Effective hourly rate, the one number worth calculating

The rate on the tile and the rate you actually get

A survey tile says 15 minutes and 150 points. If 100 points is a dollar, that reads as six dollars an hour. Almost nobody earns that figure, and the reason has little to do with the survey itself.

The tile describes one event: a person who qualifies, answers at a normal pace and finishes. It says nothing about the four screeners you failed before this one, the profile questionnaire you filled in last week so the panel could target you, the link that opened to a blank page, or the three weeks your points will sit as pending before you can spend them. All of that is time the panel took from you. None of it appears in the advertised rate.

So there are really two numbers. One is the headline rate, which is the reward on a single completed survey divided by that survey’s estimated length. The other is your effective hourly rate, which is everything the panel paid you divided by everything you spent on the panel. The first is a marketing figure. The second is the only one that tells you whether an hour on that site is worth an hour of your life.

This piece is about building the second number properly. It’s simple arithmetic. The hard part is being honest about the inputs, because the inputs people leave out are exactly the ones that drag the result down.

One warning before the formula. Everything below is a method, and any numbers in it are made-up examples to show the arithmetic. Surveyfarm.org doesn’t publish per-panel payout tests, and nothing here is a claim about what a specific panel pays. Your result will depend on your country, your profile and which studies happen to be in field. That is exactly why you need to calculate it yourself instead of borrowing someone else’s figure.

The formula, written out plainly

Effective hourly rate equals total cash value received, divided by total hours spent, over the same period and on the same panel.

That’s it. Two sums and a division. People get it wrong in three predictable ways.

The first mistake is measuring the wrong thing on top. They add up points earned instead of cash value received. Points aren’t money until you’ve converted them at the rate the panel actually uses, and some of them never become money at all, because they expire, get reversed or sit below a cash-out threshold you never reach.

The second mistake is measuring too little on the bottom. They count the minutes spent inside completed surveys and quietly drop everything else. Screen-outs, loading screens and profile updates all get left out, along with the five minutes spent hunting for a study that would actually open.

The third mistake is mismatching the periods. Someone counts a month of rewards against one good evening of timing, or counts a week of time against a single lucky payout. The top and bottom have to cover the same stretch.

Get those three right and the formula does its job. It won’t flatter you. For most people on most panels it produces a number well below the headline rate, and that gap is useful information. It tells you how much of the panel’s apparent generosity is being eaten by work you don’t get paid for.

It also gives you something to compare. A panel with a lower headline rate can beat one with a higher headline rate once screen-outs and friction are counted. You can’t see that from the tiles. You can only see it from your own totals.

Input one: what counts as reward

The top line of the formula is cash value received. That phrase is doing a lot of work, so it helps to break it apart.

Start with what’s actually spendable. If a panel credits you 500 points but those points are pending for 30 days, they belong in a separate column until they clear. Some pending rewards get reversed when the research buyer cleans its data, and a reversed reward is worth zero no matter how it looked the day you earned it.

Next, convert at the real rate. Many panels quote points, coins or tokens, and the conversion to cash isn’t always a flat, published ratio. It can vary by redemption option. A gift card might give you a better rate than a bank transfer, or a worse one, and some options carry a fixed fee that hits small cash-outs hardest. Use the rate you actually receive for the option you actually use.

Then count consolation credit honestly. Some panels give a handful of points for a screen-out. That’s real reward and it belongs on the top line, but it’s usually small enough that it barely moves the result. Count it, don’t celebrate it.

Finally, leave out anything you haven’t banked. A balance sitting below a cash-out minimum is a promise, and promises can be lost to expiry rules, account closures or a panel shutting down. You can track it in its own column and move it across when it becomes cash. Until then it hasn’t paid you.

Here’s a made-up example. Say a month on one panel shows 3,000 points earned. Of those, 400 are still pending, 200 were reversed, and the rest were redeemed as a gift card at 100 points to the dollar. The cash value received is 24 dollars, not 30. The pending 400 wait for next month’s sheet. The reversed 200 are gone.

Input two: the time you remember

The bottom line of the formula is total hours spent. Start with the easy part, which is the time you’d naturally remember.

That’s the minutes inside surveys you completed. It’s usually the smallest share of the real total, but it’s the share most people track, because a completed survey feels like work that happened. You opened it, you answered, you got the confirmation screen.

Even this part tends to run long against the estimate. The time printed on a tile comes from a small test sample before the study launched, and branching logic means you might see questions that the test group skipped. A survey quoted at ten minutes can take noticeably longer if your answers route you into extra sections, or if you’re on a phone and the grid questions don’t fit the screen. There’s a separate piece on why a ten-minute survey can take twenty-five, so I won’t rebuild that here. The point for this formula is just that you should record the time you actually spent, not the time the tile promised.

