The Product Life Cycle: Launch to Withdrawal
A product life cycle is a reading of one product's own volume and what each unit of it keeps, taken over time. A life cycle counts nothing else: not how many sellers the field carries, not how many buyers have taken the thing up. The reading runs backwards, from the two lines to the stage, and never forwards from a stage name to a figure nobody has seen.
Four curves in these notes look the same drawn. What separates them?
A definition would let the shape be filed under vocabulary, and attention would stop there. The shape is already familiar. A rise, a flat top and a fall can be drawn in ten seconds and recognised in one, and recognition feels almost exactly like understanding. The shape is the one part of a life cycle that carries no information, so recognition is the difficulty.
A drawn curve carries no information whatsoever until two things are stated: what is being counted along the bottom, and what is being counted up the side. Four curves share that same outline, and each of them counts something different. Four shapes that look identical drawn are about four different things, and the only way to tell them apart is to read what is being counted along the bottom. Here they are, in the order in which each is named wherever one of them is drawn.
The first counts how many sellers a field carries as time passes. The second counts how many buyers have taken a thing up. The third counts one product's own volumeThe count of units of one product actually sold in a period. A count of things, never the money those things brought in. and what each unit of it keeps, and the third is the one worked through below. The fourth counts how far performance moves for the effort spent on it, and its bottom axis is not even time.
Consider it for a moment. One rising and falling line could be the number of food stalls outside a single office building, or the number of households in a street who now carry a particular card, or one stall's own sales of one dish, or how much faster a kitchen got for each extra cook it hired. One picture, four questions, and no two of them share an answer. A curve drawn without labelled axes has taught the reader nothing at all, and has quietly invited four subjects to be merged into one.
Four curves in these notes are drawn with the same rise, the same flat top and the same fall. What separates them?
So what does this one count?
Two lines, and the reading needs both of them. The first line is the product's own volume: the count of units actually sold, period by period. The second line is what each unit of it keeps. The trade name for it is the contribution a unitWhat one unit leaves behind after everything that grows and shrinks with output has been paid for. A per unit figure, not a total..
Take either one away and the reading breaks in a specific way. Volume on its own cannot separate a product selling more at a thinner margin from a product selling more at the same margin, and those two situations lead to opposite decisions. The per unit line, on its own, cannot separate a product holding its margin because it deserves to from one holding its margin while fewer and fewer people buy it. Each line reports its own good news and stays silent about the other.
A product life cycle reading is two movements set side by side, and a single revenue line has already multiplied them together. The trap sits in the most commonly available figure of all. Revenue is volume times price, so a revenue line is the two movements folded into one number before the reading even starts. A rising revenue line shows that the product is fetching more money. A rising line does not show whether the money came from more units or from more money on each unit, and those two questions are answered by different people in different rooms.
Here is the concrete version. A stall sells twice as many plates as it did, at half the money left on each plate. Takings are unchanged. The cook is working twice as hard, the gas bill has doubled, and the takings line reports a perfectly flat year. Anybody reading only that line will conclude that nothing happened.
Theodore Levitt set out the product life cycle in the Harvard Business Review in 1965, and he is named for the frame. The shape of the curve is his, and the frame itself supplies no figures.
A product life cycle reading needs two lines. Which two?
What are the stages, and what is each one a summary of?
Four words do most of the work in most conversations, and they take about a paragraph to learn. A product arrives and volume is small. Volume climbs. Volume flattens. Volume falls. Trade vocabulary calls the four situations introduction, growth, maturity and decline, and there is nothing wrong with any of them as shorthand.
The four words matter less than what each of them quietly summarises about the second line. Early on, what each unit keeps is unsettled: the realised priceWhat a unit actually fetched on average, after every discount, rebate and write-off that was given. Usually below whatever the price list says. is still being discovered, discounts are being tried, and small volumes carry costs badly. Later, everybody selling something similar has learned what it costs to make, so what each unit keeps tends to be squeezed. And the two lines do not turn at the same moment. The picture above already showed as much.
A stage name is a summary of a reading already taken, so it can never be the evidence for that reading. Somebody looks at two lines, forms a view, and writes one word down. The word then travels on its own, and the next person to pick it up treats it as a fact about the product rather than as somebody's compression of two movements into a single syllable count.
