Followers Not Growing Need Not Be Handled as a Problem of the Business

You post every day and the figure has stopped in three digits. You look at the posts of people who are growing and hunt for what is different. You read the articles on how to increase it and tried them all, and the only movement came in the first few days. You believe that until this figure grows, none of it will become work.

The reasons offered for not growing usually sit on the side of quality or quantity. The content is weak, the frequency is short, the presentation is poor. Each is partly right, and fixing it does move things a little. What it moves is the ranking inside the place. Carry on for a few months with the business side still not moving, and what gets doubted next shifts to your aptitude: I am not suited to this place.

One step before that, something is checkable. Not growing and not selling occur independently. The responses being advertised do act accurately on the ranking inside the place. What they act on, though, is the inside of the place, not the quantity of the business held in your own hands. And what the structure makes likely is the other case: growing while the revenue does not move. If the two are independent, one of them being stalled says nothing about the other. And being independent is not the same as being unrelated. The same place takes the same hours, so effort put into one is subtracted from the other. It is not that there is no relation; the relation runs the opposite way from the one assumed.

Three things sit outside what is covered here. No procedure for growing the figure appears. Which formats grow, what hour to publish at — these differ by place, and since the place decides them, they have changed by the month after they are written. Where the display count is itself the revenue — advertising, or work taken on conditional on reach — the separation below is not needed either, since the follower count is then the variable of the business. And what the word place means here is limited to somewhere holding machinery by which the place selects who is shown. Where you decide who receives — anything that lands in everyone’s hands once sent — is outside this analysis. The division is not by the name of the medium but by who holds the decision over display.

The coverage runs to four. How many separate states the single phrase not growing folds together. Who defined the follower count, and for what purpose. Under what conditions growing and selling stop being linked. And what is counted instead. With the four dealt with, three indicators are placed for checking whether the replacement has taken hold. None of the three measures the rise or fall of the follower count.

Beneath Not Growing Lie Three Separate States

Behind the result of followers not growing lie at least three separate causes.

The first is a state of not being seen at all. The count of times a post gets displayed is low, and the occasion for being judged has not yet arisen. At this stage the quality of the content is not reflected in the result.

The second is a state of being seen with no reason to stay. Display is occurring and the content is being read. What is not visible, however, is enough consistency for someone to judge that they want to read the next one too.

The third is a state of people staying without it appearing in the figure. You are being read regularly, and no additional operation has been performed. For the reader, that operation is not required.

All three surface as the same stagnant figure. The move each one needs is different. The first requires an increase in exposure, the second a making-visible of consistency, and the third requires no operation at all.

The third is fine, since you are being read in the end — and it is fine. If there is a problem, it lies in mistaking the third state for the first or the second and carrying out work to increase exposure. Do work to increase exposure when you are already being read and only the frequency of posting rises; what comes back does not change.

The three overlap. It is not rare to be unseen by one layer while being read by another that does not show up in the figure. From the single result of not growing, therefore, the cause cannot be narrowed to one.

To tell them apart, look separately at the count of times displayed, the count of times read, and the headcount added. Few displays means the first; a low proportion read against displays means the second; being read with no additions occurring means the third. All three values can generally be checked in any place.

This division into three is a classification of causes, not an ordering of severity. The first looks like the earliest stage, and in practice people who have continued for a long time sometimes return to it. Change the rules of display and a state of already being read moves overnight into a state of not being seen. Rather than progressing as stages, they are travelled back and forth by decisions made outside.

The single phrase followers are not growing folds the three together and then places the cause on the side of the publisher’s ability. Handled as a matter of ability, the prescription becomes put out something better. And when someone in the third state carries this out, they increase what is already sufficient, and nothing moves in proportion to the effort invested.

The Responses in Circulation Optimise the Ranking Inside the Place

Responses for growing the figure in that place come to six.

Post high-quality content continuously. Understand the machinery of display and put things out in a form that suits it. Narrow the tags and use only highly relevant ones. Tidy the profile so that what you do is legible at a glance. Increase two-way exchange rather than one-directional publishing. Cooperate with people in the same territory and introduce one another.