The practical way to capture it is boring and it works. Note the clock time when you start a session and when you stop. A session is everything you did on one panel in one sitting, from logging in to logging out. Don’t try to time each survey with a stopwatch unless you enjoy that sort of thing. Session totals are enough, and they automatically include the gaps between surveys that per-survey timing misses.

If you use more than one panel in the same sitting, split the session by panel. Switching tabs between two sites and calling it one session will blur the comparison you’re trying to make.

Rounding is fine. Nobody needs to log to the second. Five-minute blocks are plenty accurate for a number you’ll average over weeks.

The time people forget: screeners and screen-outs

This is the input that does the most damage, and it’s the one most often left out.

Nearly every study starts with screening questions. The research buyer needs a particular mix of people, and the only way to find them is to ask. If you don’t match, you’re out. That might happen after 30 seconds or after several minutes, depending on how the screener is built and where the disqualifying question sits.

A screen-out usually pays nothing or close to nothing. But the minutes are real, and they come straight out of your hourly rate. If you spend two minutes on each of four screen-outs before landing a ten-minute survey that pays a dollar, that dollar paid for 18 minutes, not ten. Same survey, same reward, much worse rate.

Routers make this worse in a way that’s easy to miss. When a panel bounces you from one study to the next after a disqualification, each hop can include a fresh set of screening questions. The session feels continuous, so it’s tempting to only count the part where you finally qualified. The whole chain belongs on the bottom line.

How much screen-out time you’ll have depends on things you mostly can’t control. Common profiles tend to hit filled quotas more often. Rare profiles get more invites to hard-to-fill studies but also more targeted screeners. Your country matters, because a small market has fewer studies and tighter quotas. None of that is a judgement on you. It just means you can’t know your screen-out burden without measuring it.

The fix in your log is simple. Session timing already captures screen-out minutes if you start the clock when you log in. That’s the main reason session timing beats per-survey timing: it refuses to let the unpaid parts disappear.

If you want more detail, add one column counting screen-outs per session. It won’t change the formula, but it’ll show you where the time went.

The time people forget: profile surveys and housekeeping

Panels ask for a lot of information before they send you anything worth doing. There are profile surveys about your household, your job, your car, the brands you buy and whether you’ve got a pet. Some pay a small amount. Many pay nothing. All of them take time.

The panel isn’t asking for fun. Profile data lets it target invites so you see fewer studies you’d fail. In that sense the time is an investment, and a well-filled profile can reduce screen-outs later. But it’s still time you spent on that panel, and it belongs in the denominator for the period you spent it.

There’s a temptation to treat setup as a one-off cost that doesn’t count. That’s fine if you’re only asking whether a panel is worth continuing from here. It’s wrong if you’re asking what the panel has paid you per hour since you joined. Decide which question you’re answering and be consistent. For comparing two panels you already use, excluding setup is reasonable. For deciding whether to join a new one, include it.

Housekeeping adds up too. Think about reading and deleting invite emails, logging in to check for studies that turn out not to exist, or re-verifying your email when the panel asks. Updating a profile question the panel flagged as out of date counts, and so does contacting support about a missing reward. Each one is small. Together they’re a real share of the hours.

You don’t need to itemise them. Just include them in session time when they happen on the panel’s site, and add a rough estimate once a week for the email side if you want to be thorough.

An honest denominator counts every minute that exists because you have an account on that panel. Anything less and you’re measuring a flattering subset.

Dead links, broken surveys and quota-full pages

Some of the time lost on panels has nothing to do with qualification. The survey simply doesn’t work.

You click an invite and get a page saying the study is closed. You get partway through and the survey throws an error. The page loads but the answer buttons don’t respond. You finish and the completion redirect fails, so the panel never hears that you completed it. These happen because studies are hosted by many different research platforms, and the handoff between the panel and the survey host doesn’t always go cleanly.

Quota-full pages are the most common version. A study was sent to more people than it needed because not everyone opens invites. The fast ones fill the quota, and the rest land on a closed page. There’s a separate piece on why a study link dies an hour after it arrives. For this formula, what matters is that those minutes are unpaid and they count.

A failed completion redirect is the nastiest case, because you did the whole survey. Some panels will credit you if you raise a ticket with the survey ID and time. Some won’t. Either way, the time spent on the survey goes on the bottom line, and the reward only goes on the top line if and when it’s actually credited. Time spent chasing it with support belongs on the bottom line too.

Your own setup can cause some of this. Ad blockers and privacy extensions can stop a survey from loading or break the redirect at the end. That’s worth ruling out, but it doesn’t change the accounting. Broken is broken, whoever caused it.

If you want to learn something from these, add a column for failed sessions: studies that closed, errored or didn’t credit. A panel where that count stays high is spending your time on its technical problems, and the effective rate will show it.

