
What's in this deep dive
- What dollar-cost averaging actually is
- The mechanism in one sentence
- Why your average cost is not the average price
- A two-purchase proof you can check by hand
- The harmonic mean, and what it guarantees
- A six-month worked example
- Where the shares actually came from
- When lump sum investing wins
- When dollar-cost averaging wins
- The flat market case is a tie
- The falling market case: a smaller loss is still a loss
- What the method does not do
- Dollar-cost averaging versus waiting for a dip
- Automatic contributions are dollar-cost averaging
- How often should you buy
- Fees, minimums and fractional shares
- Averaging into a retirement account
- Value averaging and other cousins
- Common mistakes
- Setting up a plan you will not abandon
- How it fits a long-term plan
- The bottom line
Dollar-cost averaging is the practice of investing a fixed dollar amount on a fixed schedule, whatever the price is that day. That is the entire definition, and it sounds too plain to deserve a name. The interesting part is what the arithmetic does with it: because your dollars are fixed and the price is not, every purchase buys a different quantity, and those quantities are largest exactly when the price is lowest. Run that for a while and your average cost per share lands below the simple average of the prices you paid across, not sometimes but always, as a matter of algebra rather than luck.
This breakdown works the mechanism out in full. It covers what the term actually means, the two-purchase calculation you can check by hand, why the result is a harmonic mean and what that guarantees, a six-month worked example carried through end to end, and the four market paths that decide whether averaging in beats putting the money in at once, including the honest case where lump sum wins by a wide margin. It also covers scheduling, fees and fractional shares, the behavioural argument, and what averaging cannot do. It sits alongside our walkthroughs on how to start investing for beginners, how much to invest in S&P 500 index funds, and simple interest versus compound interest, and the companion below runs your own figures through the same formulas as you read. Every dollar amount here is an invented round number chosen to make the arithmetic legible, not a forecast and not a description of any real investment.
Key takeaways
- Dollar-cost averaging means a fixed dollar amount on a fixed schedule, so a falling price automatically buys more shares and a rising price buys fewer.
- Your average cost is total dollars divided by total shares, never the average of the prices: an illustrative $600 at $50 plus $600 at $30 gives 32 shares at $37.50, against a $40 price average.
- That gap is guaranteed by the harmonic mean, which is always at or below the arithmetic mean of the same prices, with equality only when every price is identical.
- Lump sum investing wins when prices rise steadily from day one, and averaging wins when prices dip and recover. Nobody knows which path is coming.
- Averaging in reduces the size of the swing, not the possibility of loss. Every figure here is illustrative arithmetic, general education rather than personalized advice.
What dollar-cost averaging actually is
Dollar-cost averaging is a purchasing rule, not an investment. It says: choose an amount, choose an interval, and buy that amount of the same thing every interval without consulting the price. Two hundred dollars every Friday is dollar-cost averaging. Six hundred dollars on the first of the month is dollar-cost averaging. Whatever is left over at the end of the month, invested whenever you remember, is not, because the amount and the timing both drift with your mood, which reintroduces exactly the judgment the rule exists to remove.
The rule is defined by what it holds constant. The dollars are constant; the shares are the variable that absorbs every price movement. This is the inversion that people find counterintuitive at first, because most purchasing habits work the other way around. When you buy a set number of shares each month, the dollars vary and the mechanism disappears entirely. It is the fixed dollar amount, and only the fixed dollar amount, that produces the effect the rest of this breakdown describes.
Most people already practise it without naming it. A payroll deduction into a workplace retirement plan takes the same percentage of the same salary every pay period and buys whatever that money buys at that moment. An automatic transfer into a brokerage account on payday does the same. The reason the term feels technical is that it is usually introduced as a decision, when in the majority of cases it is simply the default shape of investing out of income rather than out of a windfall.
The mechanism in one sentence
Fixed dollars buy variable shares, and the variation runs opposite to price. Six hundred dollars at $50 per share is 12 shares. The same $600 at $30 per share is 20 shares. You did nothing differently on either date; the price did the work of deciding how much you accumulated. A cheaper month is automatically a bigger month, measured in the only unit that matters at the end, which is how many shares you own.
