orders
I spent more time asking questions in the beginning of the on boarding process, so that I could hit the nail in the head in the first go, avoiding revisions requests. I did so even on my “templated” gigs, you know, the ones I consider to be foolproof.

Of course none of the above applies to all of us, since in order to be able to react like that means you still get some leads and some orders in, even if things have slowed down.

That’s why I think that instead of me writing more things I tried on this post, it would be better if I responded to questions so fellow forum members can ask me about the situation they are currently in, right now.

This has been a rather long post, I will stop writing here and I look forward to having a discussion with anyone who is willing to participate.

I am here to help anyway I can.

Thanks gang. :slight_smile:

69 Likes
Fiverr gig went to last page or almost not showing up on the search results properly
If Fiverr alive? - What Happened?
Huge gig problem please help I dont know what to do anymore!
How get more order in my gigs?
Gig constantly dessappearing from the search list
What to do when your gig with plenty of positive reviews thrown back to last pages?
Buyer Reported Me After 5 Star Review
What is about Dead gig?
Any update about gig search issu?
All my gig have gone from the first page to the last page
Gigs are going down day by day, frustration increasing
Not received a messages or order from a ranked gig almost 3 months
Why my Gig rank is down?
The first time in a (somewhat) long time with no new orders
Gig not showing in searches
A significant drop in my Best Selling Gig
What will happen delete all my gigs?
Gig ranking decline
Fiverr GIG on last page
No order since 3 years
Not getting new orders in June?
My Gig is not showing In search result
Gig aren't doing so well lately
Fix the Algorithm
Fiverr support is not Answering
How are your sales going?
Rank my gig please
Fiverr Sucks for Unpopular People
My gig was ranked at the top of first page but now suddenly it disappear and I saw it nowhere. Why?
Why I am not getting enough work now
Several gigs NOT showing in search results AT ALL
Why My Gigs are now Showing in Searches
Is it time to edit my gigs?
3 best selling gigs are out of search from last 2 months
My gig impression are down day by day but my gig active and running 7 active order
FIverr Gig Ranking in less than 1 month
I Search gig on Mobiile Apps but do not show my gig on search result
Delete & Create new gig vs Enhance & Optimize existing gig
My gigs are no where to be found
Fiverr Gig Disappeared From Searches

Thanks for this, Frank - last time, when you made the videos, I radically changed one of my gigs. It worked.

Probably have some questions later, but need to digest this first …

4 Likes

First of all, thanks a lot Fred for sharing regularly your point of view and advices on this, I followed them when I started to be more active on the forum, and I found them to be quite helpful. I will share my experience with this “mini crisis” for the sake of the discussion:

  • I had only 3 orders in the last 30 days, and only 4-5 messages. Before that, it was my gold period, had so many orders with which I reached level 2 today.
  • Just before these 30 days, I changed my gig price (I had one gig only) by including the extras inside my basic price. I also included a better video with my work samples. So its basically the same prices I was offering. Then, the algorythm update started. After that, I’ve updated and activated older gigs which were on pause previously with prices in line with my new increased prices in my “main” gig. The reactivated gigs brought me only one order in the last 30 days.

What is interesting for me is that you follow your gig statistics by taking into account the daily statistics, and you react if you don’t have order for several days. I am still far from that level of incoming orders, but is worth asking: how many days should we wait before we take action and tweak our gigs? I guess it is one of the milion dollar questions.

Yeah, it is my impression that there must be milion other things beside those you were mentioning taken into account by the algorythm. Even these four main ones you are mentioning they evlolve quickly and they may require an update after x months or definitely after new change has been implemented for the algorythm. The system seems to be very dynamic, impossible to follow and many people don’t know how to react to it, including me. Hence all the complaining and panic.

I have, for example one open gig that will stay open for some time, the buyer is still working on the product I need as an input in order to finish tmy work. I think that even that influences the algorythm.

What do you call turnaround efficiency?

Anyway, it is the second time I am experiencing this type of longer periods of “almost total silence” in one year that I am active on fiverr. The first one was because of cancelled gig and my order completion rate went down to 50, 60%. And I still don’t know how to react when this happens.

6 Likes

Well, here’s one reaction - the tsunami of new sellers seems to have been matched with a large wave of new buyers. The new buyers are resulting in a lower average spend per order. Presumably that’s why you had to react the way you did.

Stupid thing is, those buyers were probably spending more on permanent employees than they expect to spend here.

If that’s the case, then cyaxrex was probably right about increasing prices precipitately … more value is key, rather than less access. For example, fewer words for $20/25/whatever with more extras available … is that more or less what you’ve discovered. Not that you “do words” any more than I do, but you get the idea …

5 Likes

Hey @coerdelion I am glad to hear that you benefited from my live streams and got some results.