Cash-out waits and what a delay costs

Waiting isn’t time spent. You don’t sit at the screen for the 30 days your reward is pending. So why bring it into an hourly rate at all?

Because delay changes what the reward is worth to you, and some of that change is permanent. A reward you receive now can be spent now. A reward you receive in six weeks carries a risk that it never arrives. Reversals, expiry rules, closed accounts and panels that shut down all take their share. The longer the gap between earning and banking, the more of those risks get a chance to apply.

The cleanest way to handle this is the one already described: only count cash that has actually cleared. That automatically accounts for anything lost along the way. If ten percent of your pending rewards get reversed or expire, your effective rate will be ten percent lower than it would have looked on the day you earned them. The formula takes care of it without any extra maths.

There’s also a threshold effect. If a panel’s minimum cash-out is high relative to how fast you earn, you’ll carry a balance for a long time before you can move any of it. Until then, your effective rate for that panel is zero, however many hours you’ve put in. That’s a real result, not a quirk of the method. You’ve worked, and you haven’t been paid yet.

Some people prefer to keep two versions of the number. One uses only cleared cash. The other adds pending rewards at face value. The gap between the two tells you how much you’re trusting the panel to deliver on what it owes. A small gap is comfortable. A large gap that never seems to close is worth paying attention to.

The pending wait itself costs nothing per hour. What it costs is certainty, and cleared-cash accounting puts a number on that.

A worked example, with the numbers made up

Here’s the whole thing in one place, using invented figures. None of them come from a real panel. I picked them to make the arithmetic easy to follow.

Say you use a panel for four weeks. Over that time you log 12 sessions. Session times add up to 9 hours, which includes screeners, router hops, two surveys that crashed and the time spent filling in two profile questionnaires. You also estimate 30 minutes of email and support. So the denominator is 9.5 hours.

On the top line, the panel shows 4,200 points earned. Of those, 600 are still pending at the end of the month and 300 were reversed. That leaves 3,300 points cleared, and you redeem 3,000 of them for a gift card at 100 points to the dollar. You get 30 dollars. The other 300 cleared points sit below the next redemption step.

The effective hourly rate is 30 dollars divided by 9.5 hours, or about 3.16 dollars an hour.

Now compare that with the headline rate. The surveys you completed were advertised at an average of 12 minutes for 120 points, which reads as 6 dollars an hour. The gap between 6 and 3.16 is the cost of everything the tiles didn’t mention.

If you kept the second version of the number, counting pending and banked-but-unredeemed points at face value, the top line becomes 39 dollars and the rate rises to about 4.11 an hour. That’s the optimistic version. It assumes every pending point clears and every remaining point eventually gets redeemed.

Neither of these numbers is good or bad in the abstract. They become useful next to your other options. If a different panel comes out at 4.50 on the same method, or a different use of those nine and a half hours pays better, now you know. If nothing does, at least you know what you’re choosing.

Why comparing per survey misleads you

The most common way people compare panels is by the reward on a typical survey. Panel A pays 2 dollars for 15 minutes. Panel B pays 1 dollar for 15 minutes. A must be better.

It might be. It might not. The per-survey figure only holds for surveys you complete, and it tells you nothing about how many screeners you had to fail to find one.

Picture two made-up panels. Panel A pays more per completion, but its invites go out broadly and most of them screen you out. You spend 40 minutes of screeners for every completed 15-minute survey. That’s 55 minutes for 2 dollars, or about 2.18 an hour. Panel B pays less per completion, but it targets well and you qualify most of the time. Say 10 minutes of screeners for every 15-minute survey. That’s 25 minutes for 1 dollar, or 2.40 an hour. The cheaper tile wins.

The same thing happens with cash-out rules. A panel with generous per-survey rewards and a high redemption minimum can leave you with less banked cash than a stingier panel with a low one. And a panel that reverses a noticeable share of completes can look excellent in points and mediocre in cash.

This is why the effective hourly rate is worth calculating even though it takes a bit of discipline. It’s the only figure that includes all the ways a panel can cost you time or withhold value. Per-survey rewards, bonus tiers, streak multipliers and point totals all measure one slice. The effective rate measures the whole thing.

It also cuts through bonus schemes. A loyalty tier that adds ten percent to rewards is worth ten percent of the top line. If reaching that tier means spending more time on low-value studies, the effective rate will tell you whether the bonus actually paid for itself.

How much data before the number means anything

One evening of timing tells you almost nothing. Survey supply is lumpy. Some days a panel has several studies that fit your profile. Other days it has none, and every invite leads to a screen-out or a closed quota. A single session can come out far above or far below your real average.

So give it time. A few weeks on each panel is a reasonable minimum, and more sessions are better than longer ones. Ten short sessions spread across different days and times will give you a steadier figure than two marathon evenings, because they sample more of the panel’s varying supply.