That sentence is the whole engine, and everything else in this breakdown is a consequence of it. It has a useful corollary that is worth stating plainly: dollar-cost averaging quietly enforces a mild version of buying more when things are on sale, without requiring you to decide that anything is on sale. The decision is delegated to arithmetic. There is no forecast in it, no judgment about whether a decline is finished, and no moment at which you have to be right.
The corollary also explains why the effect vanishes in a flat market. If the price never moves, every purchase buys the same quantity, and the averaging has nothing to average. The mechanism needs volatility to produce any distinction at all, which is the opposite of how it is usually pitched. Price movement is not the enemy of a steady buyer; it is the raw material the method feeds on.
Why your average cost is not the average price
The most common error in this topic is quiet and costs people nothing except an incorrect belief, which is bad enough on a topic where beliefs compound. It is averaging the prices. If you bought at $50 and at $30, the intuitive answer for your cost per share is $40, because that is the midpoint. It is wrong, and the reason is that you did not buy equal quantities at those two prices. You bought 20 shares at $30 and only 12 at $50, so the cheap price carries more weight in your position than the expensive one.
The correct calculation has one form and no variants: total dollars invested, divided by total shares acquired. Nothing about the schedule, the number of purchases, or the price path changes that. Add up what went in, add up what you got, divide. Every other formula you might see is either this one rearranged or a mistake.
That definition is also what makes the result robust. It does not care whether your purchases were monthly or irregular, whether some were larger than others, or whether the price path was smooth. It handles reinvested distributions the same way, treating them as additional dollars that bought additional shares. Our note on dividend reinvestment plans covers that case in its own right, but the cost arithmetic there is identical to the arithmetic here.
A two-purchase proof you can check by hand
Take the smallest example that shows the effect, because it is easier to trust something you can verify on paper. You invest $600 when the price is $50, and another $600 when the price is $30. Both amounts are illustrative round numbers.
The first purchase gives you $600 divided by $50, which is 12 shares. The second gives you $600 divided by $30, which is 20 shares. Your total outlay is $1,200 and your total holding is 32 shares. Your average cost per share is $1,200 divided by 32, which is $37.50 exactly.
Now compare that to the simple average of the two prices, which is $50 plus $30 divided by two, or $40.00. Your actual cost is $2.50 below it, which is 6.25 percent below. You did not time anything, predict anything, or choose anything beyond the fixed $600. The gap appeared because the $30 purchase bought 20 of your 32 shares, roughly 62.5 percent of the position, while the $50 purchase bought the other 12.
Push the numbers to an extreme and the intuition sharpens. If the second purchase had happened at $10 instead of $30, you would own 12 plus 60, or 72 shares, for the same $1,200, an average cost of $16.67 against a price average of $30. The cheaper the low price gets, the more shares it drags into the total, and the harder it pulls the cost average toward itself. Enter your own two prices in the companion calculator on this page and watch the two averages separate.
The harmonic mean, and what it guarantees
The formal name for what you just calculated is the harmonic mean of the prices, weighted by the equal dollar amounts. The harmonic mean of a set of numbers is the count divided by the sum of their reciprocals. For $50 and $30, that is 2 divided by one fiftieth plus one thirtieth, which works out to exactly $37.50, matching the calculation above. It is the correct average whenever you are averaging rates or prices under a fixed numerator, which is precisely the dollar-cost averaging situation.
The useful property is an inequality that holds for any set of positive numbers: the harmonic mean is always less than or equal to the arithmetic mean, and the two are equal only when every number in the set is identical. Translated back into money, this means your average cost per share can never exceed the simple average of the prices you bought at, and will sit strictly below it whenever those prices differ at all. There is no market condition, no sequence, and no length of time that breaks it, because it is not a claim about markets.
It is worth being precise about what that guarantee is not. It is not a guarantee of profit, or of beating any alternative, or of a better outcome than investing all at once. It says only that one particular average sits below another particular average. Whether you made money depends entirely on where the price finished relative to your $37.50, and the arithmetic has nothing to say about that. Confusing a statement about averages with a statement about returns is the single biggest overclaim made on this topic, and it is usually made by people who like the conclusion.