All six do operate inside that place. The observation that one-directional publishing grows poorly is particularly accurate. The machinery of display responds to volume of reaction, so posts generating exchange are displayed more widely. Tidying the profile also acts directly on the second state — no visible reason to stay.

What the six optimise is a single thing. All of them are operations for raising a ranking inside that place. Suit the machinery of display, optimise the tags, increase reactions. Input and output alike close inside that place.

As a rule of the place, this holds together. If you want to be found in that place, following the rules of that place is the shortest path. The problem lies elsewhere: what moves on the side of the business when that ranking rises has not been verified.

An objection can be anticipated here: that if the ranking rises the number of people seeing you rises, so surely something comes back to the business.

Sometimes it comes back. Sometimes it does not. And what settles whether it comes back is not the ranking but whether a route exists to hand people onward once the ranking has risen. Let the ranking alone rise with no route in place and impressions increase, and that is where it ends.

Calling the six wasted would be wrong. For someone in the first state — not being seen — they are squarely the required operation. What is under discussion is range. And the range ends at the boundary of that place.

Only the prescription about cooperating and introducing one another has a slightly different property. It is not an optimisation inside the place but an operation borrowing a route someone else holds. It also acts differently: the character of the people who gather is settled by the source of the introduction. Handle it in the same frame as internal optimisation and this difference disappears from view.

This family of prescriptions takes a shape in which the effort invested is easily connected to the result. Increase posts and displays increase; increase exchange and reactions increase. The connection is visible, so the work on your side continues. And while it continues, the fact that the figures on the side of the business are not moving is interpreted as a shortfall of quantity.

The Follower Count Is a Value the Place Defined for the Purposes of the Place

The follower count is the number of people who enter the pool of candidates for having your posts displayed to them, in that place.

Candidates, not the number of people reached. Whether display actually occurs is decided by the place on each occasion. And the procedure is not published, and it changes whenever the place wishes. That is, this value is not even a ceiling on how much arrives; it is no more than the size of a set that might possibly be reached.

Consider why this value is displayed so prominently and its property becomes clear. For the side of the place, this value is effective at making participants stay. Keep a figure that pleases when it rises and nags when it falls permanently in view, and a reason to come back and check it is generated. The position in which it is displayed states the purpose of the value.

A general property of indicators comes into force here. An indicator placed under pressure for the sake of control loses the relation it held before the pressure. The moment raising the number becomes the aim in itself, the operation that raises the number and the operation that increases what the number was originally measuring diverge. Where this observation comes from, and whose words the familiar one-line version actually are, is handled directly in how to read customer acquisition cost. What is used here is only the property that the divergence occurs.

This divergence is in fact observable. Reports of a figure grown through highly topical posts failing to return anything to the business come even from the side of those explaining how to grow it. The people who gathered are not necessarily interested in what you supply.

The figure is not without a use of its own. Past a certain scale there are situations in which the figure itself works as material for judgement. It becomes a cue for someone seeing the name for the first time when deciding whether to read on. That, however, is a working as a signal, not a quantity of business.

Even if they are only candidates, more candidates must mean more people reached. On average, yes. The proportion arriving falls with scale, however, so the increase is not proportional. And since the side setting that proportion is the place, how many times the reached headcount multiplies when your figure multiplies tenfold cannot be calculated in advance, or afterwards.

This value has one more property. You cannot explain its rises and falls yourself. What worked on the day it rose, what happened in the week it fell, can be no more than conjecture so long as the machinery of the place cannot be inspected. Place an unexplainable figure as the target and neither the reasons it went well nor the reasons it went badly accumulate.

The Number Who Can See You and the Number You Can Contact Are Different Quantities

The number who can see you is the number of people to whom your publishing might possibly be displayed. The authority over whether it is displayed lies with the place. This number grows easily, and the proportion arriving is small relative to its size.