Watch for seasonality too. Research budgets aren’t spent evenly across the year, and some months go quiet. If you measure only during a slow stretch, you’ll underrate the panel. If you measure only during a busy one, you’ll overrate it. Keeping a running total over months smooths this out.

It helps to track the rate over time rather than recalculating from scratch. Keep cumulative totals for cash received and hours spent, and recompute the division each week. The number will jump around at first and then settle. When a few weeks of new data barely move it, you’ve got enough to trust.

Be careful with changes you make along the way. If you finish your profile surveys halfway through, your screen-out rate might drop afterwards. That’s a real improvement, but it means the early weeks and the later weeks are measuring slightly different things. Note the date of any change like that so you can compare before and after.

And don’t compare panels measured over wildly different periods. Three months of data on one site and one week on another isn’t a fair fight. Match the windows as closely as you can.

Why your number won’t match anyone else’s

Two people using the same panel for the same month can end up with very different effective rates, and neither of them is doing it wrong.

Panels route studies based on your profile. A research buyer looking for parents of young children, or people who recently bought a car, or a certain age band in a certain region, will pay to reach those people and ignore everyone else. If your profile matches what buyers want this month, you’ll see more completions and fewer screen-outs. If it doesn’t, the opposite. Your answers about age, household, job and region shape the invites you get.

Country matters a lot. Research spending follows large consumer markets, so a respondent in a smaller market may see fewer studies, tighter quotas and more closed links. The per-survey rewards might look similar across countries, while the effective rate is quite different.

Device matters too. Some studies are desktop only and some are built for phones. If you only use one, you’ll be excluded from part of the supply, and the ones you do get might run slower on the wrong screen.

Then there’s account history. Panels keep internal quality ratings based on things like speeding, inconsistent answers and reversals. A strong history can mean more invites. A weak one can mean fewer, without any warning.

All of this is why a published figure for what a panel pays is of limited use. Even an accurate figure describes someone else’s profile, country and history. The effective hourly rate you calculate describes yours, which is the only one that matters for deciding where your own time goes.

It’s also why the number is worth recalculating when your circumstances change. A move, a new job or a new device can shift which studies reach you.

What the number leaves out

The effective hourly rate is useful because it’s narrow. That also means it misses some things.

It doesn’t measure how much you enjoy the work. Some people like answering surveys while half-watching television. For them, an hour on a panel isn’t really an hour taken from something else, and a low rate might be perfectly acceptable. Others find surveys tedious enough that even a decent rate isn’t worth it. The number can’t decide that for you.

It doesn’t capture the value of flexibility. You can open a panel for five minutes while waiting for a bus. Most paid work doesn’t allow that, and that convenience has some worth that won’t show up in dollars per hour.

It doesn’t price privacy. Profile surveys and some study types ask for detailed personal information. Handing that over is a cost, and some people would rather earn less than share more. The formula treats a question about your health the same as a question about your breakfast cereal.

It doesn’t fully capture risk. Cleared-cash accounting protects you against rewards that disappear before they’re paid, but it can’t protect you against a balance you’re still building. If a panel closes while you’re below the cash-out minimum, that balance is gone, and past effective rates won’t bring it back.

And it can’t tell you about studies you never saw. If a panel is quietly throttling your invites because of a quality score, your measured rate reflects the reduced supply. You’ll see the effect but not the cause.

None of this makes the number less worth having. It means using it as one input. Where the rate is poor and nothing else redeems the panel, that’s a clear signal. Where the rate is poor but you enjoy it, you’ve at least made the choice with open eyes.

What to do this week

Keep it small. You’re building a habit, not a research project.

Pick the panels you actually use, two or three at most. For each one, start a simple sheet with the date, the panel name, the session start time and end time, what you completed, what screened you out or broke, and points earned. Once a week, add a row for cash received that week, meaning money or gift cards that actually landed.

Time sessions from login to logout. Don’t try to separate the paid minutes from the unpaid ones, because the whole point is that they all count. If you do panel email in batches, add a rough weekly estimate for that too.

After two weeks, do the division for each panel: cumulative cash received over cumulative hours. Write the result next to the headline rate the tiles suggest. Look at the gap.

Then keep going. After a month or so, the numbers should settle enough to compare. At that point, you’ve got three options for each panel. Keep it if the rate is acceptable to you. Change how you use it if a pattern is obvious, such as always screening out in one category or losing time to studies that crash on your phone. Or drop it, and give those hours to a panel that pays better or to something else entirely.

Don’t expect the result to be impressive. For many people on many panels, it won’t be. Surveys aren’t a full-time income, and the effective rate usually makes that plain within a few weeks. What it does do is stop you spending time on the worst options by accident, which is the realistic goal.

For more on how panels set pay, screen people out and decide what your time is worth to them, head back to the Survey Farm home page.