The wider the spread of prices, the wider the gap between the two means. A calm stretch produces a cost that is barely below the price average. A violent one produces a noticeably lower cost, and also a noticeably more uncomfortable statement of account along the way. The two effects are the same effect.
A six-month worked example
Scale the two-purchase case up to something that looks more like a real year of contributions. You invest an illustrative $600 on the first of each month for six months, into an investment whose price does what prices do. The prices, all illustrative, are $50, $40, $30, $40, $60, and $60.
Month one: $600 at $50 buys 12 shares. Month two: $600 at $40 buys 15 shares. Month three: $600 at $30 buys 20 shares. Month four: $600 at $40 buys 15 again. Month five: $600 at $60 buys 10 shares. Month six: $600 at $60 buys 10 more. Add the shares and you hold 12 plus 15 plus 20 plus 15 plus 10 plus 10, which is 82 shares. Add the dollars and you invested $3,600.
Your average cost is $3,600 divided by 82 shares, which is $43.90 per share when rounded to the cent. The simple average of the six prices is $280 divided by 6, or $46.67. Your cost sits $2.77 below the price average, about 5.9 percent, for the same reason as before: the $30 month alone delivered 20 shares, more than either $60 month managed at a third of the quantity.
At the month-six price of $60, your 82 shares are worth $4,920 against $3,600 invested, an illustrative gain of $1,320, or about 36.7 percent on the dollars you put in. The next section compares that to what a lump sum would have done on the same price path, which is where the honest part of this topic begins.
Where the shares actually came from
The chart below counts the shares each of the six monthly purchases bought. Every bar is the same $600; only the price changed. The bar widths scale to the largest month, which is the $30 purchase at 20 shares.
Illustrative shares bought by an identical $600 each month
Six equal purchases of $600 at illustrative prices of $50, $40, $30, $40, $60 and $60. Bar width scales to the largest month. Not a forecast or any specific investment.
Illustrative only. The six purchases total $3,600 and 82 shares, an average cost of $43.90 against a $46.67 simple average of the prices.
Read the shape rather than the numbers and the method explains itself. The tallest bar is the cheapest month and the shortest bars are the most expensive ones, and you made no decision that produced that pattern. The worst month to be a holder was the best month to be a buyer, which is the sentence that makes dollar-cost averaging psychologically survivable when a balance is falling.
The same data expressed as shares of the final position makes the weighting explicit. The $30 month contributed 20 of the 82 shares you ended up with, or 24.4 percent of everything you own, from one sixth of the money. Each $60 month contributed 10 shares, or 12.2 percent. That two-to-one difference in influence is the harmonic mean doing its work, and it is why averaging the prices gives the wrong answer.
Illustrative share of the final position bought by each month
Each month spent the identical $600, but contributed a different slice of the 82 shares held at the end. Segments sum to 100. Illustrative arithmetic only.
Illustrative only. Equal money, unequal ownership: the cheapest month bought twice the slice of the most expensive ones, which is exactly why the cost average sits below the price average.
When lump sum investing wins
Here is the case that most explanations of dollar-cost averaging skip, and skipping it is a form of dishonesty. Take the same $3,600 and the same six monthly purchases, but put them into a price path that rises steadily: $50, $55, $60, $65, $70, and $75, all illustrative.
The monthly purchases buy 12 shares, then about 10.91, then 10, then about 9.23, then about 8.57, then 8. The total is roughly 58.71 shares for the full $3,600, an average cost of about $61.32. The simple average of those six prices is $62.50, so the harmonic mean guarantee still held: your cost is below the price average, exactly as promised. At the final price of $75, your position is worth about $4,403, an illustrative gain of around $803.
Now run the alternative. The same $3,600 invested entirely in month one at $50 buys 72 shares, and 72 shares at the final price of $75 is $5,400, an illustrative gain of $1,800. The lump sum finished roughly $997 ahead, more than twice the gain, on identical money and an identical price path. The averaging did nothing wrong; it simply left most of the money uninvested through a rise, and cash that is not invested does not participate.
