By giving a number first, you're either too high and you won't have an opportunity to sell yourself, you're too low and you've literally already negotiated against yourself, or you're lucky and right at their expectation, which you could have gotten to by simply not giving the first number.

I do agree with the wasted negotiating time thing.

There are a lot of bad jobs in the world, you should be happy to turn down or be turned down by many jobs that you apply to. It's kind of a sad cultural trope that the employee is desperate for a 'yes' at every job they apply for, when that is often not true at all.

Yes, salary is a filter, because money isn't infinite and the hiring budget is only so much. Yes, a certain category of hyped-up, massively funded startup may (temporarily) exist in an alternate universe where the laws of economics don't apply and all budgets are imaginary ("for the right people..."), but the rest of the universe doesn't work like that.

For example, a small company bootstrapped by the founder's limited personal funds simply isn't going to offer $250k pay packages to Full Stack JavaScript Rock Star Ninjas. In some cases, that's close to the entire gross revenue of the company. Yet there may be advantages to working for a company like that.

Likewise, large orgs and state agencies function in normal-law economics, not RockStar VC economics. There's only so much they can afford to pay, and that's that.

Negotiations aren't just a staring contest where, if you do the right things, you get paid like a banker, otherwise you toil in the lowly ranks of the unwashed proletariat. In theory, negotiation leads to compromise. Yes, employers have more leverage in these conversations often, but it doesn't change the fact that negotiation requires credible and bona fide compromise intentions from both sides.

Take the time to understand the nature of the business to which you are applying. Yeah, you might just be getting squeezed by MBA frat boys sitting on top of tens of millions in VC who are going to get rich off the sweat of your toil for peanuts, but in 90% of the country, that probably shouldn't be your first thought.

"They try to filter people by salary" is not at the top of my list of reasons for why I wouldn't want to work somewhere. It's a reason, but it is dwarfed by many other ones.

It's an absurd example, but let's say you don't get hired because the hiring manager thinks your shirt is super ugly. That's insane and obviously completely improper and irrelevant -- but you're probably better off not getting that job!

Anecdotally, even as a software & full-stack web developer that's worked in half a dozen languages and has a math degree, I've been in both positions -- I've had the freedom to complain about people annoying me with unsolicited offers to interview while banking a third of my paycheck and treating another third as disposable... and I've been pavement-pounding in desperation.

The former has been more often the case than the latter for me, but I think it's probable that like a large portion of the audience here, I've been insulated by circumstance.

But it's not a rule that applicants are desperate, and hiring parties aren't. It's absolutely a stereotype of the hiring process though.

And that's why salary is a binary search!

https://thehftguy.com/2017/01/23/career-advice-and-salary-ne...

Except that isn't a binary search.

In an otherwise excellent article, the missing negotiation calibration and example is glaring. The rest of the article would seem to suggest just as you will get your next big promotion from another company, if you find yourself considering coming down walk away. The heart of the approach is getting away from you becoming invested in being hired by a specific company.

I am not sure I understand your second paragraph.

Yes, networking and insider information is much better than any of this.

First, do you really want to work for the lowest bidder? I have before, and it was a complete waste of my time.

Second, Name a number that you can be happy with. If they meet it, then you should be happy.

Third, many companies use banding for salary range, and if you're below the median, then you're raises will be bigger than if you're at the top of the range. In effect, naming too small of a number is often nullified within 2-3 years.

I do agree that it's a psychological game, but don't play it. Name a number you want. If they are reasonable and they like you, they'll negotiate.

In one case, I was too high for the position, and they created a new position (title, really) to put me in. I thought that was good faith.

Maybe I am naive, but I feel like this creates a dynamic that hurts otherwise promising prospects.

You've hit upon a fundamental tenet of Marxism: alienation [0]. You haven't said whether or not you offer equity as part of your compensation. If not, then it is somewhat unreasonable to expect that sort of loyalty from your employees. If so, then you have a challenging task of convincing people to buy in to your vision at a potentially high personal risk.

Today's economy is rife with alienation. So many people are constantly changing jobs or dropping out of the workforce [1]. Meanwhile, they're struggling inside a system that produces rampant cost disease [2] and staggering inequality, as illustrated by the dreaded "elephant graph" [3]. Overall, I think there's a growing sense of discontent with capitalism and this can manifest in suspicion towards employers.

Now I'll grant you that it's entirely unfair to lay all this at the feet of just a single employer such as yourself. I don't intend for this to come across as some kind of blame game. I merely hope to provide a larger context for ideas that many people may not even be consciously aware of but contribute to their feelings and instincts nonetheless.