The number you can contact is the number of people you can deliver to, all at once, from your own side. The authority lies with you, and the proportion arriving is higher by an order of magnitude. This number grows with difficulty, and what is added accumulates as it is.

The two differ in the nature of their growth. The value of the number who can see you changes overnight if the machinery of the place changes. The quantity arriving changes when the rules of display change, even if you have done nothing. The number you can contact remains as it is when you do nothing.

The two part on one question. Could you contact these people if that place became unusable from tomorrow? If you could, they belong to the number you can contact. If you could not, they are the number who can see you.

Building a contactable state requires putting the other party to some trouble, and that may look impolite. There is trouble involved. And to have someone take on that trouble, you must have handed something over first. Demand the trouble before handing anything over and it genuinely does not hold. This is a matter of order, not of politeness. The design of what gets handed over is treated in designing a lead magnet.

Nor is it the case that a larger number of contactable people is simply better. Holding the contact details of large numbers of uninterested parties builds a state where only cost rises and nothing comes back. What should be measured is not the number but the proportion read on arrival. This point is handled squarely in the premises of list building.

The moment the two are divided, one of the figures looks small. Move your viewpoint from a state of several thousand people being able to see you to a state of being able to contact several dozen and it feels like a retreat. Compare them by the real count arriving, however, and the latter being larger is not unusual. Apparent size and quantity arriving have inverted.

This inversion can be checked. Count the number of people who actually opened something you sent to everyone at once, and set it beside the number of people to whom a post on the same day was displayed. One may be several thousand, the other several dozen. Even so, the number within the former who read the content does not come out of the display count, whereas the latter yields the number who read it directly. The kinds of quantity being counted differ, so comparing size alone does not settle an order.

Set side by side, the former is sometimes the larger. Even then the difference in the nature of the two figures remains. One changes next month by a decision on the side of the place; the other does not. Which is larger now and which will be the same size next month are answers to different questions.

The Engagement Rate Carries the Same Property

There is advice that if you doubt the figures, the rate of reaction is what you should look at instead. The direction of that advice is correct. The rate, though, is defined in the same place.

Engagement refers to the proportion of displayed posts on which an operation beyond reading was performed. It is indeed a quantity closer to substance than the follower count. It distinguishes people who merely saw from people who actually reacted.

The property, however, is the same. Both numerator and denominator are defined by the side of the place. What counts as a reaction, and how displays are counted, differ from place to place and change over time. The value changes when the definition changes, even with identical activity.

And this rate tends to move inversely with scale. As the number who can see you grows, layers with thin interest enter the denominator, so the rate falls. That is, the two values are hard to improve simultaneously, and the assessment inverts depending on which you look at.

The rate is useful in one place: comparing differences of content in the same place at the same scale. Level the conditions and differences of content show up in the rate. What it cannot do is serve as an indicator expressing the state of the business.

A post drawing many reactions must be a good post, one might say. The accurate answer is that forms which readily collect reactions exist. Strong assertion, the presentation of conflict, a surprising figure. These increase reactions independently of the quality of the content. And leaning towards this form changes the character of the people who gather. People who react readily gather, and people whose judgement is careful drift away.

A falling rate does not itself mean deterioration. There are cases where it falls because layers with thin interest entered the denominator. In such a case the rate falls while the real count of reactions rises. When rate and real count move in opposite directions, the conclusion inverts depending on which you look at, so no judgement is possible unless both are set side by side.

Place the rate as a target and the content of posts starts being selected by short-term reaction. The shorter the time until a reaction returns, the more readily a thing gets selected. And content meaningful to a business — the long explanations that become material for judgement — is generally slow to draw reactions, and draws few. What returns slowly does not arrive in time for the rate’s assessment. The cycle of assessment is selecting the character of the content.

The selection hardens the longer it runs. Once experience accumulates of figures moving on fast-reacting formats, the material for deciding what to publish next becomes that experience alone. Judgement comes to be supplied only by what returns quickly. Slow-returning content is known before publication not to grow, so it drops out of the candidates first. Dropped candidates leave no record, so the selection itself goes unobserved.