This is the general result, and it is worth stating without hedging: when prices trend upward from the day you start, investing everything at the start wins, and the more steadily they rise the more decisively it wins. Since upward drift over long horizons is the reason people invest in the first place, the base case tilts toward lump sum on expected value. Anyone who tells you averaging in is mathematically superior has either not run this path or has decided not to mention it.
When dollar-cost averaging wins
The mirror image is equally clean. Return to the first six-month path, the one that fell from $50 to $30 and recovered to $60. The steady buyer finished with 82 shares worth $4,920 on $3,600 invested, a gain of $1,320.
The lump sum buyer in that path put the full $3,600 in at $50 and bought 72 shares. At the closing price of $60, that position is worth $4,320, a gain of $720. The steady buyer finished $600 ahead, having owned less money in the market for most of the period. The advantage came entirely from the three months in the middle when the price was below the starting level and each $600 bought unusually well.
The pattern generalizes: averaging in wins when the price spends time below your starting price and then recovers, because that dip is where the extra shares are manufactured. It loses when the price mostly sits above where it began. Neither outcome is a property of the method. Both are properties of the path, and the path is not knowable in advance, which is the entire reason this comparison never resolves into a rule.
There is one asymmetry worth naming honestly. The lump sum advantage in a rising market is a matter of expected value, while the averaging advantage in a falling one is a matter of damage control, and those two are not experienced the same way. A smaller number on a statement during a decline changes behaviour in ways that a slightly larger number during a rise does not, which is why the argument for averaging is usually made on behavioural grounds rather than mathematical ones.
The flat market case is a tie
For completeness, run the degenerate path. If the price is $50 in all six months, each $600 buys exactly 12 shares, six purchases give 72 shares, and the average cost is $3,600 divided by 72, or exactly $50.00. The simple average of the prices is also exactly $50.00, and the two means coincide, which is the equality case of the harmonic mean inequality.
The lump sum buyer also ends with 72 shares, because $3,600 at $50 is 72 shares whether it happens on one day or six. Both finish with identical positions worth identical amounts. There is nothing to choose between the two approaches, because there was no volatility for either to interact with.
This case is instructive precisely because it is boring. It shows that the averaging benefit is entirely a volatility benefit, and that in the absence of price movement the method reduces to a scheduling preference. It also quietly reveals why the effect is often overstated in calm periods and then discovered dramatically after a turbulent one.
The falling market case: a smaller loss is still a loss
The fourth path is the one that gets left out of promotional material, and it deserves the same arithmetic as the others. Prices fall through the whole six months: $50, $45, $40, $35, $30, and $25, all illustrative.
Each $600 buys 12 shares, then about 13.33, then 15, then about 17.14, then 20, then 24. That is roughly 101.48 shares for $3,600, an average cost of about $35.48 against a simple price average of $37.50. At the closing price of $25, the position is worth about $2,537, an illustrative loss of roughly $1,063, or about 29.5 percent of the dollars invested.
The lump sum buyer at $50 owns 72 shares worth $1,800 at the end, a loss of $1,800, or 50 percent. Averaging in cut the loss by roughly $737 in this path. It did not prevent the loss, reduce the risk of the underlying investment, or provide any protection worth the name. It spread the entry across a decline, which mechanically produced a lower average entry price than the first day of that decline.
Sitting with all four paths together gives the honest summary. Averaging in produced the better outcome in two of them, the worse outcome in one, and a tie in one, and which path arrives is not something the method controls. What the method reliably controls is the size of the mistake available to you on any single day, and for many people that is the more useful property.
What the method does not do
A short list of things dollar-cost averaging cannot do, because the overclaims in this area are persistent. It does not raise the expected return of what you buy. The expected return belongs to the investment, and a purchasing schedule cannot improve it. If anything, holding money in cash while you phase in reduces the average time your money spends invested, which works against you when the underlying drifts upward over long periods.
It does not protect against a permanent decline. If the price never recovers, buying more of it on the way down produces more shares of something worth less, and the lower average cost is cold comfort. The method assumes nothing about the quality of what you are buying, which is why it pairs badly with concentrated positions and comfortably with broad diversified holdings. Our note on index funds compared with ETFs covers the vehicle side of that decision.