[0] https://en.m.wikipedia.org/wiki/Marx%27s_theory_of_alienatio...

[1] https://www.brookings.edu/blog/social-mobility-memos/2017/02...

[2] http://slatestarcodex.com/2017/02/09/considerations-on-cost-...

[3] http://voxeu.org/article/greatest-reshuffle-individual-incom...

Luckily, most people out there aren't purely capitalistic, do-whatever-it-takes-to-make-a-buck sociopaths, but they're definitely out there, will play you like a pawn, and it might be a while after (or never) until you realize what happened and what role you played in their twisted little game.

To me, it's sad that we have humans literally dying because they can't afford better health insurance in this country while others play pricing games having ships full of oil sail around to manipulate the market into paying them more, or toiling to shave off a tenth of a millisecond for their high-speed quote stuffing system. The idea of money has become so remote and intangible in some cases that I hate how it's also used for things people actually need.

Oh, the "I better be the most well-compensated person on the team or I'm going to get upset" mentality. If our strategy weeds those folks out, all the better.

I try to only ever apply to jobs with salary range posted. Although as a developer is becoming increasingly hard to find those posts

rip market.

Anchoring is a more general marketing concept that works well in advertisements (crossing out a black $280 and putting a red $100 underneath it). Anchoring doesn't work so well for something much more fluid (and with much more at stake) such as a salary negotiation.

In-person sales meetings (which is what an interview is, for both sides) are where "first to talk" come into play. It's a concept as old as war meetings between opposing roman generals. First to talk loses, period. Why? I don't know, but years of sales and recruitment has taught me that whoever slips is going to give the advantage to the other side of the table.

You can only hurt yourself by giving a number. The company has set a budget, either you fit or you don't. Maybe the budget is higher than you expected and you take the job for way more than you would have asked for. Maybe it's less and you don't take the job. The key point is that company is not the only company you are dealing with - you can move on to another. And another and another, and as long as you don't give your number, you will win out.

I can't share this data but when I worked at a bootcamp our analysis showed this as well. Students that gave their number got paid less. There's a big push for bootcamp transparency so hopefully that sort of info will get published soon.

I think it is difficult for employees to judge what they're truly worth. Therefore, when most people give a number, that number will be lower than what an employer may be willing to pay (and perhaps lower than the company's initial offer would be). I would guess this holds doubly true for graduates of a bootcamp, or recent college graduates, who have less industry experience on which to base their judgment of business value.

So this reasoning holds if you have an abundance of time, but you can generally figure out what their likely range is going to be. Do research, ask colleagues, etc. Anchor at the high end of that - you can also leave it vague by talking about 'total comp' which can includes subjective things like equity, location, vacation, etc. You may loose some value by doing this but it'll also mean you won't waste time and so you'll only be interviewing in places that are in your ballpark.

Asking for the candidate's expected compensation, OTOH, is fine, although a little tacky.

>>In negotiations, anchoring refers to the concept of setting a boundary that outlines the basic constraints for a negotiation; subsequently, the anchoring effect is the phenomenon in which we set our estimation for the true value of the item at hand.[5] In addition to the initial research conducted by Tversky and Kahneman, multiple other studies have shown that anchoring can greatly influence the estimated value of an object. [0]

Do you or your Roman generals (who probably wouldn't be fighting each other anyway, but that's besides the point) have any hard data/sources you can show that demonstrate a link between anchoring and coming off worse in a negotiation?

You can hurt yourself by giving a number, but you're more likely to help yourself if you choose a realistic but slightly high number and indicate that you are willing to negotiate based on non-monetary factors.

[0] : https://en.wikipedia.org/wiki/Anchoring#In_negotiations

What you're saying would be true if companies were navigating the process as blindly as candidates are. Then a high anchor could lead the employer to negotiating from a position of weakness. But, armed with data, companies don't do that.

1 - With no asymmetry both peers would have a value anchored before negotiations start.

Two stupid things on their part...I got headcount, not a budget. So when negotiating salary, it wasn't in my best interest to haggle. I'd rather have happy relationships with my direct reports than know that I screwed them out of a few dollars on behalf of the company. Had they made it my duty to make the numbers add up, I might have been a bit more frugal. Oh, and they didn't filter the data to only show me positions below my title. So I went into my reviews armed with the exact market salary data that the company used to negotiate with me.

That's far too big of a generalization and I think it's easier to talk about specific scenarios. The parent comment alluded to a quite common counterexample:

1. salary range for position is 100k-150k

2. candidate is first to talk and names anywhere from $130k to $170k

3. candidate gets talked down a bit, e.g. 10%

It seems to me the candidate will still end up higher than if the company is first to talk and names $100k!