As for your reply here I think you are right.

There definitely was a spike in buyers here that got matched with new sellers due to how the algo defined relevance.

I had a quick chat with my SM and she was hellbent on me not lowering my price, but offering less at a more accessible price point.

She offered me several examples of other Pro sellers who offer a 30 second video as a basic package and they don’t do script writing or strategy calls. They had more orders than me during this drought.

I was extremely hesitant to consider offering a service that forces the buyer to purchase add-ons in order to get a complete 1-minute explainer video, but I am not going to lie: I am now considering it.

@audioboon Hey there!

Nice to hear from you again!

By turnaround efficiency I mean how efficient your customer journey is from the moment a purchase has been made till the second the buyer clicks on the “accept delivery button”. That’s all that matters to Fiverr.

So if I deliver relatively fast, and there are few or none requests for modifications, then to the eyes of the algorithm I am an efficient seller, since I am streamlining the process from purchase to Fiverr pocketing the dough.

As for your question, I don’t have any recommendation as this is something that differs between you and me.

The drought is perceived differently by me and my account’s standards.

That’s why I check my analytics on a daily basis. That makes sense to me since I assume I get more messages from prospects and I have more gigs across several categories.

There is no rule of thumb either.

If you were getting regular orders and/or inquiries and that suddenly changed, then you know something is up IF that persists for several days.

All sellers are bound by 3 factors that are vastly different for each and everyone of us:

  1. Probability
  2. Context
  3. Skill

So in essence: Luck, operating conditions and ability to perform a task at a high level of proficiency.

7 Likes

Something that has been on my mind about the whole algorithm thing is a combination of the following two things.

Situation 1:

  • many people rarely or have never edited their gigs.
  • Fiverr has had many changes to gig options in the last few years with new sections, subsections, features, “included” sections etc.

This means that there are likely to be a lot of sellers who have incomplete gig descriptions - eg. Have an SEO gig that doesn’t specify what type of SEO they offer (on-site, off-site etc), or have a writing gig that doesn’t specify the writing style or content specialization etc.

Situation 2:
Algorithms are constantly being tested to see what effects things have. This can mean that (for example) in January, a lot of importance is given to one factor but then in March, the importance is given to something else. This could cause a complete change in rankings.

Now, if you combine the two situations here, someone who has never selected certain features for their gig will be moved to the very last page when that feature is prioritized in the algorithm, as they have a zero/null value on that.
What’s more, very often when I have edited gigs, I have noticed a new feature and just entered it, thinking maybe I missed it the first time or something. So when some people edit their gig, they do the same and magically, they are back ranking again!

Everything in this comment is my own opinion and conjecture, I haven’t done anything much in terms of researching it but there is a logic to it. If someone’s sales have tanked then there is likely to be no harm in checking if there’s a feature they haven’t got included on their gig. Note: Editing a gig will likely make it disappear from search for 48 hours.

12 Likes

Could you provide me the link to your streams or tell me the platform you were streaming on, @frank_d, I would love to watch your streams and have a guidance of a Pro+ TOP Rated seller

2 Likes

Here: https://www.youtube.com/channel/UChRBiv_tDYKzyPNYuJkAQzA

I was rather hoping Frank would do more of these …

8 Likes

Hey @eoinfinnegan !

Always happy when you chime in.

Great insight. I used to wait for Fiverr’s prompts in order to edit my gigs.

So whenever a notification would pop up that they moved one of my gigs (happens more often than you’d think as the marketplace is expanding) or I got an email from my SM about a new feature, then and only then would I edit my gig to fit the new things the algo looks over.

However I did notice that after tweaking my “dead” gigs, things started picking up again a couple of days later. (As shown in my analytics snapshot)

I think there’s definitely some truth to your assumption and it makes sense that if Fiverr changes its categories or adds more meta data, our gigs need to be adjusted accordingly.

I also think that due to the amount of gigs we are now seeing across all categories in the marketplace, Fiverr may also favor “active gigs” more, meaning gigs that we tend to vs gigs we haven’t touched in months.

There may be something there too, as Fiverr is trying to make sure things stay relevant for buyers.

Once again: YMMV and there’s no recipe for amount of edits per days or formulas I see other forum members share.

You need to be able to improve your ability to understand your unique situation by observing and then make decisions.

3 Likes

Can you elaborate on what made you develop that theory, as greater risk and longer turnaround time are kind of built into big-ticket items, and those factors were there all the time, not just now? You think Fiverr is just more risk-averse now, because they sometimes refund sellers?