It does not time anything. There is no version of dollar-cost averaging that identifies bottoms, and treating a scheduled purchase as a prediction misreads it entirely. The rule is valuable because it removes the forecast, not because it makes a good one.
Finally, it does not eliminate the emotional difficulty of investing during a decline. It reframes it, sometimes helpfully, by giving the falling price a job to do. But the balance still falls, and continuing to send money into an account that is shrinking remains the hardest part of the whole practice.
Dollar-cost averaging versus waiting for a dip
A tempting alternative sits between the two approaches: hold the money in cash and deploy it when the price drops meaningfully. This sounds like averaging with better aim, and it fails for a reason worth spelling out. It requires two correct judgments rather than none: how large a decline qualifies, and whether the decline you are looking at is the one you were waiting for or the beginning of a larger one.
The failure mode is not usually buying at the wrong moment. It is never buying at all. A threshold that has not been reached produces no action, and a threshold that has been badly overshot produces a sense that waiting a little longer is now obviously correct. Cash that is waiting for a signal frequently waits through the entire period you were saving for, and does so while feeling prudent.
The scheduled version has no threshold to be wrong about. It converts an open question into a recurring transfer, and it is fully specified in advance, which means it can be automated and then ignored. That is not a claim of superiority over a perfectly executed dip-buying plan. It is an observation that perfectly executed plans requiring repeated judgment calls are rare, and the comparison should be against what people actually do rather than what they intend.
Automatic contributions are dollar-cost averaging
The most reliable way to practise this is to stop practising it and let a transfer do it. Most brokerage platforms and retirement plans support a recurring contribution of a fixed dollar amount on a fixed date, and once that is set the method runs whether or not you are paying attention, which is its main advantage over an intention.
Automation changes the failure mode in a specific way. A manual plan fails when the price is frightening, which is exactly when the purchases matter most, because the decision to skip a month is made under the same emotions that make the month cheap. An automatic plan fails only when you actively intervene, which raises the bar from inertia to a deliberate act. That asymmetry is worth more than any refinement of the schedule.
Two practical cautions apply. First, make sure the contribution actually invests rather than merely transferring cash into the account, because plenty of people discover months of deposits sitting uninvested in a settlement balance. Second, revisit the amount when your income changes, since a figure set years ago quietly becomes a smaller share of what you earn. Our walkthrough on opening a brokerage account covers where those settings usually live.
How often should you buy
Frequency matters much less than most people assume, and the reason is that weekly, twice-monthly and monthly purchases of the same annual total are all sampling the same price path, just at different resolutions. The averages they produce land close together over any meaningful stretch, and the differences between them are noise rather than signal.
The practical answer is therefore to match the schedule to your income. If you are paid twice a month, buy twice a month, because that is when the money exists and the transfer requires no separate act of saving. Money that sits in a checking account waiting for the first of the month is money exposed to being spent, and that risk is considerably more real than any difference in average cost between two schedules.
Where frequency genuinely matters is friction. If your platform charges a commission per trade, splitting the same money across four purchases instead of one multiplies that cost, and on small contributions the fee can swamp any averaging effect. If there is a minimum purchase amount, a schedule that is too frequent may simply fail to execute. Check both before choosing an interval, and prefer the least frequent schedule that still keeps the money out of reach of your spending.
One further note on timing within the period: choosing a specific day of the month because it seems statistically favourable is a forecast wearing a schedule’s clothing. Pick the day after you are paid and stop optimizing it.
Fees, minimums and fractional shares
The clean arithmetic above assumes your $600 buys exactly $600 worth. Three real-world details bend that, none of them fatal, all of them worth knowing before you are surprised.
Commissions, where they still exist, come off the top. A per-trade fee raises your true average cost by the fee divided by the shares bought, which hits hardest on small purchases and on expensive shares. The effect is straightforward to check: add the fees to your total dollars invested before dividing by shares, and the resulting number is your honest cost per share.