Two things follow, and both are refusals about evidence rather than about the idea. The first: no dates. No published year attaches a launch, a withdrawalThe decision to stop offering a product at all. A choice somebody makes on a particular day, rather than something that happens to the product by itself. or a stage to the business used here, and a supplied date would invent the only fact that would have made the label checkable. The second: a stage name may never be used to predict its own input. The word was read off the lines, so it cannot then be turned round and used to forecast them. The reversal is worked through below, and it is not a rare or exotic error.
A plan sheet carries a stage name in its heading and a flat volume forecast beneath it, on the grounds that products at that stage sell flat. What is wrong?
How to Analyse a Product Life Cycle
Six steps, in order, and not one of them is choose a word. Read them once as a sequence, then look at where the sequence stops when it meets real evidence. Most treatments leave that part out.
One. Name the product, singly, and say what one unit of it is. Two. Get the volume line: the count of units sold, period by period, and write down where each count came from. Three. Get the per unit line: what one unit left behind after everything moving with output was paid, period by period, and state which period's rate priced which figure. Four. Set the two movements side by side and say in plain words what each one did. Five. Only now, write the summary word if one fits, and write it as a summary of those two movements. Six. State which of the two lines would have to move for the summary to be wrong.
Steps two and three are where real readings die, and step five is where invented ones are born. Almost nobody fails at step four; comparing two movements is easy once both are held. A word is always available and a count is not, so when step two returns a single figure or nothing, the person doing the work quietly slides forward to step five.
The sequence produces an honest close when a line comes up short, rather than an admission of defeat. The close carries three things: the movements that could be established, the line that could not, and the reason it could not. Handed over in that form, the work is finished. Somebody else can pick it up, disagree with the first part, and go and find the second part, and neither of those is possible if a word was written instead.
The reading sequence is run on a business publishing two years of margin and one year of volume. Where does it stop, and what goes on the sheet?
What happens when the two lines move in opposite directions?
Four combinations exist and only four, so work them rather than describing them. Both lines up: more units sold and more kept on each of them. Nobody argues about that one. Both lines down: fewer units and less kept on each, and the two movements point the same way. Both lines falling together is rarer than it sounds. Volume up and per unit down: the units are moving and each one is worth less than it was. Volume down and per unit up: fewer units, each keeping more of what it fetches. Often somebody stopped discounting in order to hold the price.
Two of the four combinations carry no single stage name that fits them, and those two are the ones a decision usually turns on. A word has to throw one of the two movements away in order to fit, and whichever movement it discards is generally the one that would have changed what anybody did next.
So the instruction is short: name both movements and stop. Two movements printed plainly can be acted on separately, and somebody can disagree with one of them without discarding the other. A single word cannot be argued with in halves.
The household version is close to hand. A household spends more on groceries every month while carrying fewer bags home. Two movements, one bill, and the bill on its own hides both of them. Nobody would accept a one word summary of that household's year, and yet the same one word summary of a product's year travels through meetings unchallenged.
Volume rose over the period and what each unit keeps fell. Which of these is a legitimate output of the reading?
What does the reading produce when one of the two lines carries a single point?
Now run the sequence on the business these notes carry, and watch where it stops. The product is the hard-bound register made by Anjani Stationers Private Limited, an invented business used throughout these notes, and one unit of it is one register.
Step three goes well. The per unit line is available for both published years. The contribution marginThe part of each rupee of revenue still standing once the costs that move with volume have been paid. Worked in full earlier in these notes. was 42.75 per cent in the earlier of the two published years and 42.78 per cent in the later one, so the two published figures differ by 0.03 points. In the later year, one register fetched a realised Rs 108.00/- and carried Rs 46.20/- of contribution. Multiply Rs 46.20/- across the 2,50,000 registers the works made and the year's contribution comes to Rs 1,15,50,000/-, exactly the published figure. The check confirms that the volume point and the per unit figure belong to the same year rather than to two different ones.
Step two does not go well at all. The volume line carries exactly one point: 2,50,000 registers, published for the later year and for no other period whatsoever. The single published point permits one sentence about the per unit line. The per unit line did not really move. One point cannot make a movement, so the volume line gets one figure and no direction at all.