Average gig price is pretty prominently featured on the Analytics page, and I assume it’s a ranking factor as well. Then I’d guess that probably many buyers who order high-ticket items are companies, and as such perhaps rather more reliable and pose a lesser risk than reseller-buyer-scammers we read about on the forum so often and who buy small-ticket items usually, or?

Also, what is a big-ticket item/longer turnaround time for you, respectively, what do you think is it for Fiverr?

I’m not quite sure about the correlation of order amount and delivery time, too.
Maybe not a big-ticket item for you, but for brevity, let’s say seller A sells a $500 gig with 7 days and B sells a $100 gig with 1 day turnaround time, a $300 gig with 5 days and a 100 gig with 7 days turnaround time. Fiverr gets the same money from both sellers within the same time frame. A's buyer has to wait 7 days for his 500 gig, while B’s first buyer only needs to wait 1 day but also gets only $100 worth of work and B’s 3rd buyer needs to wait 7 days, like A’s buyer, but for a $100 order instead of a $500 one.
On the surface, it would seem that seller B is better off, and the algorithm works in favour of seller B, as B makes Fiverr the same money, can deliver 3x and potentially gets 3 good reviews while A only delivers 1x and potentially gets only 1 review, however, the “average gig price” metric might be there to balance that out a bit in favour of A? After all, A makes Fiverr the same money in the end. Maybe the risk that A’s one buyer cancels is higher (not necessarily, I think) than that all 3 buyers of B cancel but on the other hand, you could also say that the risk that something with one of B’s 3 buyers goes awry and B needs support time is higher.


I sound like a broken record, but I agree with Eoin :wink: Absolutely even, on that post. I have also a very good example; when I edited my gig after Fiverr had made some changes to the category once, I found that my gig had somehow lost its language pairs, which obviously is pretty dramatic, so, yes, whenever Fiverr makes changes to a category or subcategory, that might be a good time to check on your gig (and, before you even hit the edit button, take a good look at your gig/s and ponder if there’s anything you might want to change of your own volition too, so you won’t have to edit it more often than needed while things are going well for you, also, a chance to revisit your gig requirements and FAQ, with passing time and buyers, you might want to adjust them as well).

5 Likes

Just, y’no, my 2p … but something changed around new year. I had a couple of big (ish) ticket orders go south at that time … and that was also the time that the tsunami began to grow …

I was compensated for both - eventually - but now Frank’s put all of this together - possibly.

Lots of new sellers + lots of new buyers = lots more risk. Big ticket orders may be more likely to come apart … which costs Fiverr money. Big money, if there’s a lot of them …

PS - I’m working on a big order right now which is not likely to crash. But divided it in two, just to be safe. The second order is not as big, but still. Looks like I’ll have to regig that gig again …

4 Likes

I concur with your assessment @frank_d, I think fiver is striving for cashflow conversion. Quicker deliveries with less revisions mean less probability of the buyers cancelling and a better buyer impression. I already vet in-depth proding questions for prospective buyers in order to cut revisions, as I hate revisions with a passion. This has forced me to focus much more on the sale to keep my conversions up.

5 Likes

“Yes but”, the majority of cancelling and endless revision stories we read on the forum are about low-ticket items (I suppose usually with quick deliveries), at least that’s my impression. Personally, I’ve never had issues with (for me) higher-ticket items (doesn’t mean it won’t happen eventually, of course).

4 Likes

@miiila yes but I think a lot of those low ticket sales are buy new sellers and sellers dealing with buyers asking the world for next to nothing.

1 Like

Based on what we read on the forum, yes, sure, but never mind the reason, the majority of cancelling and endless revision stories seem to be for low-ticket orders.

It’s possible, of course, that sellers who have those issues with high-ticket orders don’t post about it so often, or never post on the forum, but there’ll be also lots of sellers who have those issues with low-ticket orders and don’t post on the forum. And, generally, there probably are a lot more low-ticket orders than there are high-ticket orders. It’s also possible, though, that few sellers will bother support about cancelled orders of low value, while probably many if not all will do so with high-value, or even “highish”-value orders.

We need to define what a high-value order is :wink: first, though, if we want to dive deeper into that, or want to pull some actionable advice for ourselves out of it :slight_smile:

3 Likes

I don’t usually read long posts but man this was just so interesting piece of writing. I absolutely loved reading it. This is so true and so relevant. Fiverr is dynamic and you gotta stay relevant in order to keep making money. Sitting and waiting for the gig to show up on top or waiting for the buyer to magically find you isn’t going to get any seller anywhere. You gotta study, analyze, assess and then strategize.