Also, even if a company can't hire me at the moment due to real budget constraints, they might be able to do so in the future, or perhaps one of the people I interviewed and impressed enough to get to the salary negotiation phase will remember me when they move to another company and want to hire me then.

If I name a number way out of their range first, though, they could reject me right off the bat, and I'd never get an interview in the first place, never get the practice, and never get a chance to impress anyone. All of these are worth more to me than the chance of a salary negotiation going sour because I wind up being too expensive. Besides, I'll always have the chance to decide whether I want to work for the company despite them being below my ideal salary. Salary isn't everything.

As for anchoring, that might work if you're negotiating with novices, but many employers have professional salary negotiators (either HR or recruiters themselves) who know about anchoring, and they're not going to get fooled by such psychological tricks. If I don't reveal a number first, however, they are forced to name one, and then the ball is in my court, and I can negotiate up from there as I see fit. There's really no downside.

This. I get the impression that most software devs don't make an effort to learn this incredibly valuable skill. It doesn't even have to be haggling, per se. Just asking for a higher salary is often going to increase it by thousands of dollars. It just boggles my mind at how so many people will leave so much money at the table...

General life rule that so many people seem to have zero clue about - you only get what you ask for. Instead, so many people turn into passive aggressive assholes or sit in silence and stew. If you want more money, ask for it (in a polite professional way obviously). If the company says no, you now have a clear next step.

As an employer, I do.

I once made the mistake of spending a lot of time interviewing a candidate assuming that their salary expectations would be around market rates...they weren't. This was years ago and it chaps me a bit to this day that I made such a rookie blunder. I won't make that mistake again. I ask for a range right off and give one in return. This saves a lot of time if expectations are way off.

It's like buying milk. I couldn't tell you exactly how much I could buy a gallon of milk for right now (without looking it up). But I could tell you that it is for around five or six dollars. I could also tell you that organic milk is a bit more and the store brand is a bit less. I could also tell you it is likely to be a bit more in a small, boutique store than a large chain store and cost a bit more in the middle of NYC or SF than it would cost in the outskirts of St. Louis.

In short, I know enough about the cost of milk that if it is priced at $16 a gallon or $1.50 a gallon that it is unusual and, despite not knowing the exact 'normal' cost of a gallon of milk, that I can assume I can walk away and get some somewhere else in the market for a 'normal' price.

I think most people are in my shoes when it comes to milk and I think most employers in tech have a pretty good idea what a normal salary for a position is, especially those who have written checks for those positions for a while.

If you are finding that is not the case, I would wonder why.

The "10X developer" may be a bit of overhyped mythology, but underneath it is the very real idea that developers vary enormously in their abilities and the amount of value they can add. While the variance in their salaries doesn't capture the variance in their productivity, perhaps, there are many situations where there is profound business sense in paying one developer 2-3x more than another.

The problems are: (1) Everyone thinks they're that 2-3x guy; (2) office politics would sooner or later lead to the news getting out and a profound sense of injustice spreading like a wildfire.

I've had employers think that $80k was a reasonable offer for me (when I was making $200k).

Otherwise, what I'm talking about is the fat part of the salary curve...there are always going to be interesting situations at the edge.

It's like mechanics: F1 mechanics make a hell of a lot more than your average wrench turner (for good reason). If you need someone to look at your Corolla you don't need that extra skill and you couldn't afford it even if you wanted it.

I've seen way too many companies where they have 3 or 4 developers doing work which 1 competent developer could do, but they're unable to hire or retain competent developers due to poor pay. They'd actually save a lot of money by paying better.

But I really like that you approach it this way - practical, sensible, and acknowledges that both parties' time has value.

That's comically naive. External recruiters routinely share candidate ranges with employers.

Believe me, if I told an external recruiter I was looking to make X and they sent me on an interview and I got a job offer at 0.5X, not only would that be the last time we ever spoke but wherever I eventually landed would not be using that recruiter or their firm ever again.

I can't think of a job I've been contacted about in the last 15+ years where I didn't know what the salary range was before I even agreed to have my resume submitted.

I won't object.

To take a silly but illustrative example: if I'm looking for a javascript developer, I don't care if the person I'm interviewing also has a law degree from a top tier school and can craft airtight patent applications even if I may have some use for that skill. I am hiring for a javascript developer. That is the role I need filled. I will offer a salary range based on that need.

It is a silly example but I see the issue so often: someone will apply for a role and then expect to be compensated for their other skills. While, sometimes, it may be worth it to do a bit of management gymnastics and get someone brilliant through the door and find out how to fully leverage their talents later, most of the time that's just wishful thinking on everyone's part.