Whole-share-only platforms create a remainder. If shares cost $50 and you send $600, you buy 12 and nothing is left over, but at $55 you buy 10 shares for $550 and $50 sits in cash. That cash is not invested, so your effective average cost drifts slightly from the formula and, more importantly, some of your money is not working. Fractional-share support removes this entirely by letting the full amount buy a fractional quantity, which is why the rising-market example above shows share counts like 10.91.
Ongoing fund costs are a separate drag that no purchasing schedule affects. They come out of the investment’s returns year after year regardless of how you bought in, and over long horizons they matter far more than any averaging effect. Our breakdown of what an expense ratio is works through that arithmetic, and it is the one number in this whole area you can control with certainty.
Averaging into a retirement account
Inside a workplace retirement plan, dollar-cost averaging is not a choice you make but the structure you are handed. Contributions arrive with each pay period, are invested according to your election, and buy whatever the price is that day. The mechanism is running whether or not anyone involved has heard the term.
Two features of that setting deserve mention. Employer matching contributions, where offered, arrive on the same schedule and are commonly described as the highest-certainty return available in a retirement plan, which is a separate matter from averaging but interacts with it, since both reward consistency. And the tax treatment inside such accounts means the cost basis arithmetic above generally has no tax consequence, because you are not tracking gains per lot for a taxable sale in the same way.
In a taxable brokerage account the opposite is true: every purchase creates a separate tax lot with its own cost and date, and a long dollar-cost averaging habit produces a great many of them. That is not a problem, and brokers track it, but it is worth understanding when you eventually sell, and it interacts with rules on realizing losses. Our coverage of portfolio management touches on how those records fit the broader picture, and a tax professional is the right person to ask about your own situation.
Value averaging and other cousins
A few variations on the theme circulate, and it helps to know how they differ. Value averaging sets a target portfolio value for each date rather than a fixed contribution, then invests whatever amount is needed to reach it, which means contributing more after a decline and less, or even selling, after a rise. In theory it sharpens the buy-low effect. In practice it demands a variable and sometimes large contribution exactly when markets are falling, which is when spare money is least likely to be available, and it can generate sales and tax events.
Share averaging, buying a fixed number of shares each period, is the anti-method described earlier: the dollars vary with price and the harmonic mean effect disappears completely, since you are now buying equal quantities at every price. It is not wrong, but it has none of the properties this breakdown has described.
Percentage-of-income contributing is the variation most worth considering, because it keeps the discipline while letting the amount grow with earnings. It is still dollar-cost averaging in every period; the fixed amount simply resets when your income does. For most people saving out of a salary, that is the version that survives contact with real life over decades, and it pairs naturally with a target you can actually check. Our breakdown on how much you need to retire sets that target, and the retirement number calculator on our homepage puts a figure on it.
Common mistakes
The first mistake is the one this breakdown opened with: averaging the prices instead of dividing dollars by shares. It produces a cost estimate that is always too high, which makes people believe they are further behind than they are.
The second is stopping during declines. This is the most expensive error available in the whole method, because it removes precisely the purchases that generate the low average cost. A plan abandoned in month three of the six-month dip example would have missed the 20-share month, the single largest contributor to the final position.
The third is confusing a lower average cost with a gain. Your cost sits below the price average in every fluctuating path, including the one where you lost 30 percent. The two facts are unrelated, and a marketing line that mentions the first without the second is incomplete.
The fourth is applying the method to something you would not want to own more of. Averaging into a declining position is only sensible if the decline is price movement rather than deterioration, and the method contains no way to tell the difference. It presumes a judgment about what you are buying that must be made elsewhere.
The fifth is over-engineering the schedule: switching intervals, moving purchase dates, or pausing to wait for clarity. Every one of those reintroduces the forecast the rule was designed to eliminate, and the gains available from perfect scheduling are far smaller than the losses available from an interrupted habit.
Setting up a plan you will not abandon
The construction is short. Decide the amount from your budget rather than from a view about prices, and choose a figure you can sustain through a bad year, since a smaller amount that survives is worth more than a larger one that stops. Choose the date to fall just after you are paid. Set the transfer and the investment instruction together so the money does not stall as cash. Then leave it.