Thank you for such a quality piece. Had so much that I always wanted to share. Glad someone did. :+1:

6 Likes

As @eoinfinnegan pointed out, the algorithm is highly dynamic. Something that had a weight of 0.5 may have a weight of 6 just two months away.

Once again this is definitely biased on my end, I have to admit.
My assumptions are based on the fact that I sell high priced gigs, and I have an average gig price of $60 (definitely above average) and yet I see lower priced gigs performing better.

I think it’s not just a matter of high price, long delivery times, it’s also a matter of converting leads to sales. Someone messages me and says “wow, I loved your portfolio. How much to get a 1 minute video done?” To which I respond with my +$4k rate.

The buyer then either politely declines, or just stops responding altogether.
That same buyer then messages a seller with a $30 price tag, and happily buys their service.

Fiverr now “learned” that I didn’t close, whereas “explainers4cheapz” did.
If this happens 3-4 times a week, I may be labeled as a low-performing account or a liability.

I can only answer re: my own standards. I sold a gig for $13,000 and it took me 30 days to deliver, then about 12 more to get 3 rounds of revisions done. (big clients take days to figure out changes even on a 60-second video)

So it took a month and a half, for Fiverr to make $2,600 off of that sale.

During that period, I am sure “explainers4lolz” made roughly the same amount, selling gigs at approx. $150 a piece, and their queue is always up to double digits. (this is an actual Pro seller in my category, I changed the name to protect them)

Bottom line is: it’s not just how fast I make Fiverr money.

It’s how I teach Fiverr about how I perform every time it sends me leads.

If they keep sending me assists, and I don’t dunk the ball, they will start passing the ball to other players.

That’s my assumption in a nutshell.

Oh one more thing worth exploring: I firmly believe that each vertical has different weights for the same metrics. Explainer videos and websites are treated differently than logo design and voice overs. :slight_smile:

7 Likes

Some really interesting thoughts here @frank_d , thanks for taking the time to put it together.

Your theories make a lot of sense. From our perspective, we’ve been operating pretty successful voice over gigs for the past 3 years, and whilst we’ve had the standard ebbs and flows on both our male and female VO gigs, we’ve noticed they’ve both been pretty steady over the years.

Contrast that with a screencast video gig that we launched toward the back-end of last year - despite being in a much smaller talent pool, we’ve found that much, much harder to grow and develop.

We charge ‘big ticket’ costs (compared to our voice over gigs at least), we had some cancellations a few months ago (some where we were to blame, others where the client was) and find that 9 out of 10 of the enquiries we get are tire kickers, who do exactly what you describe - send a ‘how much’ email, then disappear when you try to engage them, especially if you mention price.

The gig went from around 5 enquiries a day, which would allow us to convert a handful each week, to enquiries dropping off a cliff. We’ve made some changes, and managed to complete a couple of orders lately, so we’ll see if our recent order completion combined with lead-conversion you talk about gives the gig the shot-in-the-arm it obviously needs.

One of our VO gigs is also massively out-performing the other at the moment. We do turn down a lot of work, usually due to client budgets being too low for our prices. Perhaps it’s a good time for us to experiment with compromising for a while and seeing what happens.

4 Likes

Thank you for taking the time to respond @cubittaudio

Always nice to see what you think.

I really feel like most sellers on Fiverr keep forgetting about the different types of performance that are being measured.

Everyone’s question is: “I created a fantastic gig, with an amazing thumbnail, so where is my top ranking spot?”

I think that most people who can sense the pulse of things here on Fiverr as sellers, definitely can relate to what you described as “inquiries dropping off a cliff”.

It’s like someone flicks a switch and no messages go through. Radio silence.

I think that when that happens, there’s definitely something going on related to performance.

1 Like

Thank you, some interesting things in there, including your average gig price in comparison to your high-ticket items. My average gig price is higher but my high-ticket items aren’t nearly as high, that probably means I have more medium-range orders, which probably don’t take quite that long to deliver usually - though I do have gigs with 30 days delivery time sometimes, and I definitely agree that the algorithm probably doesn’t like that so much.

I also think that time to react to revision requests and to deliver revisions and the final completion of the gig certainly weigh a lot, I think you called it turnaround efficiency, and probably the factor secret reviews is highly influenced by that too (I imagine, even if buyers give a great 5* public review, they might say things like “great but could have been faster/but with the revision, it took much longer”, etc.).

So, was your worry about your theory because you shifted your core business to bigger projects linked to the actions you took, i.e., did you re-focus to get smaller projects because of that?

1 Like