How often does your eventual accepted offer land at the high end of your offered range? How often at the low end?

I ask, because it often seems unproductive to negotiate within-range on merits. If you're willing to pay up to X, I as a candidate am going to expect that amount. Anything less is letting you, the employer, capture the surplus...

I understand the concept of probabilistic valuation of risks/opportunities of a given candidate, and if that were practically quantifiable in the real world, I agree that it would represent a sound way to price an offer (provided you also factored in the cost of losing a candidate and having to put more resources towards finding an equivalent or better candidate)... [1]

Unfortunately, in practice, a candidate has to be evaluated more coarsely as "good enough to hire" or "not good enough to hire". There's no way you could in good conscience employ someone who couldn't do the work - even if they came at a discount. That means that there definitely is budget surplus any time you hire someone at less than your maximum budget. There is a fixed amount you're able to pay to get the work done, and that is at the max of your budget range.

Strategically, then, the max of the range offered represents the minimum a qualified candidate should target in a negotiation... Your mileage may vary, of course, with the skill of the negotiator.

[1] By the way, I highly recommend Algorithms to Live By for an interesting discussion around the use of such methods in practice.

Oh, if it was only so simple as 'can do the work' | 'cannot do the work'!

It is, of course, a very course and rough estimate but it is more like 'This person is likely to be great! They will not only do the work but surprise and delight me and pull our organization forward more solidly and quickly then we even hoped. We are lucky they happened to be looking just at the time we need their skills!' or 'This person looks like a good candidate but they tend to jump jobs on a yearly basis and has experience in C# rather than Java much less Scala but we've been looking for a while and we really need to get started and they seem smart enough to get up to speed.'

Point is, it's a complicated process with complicated factors.

Also, there is the wider market to consider. If I don't think a candidate can go out and get another job for the same price or more, I am unlikely to offer a premium on top of that market price.

So I have to disagree; skilled jobs are not binary in nature -- the max of the range offered represents how much a top tier candidate that fits well with the position should target. For those others without the same risk/reward profile or market power, they will have to keep from overselling themselves and take those factors into account.

If you like Algorithms to Live By, I think you'd enjoy 'How to Measure Anything' which really gets into the nitty gritty of how to reduce uncertainty and how even modest reductions can lead to much more solid decisions with these kinds of inherently fuzzy and complex problems.

If I tell you at the beginning of our conversation 'The salary for this position is between $80K and $120K depending on your experience, education, and demonstrable skill etc'...

Throughout the course of our interview it is revealed that your education is a good fit for the position and your experience, while decent and potentially transferrable, is not a great fit for the position. Also, the technical part of the interview was good in some areas and poor in others (though I suspect through mere ignorance and not raw ability) and you have no public code under your belt that I can examine. But, you seem to communicate well and get along with the team.

Well, you are a good candidate but represent a significant risk relative to some other candidates. I could hire you and it turn out that those skills are not as transferrable as I'd hoped. You could take significantly more training than I was expecting for the role which is a cost to me.

The 'surplus' goes toward the expected cost of the risk. It is possible that you could find a better fit employment wise and that I could find a better fit employee wise but that carries its own risk. We both weigh all of that call it good and are happy how it turns out. Or we don't and the search continues.

I wish more employers were willing to do that. Many that I've talked to want the candidate to state their salary expectation, but aren't willing to say what range they are looking to pay.

I also don't think salary is everything - far from it! Yet I've never had a problem saying "I believe my market value is $XYZk, however if the culture, team and product is a good fit for me then I'm sure we'll be able to find a salary level that works for both of us."

Normally I say this right off the bat and have not yet had bad feedback about it. I don't see anchoring as a psychological trick, rather an opportunity to save both you and your potential employer a lot of time.

This. I'm very upfront that if a company can't meet at least a particular dollar amount expected, it's not worth wasting both of our time. This saves a ton of time, and also serves as a nice red flag when a company agrees and then later tries to negotiate down at the last minute (if they're going to penny-pinch at the last minute, things like bonuses and vacation approval are basically never happening).

Let them see what you can do for them, and let them show you what it's like to work for them, then you should find it easier to come to an agreement on a number.

In one place that I worked at, salary for a common technical job title (n=100) varied by 50%. The mechanics were pretty obvious. People who worked their way up got screwed with the 10% raise bullshit, and people who came in through some other means were able to extract higher salaries.

Where I find this stuff annoying is where there is no signal about the salary level. One placed I spoke to many years ago was advertising for a position that was probably worth $100-105k in the market, but they were hiring at $50k.