The only scheduled maintenance is an annual look: raise the contribution if your income rose, confirm the purchases are executing and investing, and check that what you are buying still matches the allocation you intended. That last point is where rebalancing enters, and our walkthrough on how to rebalance your portfolio covers it in detail. None of this requires an opinion about where prices are going, which is the point.
If you are starting from nothing, the sequence in our beginner walkthrough covers the account and allocation decisions that come before the schedule. If you are starting from a lump of cash, the honest framing is the one from earlier in this breakdown: the arithmetic favours investing it, and phasing it in is a way of buying comfort at some expected cost. Deciding how much that comfort is worth is a personal question, and a fee-only financial professional is better placed to help with it than any article.
How it fits a long-term plan
Zoom out far enough and dollar-cost averaging stops looking like a strategy and starts looking like plumbing. It is the mechanism by which income becomes invested capital, running quietly underneath every decision that actually determines outcomes: how much you save, what you buy, what it costs to hold, and how long you leave it alone.
Those four levers do the heavy lifting, and the averaging effect is small beside them. Raising a monthly contribution meaningfully changes a thirty-year result. Holding on through a decline changes it. A lower ongoing cost changes it. Whether your average cost per share was $43.90 or $46.67 in one six-month window is a rounding detail by comparison, and treating the averaging as the strategy rather than the delivery system inverts their importance.
What the method is genuinely excellent at is making the other four possible. It converts saving from a decision into a default, it removes the entry-timing question that keeps people in cash, and it survives inattention, which no plan requiring judgment does. Compounding needs time above all else, as our comparison of simple and compound interest works through, and the main practical obstacle to time is waiting to begin. A scheduled amount is the cheapest available cure for that.
The bottom line
Dollar-cost averaging means putting a fixed dollar amount in on a fixed schedule, so that a falling price automatically buys more shares and a rising price buys fewer. Your average cost is total dollars divided by total shares, never the average of the prices, and because a fixed sum buys a quantity that moves inversely with price, that cost is a harmonic mean and therefore always sits at or below the simple price average. An illustrative $600 at $50 and $600 at $30 gives 32 shares at $37.50 against a $40 price average, and the six-month example gives 82 shares at $43.90 against $46.67. Those gaps are algebra, not skill.
What the arithmetic will not do is tell you it beats the alternative. On a steadily rising path the same $3,600 invested at once finished roughly $997 ahead in the illustrative example above, and on a steadily falling one the steady buyer still lost about 30 percent of what went in. Averaging shrinks the swing around whatever the investment does; it does not change what the investment does. The strongest honest case for it is that it removes the timing decision, runs without supervision, and keeps money flowing in during the stretches when a judgment call would most likely stop it. Treat every figure here as illustrative arithmetic, put your own into the companion, and judge the schedule on whether you will still be running it in ten years rather than on which path it would have won.
Dividora writes for readers who want the working shown, and that is all this breakdown is: educational general information about how a purchasing formula behaves, not financial, tax, or investment advice, and not a recommendation to buy, hold, or sell anything. The prices, contribution amounts, share counts, and outcomes above are invented round numbers selected to make the arithmetic checkable, not observed results, not typical results, and not a forecast of any kind. Investments that fluctuate can and do lose value on any schedule, a lower average cost per share is not a gain, and platform fees, minimum purchase rules, fractional-share support, and tax treatment of individual lots all vary and change over time, so confirm the current terms that apply to your own accounts. Before committing to any contribution plan, put your actual timeline, savings, and circumstances in front of a qualified financial professional.
Frequently asked questions
What is dollar-cost averaging in simple terms?
Dollar-cost averaging means investing a fixed dollar amount on a fixed schedule, regardless of what the price happens to be that day. Because the dollar amount is fixed and the price is not, each purchase buys a different number of shares: more when the price is low, fewer when the price is high. That single asymmetry is the whole mechanism, and it is why your average cost per share ends up below the simple average of the prices you paid across. Most people already do this without naming it, because a monthly payroll contribution into a retirement plan is dollar-cost averaging by construction. Everything in this breakdown is illustrative arithmetic and general education, not advice about any particular investment.