And here is the temptation. A careful reader is already reaching for it, not a careless one. Revenue rose 12.50 per cent, from Rs 2,40,00,000/- to Rs 2,70,00,000/-. Surely the earlier year's revenue can be turned into the earlier year's volume? There are two obvious ways to do it, both using nothing but published figures, and both looking completely sound. The two ways disagree with each other, so neither of them is the volume, and neither answer is printed. The two answers disagree because the margin moved between the years, by a very small amount, and a small disagreement in a divisor is quite enough. Printing either answer would put a candidate volume in front of the reader that would be remembered afterwards, so it stays unprinted.
The reading therefore stops at step two. The sheet carries one point, one flat line, and the reason the second point cannot be built.
The panel below holds the year's volume exactly where the published year put it and moves only what each unit keeps. What appears in the stage slot as it moves?
One line moves as far as the control allows, and the other one refuses to budge
One control, and it moves what each register keeps, away from the published Rs 46.20/-. The volume stays exactly where the published year put it, at 2,50,000 registers, at every one of the forty one settings. Watch three things: the flat volume line that never redraws, the per unit line that travels, and the empty box on the right.
Rs 46.20/- a register Rs 1,15,50,000/- of contribution for the year the published year, exactly
At this setting one register keeps Rs 46.20/-, which is the published figure itself, and the year's contribution comes to Rs 1,15,50,000/- across a volume that has not moved at all. The volume line is still at 2,50,000 registers, so no stage can be read from this.
Assumptions. Only one volume is published for this business, so the volume is held at 2,50,000 registers at every setting. Rs 46.20/- of contribution a register is published for the same year as that volume, and the two multiply back to the published contribution exactly. Every contribution a register other than Rs 46.20/- shown here is a demonstration figure and is published nowhere. The register figure is rounded to the nearest paisa and the year's total follows from the rounded figure, so the two on screen always agree. The published year is the only period with a volume behind it, so a moved contribution figure is arithmetic rather than a later year. One line can travel a long way while the other stands still, and neither movement on its own puts a product on a curve or names a stage.
One duration is published in these notes. May it stand as a product's life?
No dates have been attached to the lines, and yet these notes do publish one span. The Sunrise Public School group has bought from this business for eleven years and is 30.00 per cent of the published year's revenue. Eleven years is the only duration published alongside this reading, so it is worth stating plainly and refusing plainly.
Eleven years counts one buyer's continued custom, and a product life cycle counts one product's own volume and margin. The two are separate counts with separate populations, and a span lifted out of one of them carries no information at all about the other. A buying relationship can run for eleven years across a product that changed completely in the middle of it. A product can finish its whole life inside one buyer's second year. Neither of those is unusual.
There is a second and smaller refusal riding alongside. One buyer at one date is one point, and one point is not a path. The same refusal is made where the pace at which an idea spreads through a group of buyers is set out, for exactly the same reason, and refusing it twice is cheaper than explaining a wrong answer once.
One duration is published anywhere in these notes: a buying relationship running eleven years. What may a product life cycle reading do with it?
What does a finished reading card look like when a row stays blank?
Here is the whole reading, filled in, with the blanks left visible. Four rows. The per unit movement is filled: 42.75 per cent then 42.78 per cent, a difference of 0.03 points between the two published figures, and the plain words are that it was untouched across the two published years. The volume movement carries one point, 2,50,000 registers in the later published year, and beside it the words no second point published rather than an ambiguous empty space. The summary word is blank, and the reason for the blank is the row above it. The last row names what would change the answer, and it is the one thing anybody can actually go and do.
A card with a stated blank can be handed to somebody who knows the trade and argued with. A plausible stage name in the same card leaves nothing to disagree with, so that card cannot. Most people expect the opposite, so the inversion is worth sitting with. The complete-looking card is the useless one.
The stage name that was used to predict the line it was read off
A team is preparing next year's plan for one product. Somebody pulls the last few periods of volume, sees a line that has stopped climbing, and writes one word at the top of the sheet: mature. The word is a fair summary of what that line did, and at that moment nothing whatsoever has gone wrong.
Then the sheet travels. A second person, who never saw the volume line, reads the heading and does the reasonable thing with it: products at that stage sell flat, so the volume forecast is set flat. A third person reads the flat forecast and sizes the spending against it. A flat line does not need supporting, so that step is reasonable too.