If specifically asked "What is your current salary?" you would typically respond with the actual current salary, not your ideal new salary.

For instance, I think the worst bonus I've ever received with my current employer is 15%. Best was 30% or so.

That's not salary. But it's fairly reliable. With the new employer the bonus might not be reliable even if it's comparable percentage-wise. Every company has different criteria they use to fund the bonus pool and it's unheard of for a potential employer to ever disclose that. Everyone at my current employer gets paid out of the same bonus pool as the CEO so I know we're getting paid :-)

Also, equity (at least in my case) plays a huge role. The last few years I've made as much or more in selling vested stock options as I did in salary.

So if someone were to ask me what my current salary is in a employment negotiation I would tell them that I don't think it's relevant and give them a few examples why. I would however, tell them what I thought I was worth and justify it by explaining what value I could bring AND-OR I would tell them what salary range I've already been offered elsewhere recently and ask if they were prepared to make an offer in that range. If they insisted my current salary, I pretty much know that's not the right place for me.

BTW, if specifically asked 'what is your current salary?' I would typically respond with 'my ideal new salary is $X'

They're basically using it as a means to sanity check and calibrate their own assessment.

My next job paid me $90,000 CAD a year. (Investment bank)

My next job paid me $160,000 USD a year. (Google)

All numbers are all-in comp. All junior dev roles.

Should the bank, and Google have offered me $33,000 and $40,000, so that the new salary isn't too much higher then the old one? Would that satisfy your notion of fairness? When new grads hired by the latter are getting paid $150,000/year?

You are hired to do your new job, not the old one. What the old one paid has no relevance.

In other words: If you don't know what you're worth, of course they won't give you what you're worth.

People at $400K a year don't apply for a job, they're recruited. Any of the problems mentioned in this thread would likely not apply.

From where does this expectation derive, and what makes it a good barometer of sanity? On the face it just sounds like the same rationale used for basing raises on percentage of an employee's initial salary. "3% over your last job" is not fair.

Based on experience, almost nobody negotiates their offers, and if you see people talking about their job moves, people are often excited about, say, a 10% increase.

That's why employers expect to be able to get away with anchoring it to your current salary.

In this case, however, the disadvantage comes from a fixed number - your previous salary - not what you'd like.

In a situation where you were underpaid, declaring your previous salary will definitely put you at a negotiating disadvantage.

You're negotiating from a position of weaker information. You have no idea what the other engineers are being paid, all you know is how much you're being paid, and what you would like to be paid. Meanwhile the recruiter/hiring manager knows exactly how much they are paying all the other engineers.

You are now tasked with trying to get as fair/high a salary as possible, without knowing what their threshold is for simply no longer continuing the negotiations, because as you said "the price is just way too high".

By hearing the other side first, you can then gauge whether or not your desired salary is acceptable, or you can try to negotiate up.

This is absolutely 100% wrong. In price negotiations you will almost always fair better if you force the other party to bring up a number first.

There is a large body of research which disagrees with you...

Oechssler, Jörg; Roider, Andreas; Schmitz, Patrick W. (2009). "Cognitive abilities and behavioral biases". Journal of Economic Behavior & Organization. 72 (1): 147–152. doi:10.1016/j.jebo.2009.04.018. Jump up ^ Orr, D. & Guthrie, C. (2005). "Anchoring, information, expertise, and negotiation: New insights from meta-analysis". Ohio St. J. Disp. Resol., 597, 21. SSRN 900152 Jump up ^ Kristensen, Henrik; Gärling, Tommy (1997). "The Effects of Anchor Points and Reference Points on Negotiation Process and Outcome". Organizational Behavior and Human Decision Processes. 71 (1): 85–94. doi:10.1006/obhd.1997.2713.

(edit as I can't reply to the child comment: Yes, you're right of course. Put it down to a mix of multitasking, iphone browser and university-instilled fear of using Wikipedia as a source)

Why not link to the Wiki article you copied the sources from? https://en.wikipedia.org/wiki/Anchoring

There is no one size fits all to negotiations, there are simply too many. I would be hesitant to even apply some kind of 80/20 rule here. There are many kinds of negotiations and not all follow the same pattern, even if all of them in one person's experience have.

I have succeeded when demanding they present a number and when presenting a number. I think there are other important things in negotiations like how numbers are presented and what else is being exchanged, but I do not have good sources for this last bit outside my own experience.

You can do this if you have a lot of leverage. But if you don't -- and most job seekers don't -- then giving a number first will put you at a disadvantage because it reveals how much you think you are worth and simply gives the other side a starting point from which to bring down the price.