How do you calculate the average cost per share?
Divide the total dollars you invested by the total shares you ended up with, and nothing else. If you put in an illustrative $600 when the price was $50 and another $600 when the price was $30, you bought 12 shares and then 20 shares, so $1,200 divided by 32 shares is $37.50 per share. Notice that the simple average of $50 and $30 is $40, so your actual cost sits $2.50 below it. Never average the prices themselves and call that your cost, because it ignores the fact that you bought more shares at the cheaper price. The companion on this page runs the same two-purchase calculation on any figures you enter.
Why is my average cost lower than the average price?
Because a fixed dollar amount buys a quantity that moves inversely with price, your cost per share is what mathematicians call a harmonic mean of the prices rather than an ordinary arithmetic mean. The harmonic mean is always less than or equal to the arithmetic mean of the same numbers, with equality only when every price is identical. In plain terms, the low-priced purchases contribute more shares to the total, so they pull the cost average down harder than the high-priced purchases pull it up. This is a property of the arithmetic itself, not a market prediction, and it holds whether prices rose, fell, or wandered. It also does not mean you made money, which is a separate question about where the price ended up.
Is dollar-cost averaging better than investing a lump sum?
Not automatically, and it is worth being direct about this. If prices rise steadily from the day you start, a lump sum invested on day one owns more shares for longer and finishes ahead, because the money you held back missed the rise. Dollar-cost averaging finishes ahead when prices fall and later recover, since the middle purchases collected shares cheaply. Nobody knows in advance which path is coming, so the choice is usually made on other grounds: whether the money exists as a lump at all, and how much regret risk you can carry if you invest everything the day before a decline. Both worked examples appear later in this breakdown, including the case where lump sum wins clearly.
Does dollar-cost averaging protect me from losing money?
No, and any explanation that implies otherwise is selling comfort rather than arithmetic. If the price of what you own is lower at the end than your average cost, you have a loss, however carefully you spaced the purchases. What averaging in changes is the size of the swing, not its direction: in an illustrative six-month decline from $50 to $25, a steady buyer might average around $35 per share and finish down roughly 30 percent, while someone who put everything in at $50 would be down 50 percent. Both lost money. Averaging reduced the damage in that path and would have reduced the gain in a rising one.
How often should I invest when dollar-cost averaging?
The schedule matters far less than most people expect, and the practical answer is usually whenever you get paid, because that is when the money exists and the decision requires no willpower. Weekly, twice monthly, and monthly purchases of the same annual total produce averages that land close together, since they are sampling the same price path at different resolutions. Where frequency does matter is friction: if your platform charges per trade or enforces a minimum purchase, more frequent buying can cost more than it gains. Match the rhythm to your income and your platform's rules, then leave it alone. Consistency over years does more work than any tuning of the interval.
Does dollar-cost averaging work with index funds and ETFs?
The arithmetic is indifferent to what you are buying, so it applies to any investment with a fluctuating price, and broad funds are simply the most common vehicle for it. The practical differences are mechanical rather than mathematical: some platforms let you schedule automatic purchases of funds in exact dollar amounts and support fractional shares, while others only fill whole shares, which leaves a small cash remainder each time. Whole-share-only buying makes your real average cost drift slightly from the clean formula, though rarely by enough to change any decision. Our separate coverage of fund structures explains where those mechanics differ. Nothing here is a recommendation to buy any specific fund.
Is dollar-cost averaging a good strategy for beginners?
It is commonly described that way for reasons that are behavioural more than mathematical, and that distinction is worth keeping straight. The averaging itself does not raise expected returns; what it does is remove the need to pick a moment, which is the decision that most often keeps new investors sitting in cash for years while they wait for clarity that never arrives. A fixed automatic amount converts an emotionally loaded judgment into a scheduled transfer, and money that is actually invested beats a perfect plan that never starts. Whether it suits your situation depends on your timeline, your emergency savings, and your tolerance for seeing a balance fall, which is a conversation for a qualified financial professional rather than an article.
Find a fiduciary financial advisor
Tell us about your portfolio and what you want it to do. We will connect you with fiduciary advisors who work in your interest.