The tempting diagnosis is the wrong one, so state exactly what happened. Nobody made an arithmetic error. Nobody invented a number. Every step after the first was a sensible inference from what was in front of it. A summary of a reading was promoted into a cause of that reading, and once the promotion happened the plan rested on itself. The label was read off the flat line it is now being used to predict, and a forecast that rests on itself contains no outside fact to be wrong about.
The cost lands somewhere specific rather than in general worseness. The flat forecast becomes the denominatorThe figure everything else on a sheet is measured against or divided by. Change it and every ratio built on top of it changes, without any of them looking different. for everything downstream, so the spending is sized to it. When volume then moves in either direction, the review asks why the plan was missed. The question is about execution rather than about the sheet. The finding that never gets made is that nobody checked the second line at all: what each unit was keeping may have been moving the whole time, in either direction, and it was never on the sheet in the first place.
And the part worth sitting with. The circle is invisible precisely because the first step was honest. Somebody eventually asks where an invented number came from, so a plan built on one invites challenge. A plan built on a fair summary of a real reading looks as though it has already been checked, and in one narrow sense it has.
The fix is one line, and it is not a better forecast. Print the two movements beside the word. Anybody picking the sheet up later can then see what the word summarised, and can disagree with the reading rather than with the label.
Four lines that travel with any stage name, in this order
An analyst reading somebody else's note runs four checks, and they take under a minute. The same missing line stops a person inside a business deciding what to spend on one product, so that reader runs exactly the same four.
First, the volume movement it was read from, with the periods named. Not the direction in words: the count, in each of at least two periods, and where each count came from. Somebody who answers that the volume has been solid has not answered it.
Second, the per unit movement it was read from, with the convention stated. Which period's rate priced which figure, said out loud. Two figures spanning two periods, priced at two different rates, will move for reasons that have nothing to do with the product.
Third, which of the two lines was estimated, marked line by line. The reader remembers the label and forgets the caveat, so a general caveat at the bottom of a sheet protects nobody. Marking the estimated line where it sits survives the sheet being passed on.
Fourth, what would make the summary wrong. One sentence naming the movement that would break it. A stage name with all four lines blank is a word rather than a finding, and the first two lines alone would have stopped the business used here from ever acquiring a stage name, without anybody needing to argue about the shape at all.
A stage name for a product arrives with nothing beside it. Which single line, printed alongside, does most to make it checkable?
What is local here, and what has to be confirmed at source
| What is set here | What it looks like here | Where it is settled |
|---|---|---|
| The way money is written and the way digits are grouped | Indian convention throughout, on figures belonging to an invented business | Local convention only, carrying no rule |
| The legal form Private Limited, carried by one invented business | A form of words attached to nothing that trades | Indian company law, named for nothing numerical |
| The existence of filed annual accounts | Named once, with no figure attached to it | Ministry of Corporate Affairs, read at its own site |
| The shape and the man who set it out | Named in the body and in the table below | Theodore Levitt, 1965, confirmed at hbr.org |
The mechanism itself is entirely universal. A product's volume and what each unit of it keeps are two separate lines in every market on earth, and a stage name summarises a reading everywhere. No rate, threshold, period or statutory definition arises in this subject, so no regulator appears in this block, and making it concrete with an Indian disclosure limit would require a figure written from memory.
Which of the borrowed names here can be checked, and at what address?
| Source | Document | Site | How it is treated here |
|---|---|---|---|
| Theodore Levitt | His article on the product life cycle, Harvard Business Review, 1965 | hbr.org | Named for the frame, not quoted. The four working words for the stages are ordinary trade vocabulary and nobody is credited with them here. |
| Ministry of Corporate Affairs | The register of annual accounts that Indian companies file | mca.gov.in | Named for the existence of filed accounts and for nothing numerical. A filing carries money and rarely carries the count of units underneath it, so the line this reading needs most is the line no filing requires. The missing line is the ordinary situation rather than a fault of the business used here. |
| The arithmetic above | Earlier in these notes, where each figure was first published | finmaverick.com | Two divisions, one subtraction and one multiplication, every one of them repeatable by hand in under a minute. |
Anjani Stationers Private Limited and the Sunrise Public School group are invented.
Educational material. Not advice on any investment, tax, budget or market position.