Research on organisational learning has a name for this shape. Denrell and March set it out in a 2001 paper as the hot stove effect, treating what happens when experience is used both to evaluate alternatives and to choose which one to sample next (Denrell & March, 2001, Organization Science, 12(5), 523–538). An alternative that returned a poor result once gets chosen less often afterwards. Where it is not chosen, no new result comes back. So long as your own record is the only material, an initial estimate that was wrong stays wrong. The two set this out as a shape in which adaptation restricts information. Publish something slow-returning once, watch it fail to grow, and that format becomes less likely to be chosen next time — and what stops arriving is the occasion for measuring how much it actually works. The fact of not having measured appears nowhere in the figures.

Not Growing and Not Selling Occur Independently

Here is the centre of this article. The two results appear in four combinations.

The first is growing and selling. A state in which figures and business agree.

The second is not growing and selling. A state in which the necessary exchanges hold with a few hundred parties. In businesses that stand on individual commissions, this is the standard case.

The third is growing and not selling. Impressions and reactions are both rising while nothing connects to commissions or revenue. This state is structurally likely.

The fourth is not growing and not selling. Only in this case do two problems exist at once.

Of the four, only the first and the fourth are states in which the two quantities move together. In the second and the third they do not. That is, reading the state of a business from the follower count is wrong in two of the four combinations.

Why the third is likely can be explained from the structure. The condition under which figures grow differs from the condition under which commissions arrive. Figures grow when you put out content that many people find interesting. Commissions arrive when it is conveyed to a small number of people holding a specific difficulty that the difficulty can be solved. The former widens the range; the latter narrows it.

This is not to say the two are incompatible. A state in which you handle a specific difficulty and the manner of handling it happens to interest a wide layer does hold. That, however, is the result of design, and not something that occurs naturally as a result of chasing figures.

The larger the pool, the more prospective parties it contains. It does. What increases, however, is the absolute count, not the proportion. And your own load rises in proportion to the pool. Grow the pool alone while the proportion stays low and the volume of correspondence reaches its limit first.

Someone in the second state — not growing and selling — has no need to mind the figures. And yet it is often precisely people in this state who mind them. Even with the business holding, they appear to themselves not to have reached the standard being spoken of around them. The fact that the state of a business and the appearance of it in a place are separate holds just as firmly when the business is turning.

Someone in the third state interprets the cause as a shortage of figures. Nothing comes back because there is still not enough. And continuing to work at increasing the figures advances further in the direction of widening the range, moving away from the condition under which commissions arrive. The response takes the shape of intensifying the symptom.

Divide Businesses That Need Numbers From Businesses That Do Not

Are numbers entirely unnecessary, then. They are not. The number required is settled by the shape of the business.

In a shape where impressions themselves become revenue — advertising, or commissioned placements — the number is a direct variable. In this case, chasing the follower count coincides with chasing the business. It has legitimacy as an indicator.

In a shape supplying a specific party at a comparatively high unit price, the number required is small. If a few dozen commissions a year are required, a few hundred contactable parties suffice. In this case, chasing numbers is work that moves away from the business.

In between — a shape supplying something of small unit price to many — the number required is large. What is required, however, is not the number who can see you but the number who travelled the route as far as purchase.

There is one way of dividing. Multiply the unit price by the count of transactions required, and work backwards from there. Once the count required is known, how many people are needed upstream of it emerges. Perform this calculation and a figure smaller than expected often comes out.

The arrival rates used in working backwards are not known at first. They are values that emerge only once people have actually been passed along the route. The first estimate is therefore rough, and requires correction. Even so, having a rough estimate is more usable for judgement than chasing figures with no target at all.

If the number required is small, perhaps there is no need to publish at all. What is required is not the number but the route. Even for a few dozen people, a route is required by which they find you, judge, and arrive at a contactable state. Publishing is carried out not to manufacture numbers but to build the entrance to a route. How the stages are arranged across the whole is placed as a map in the design of a marketing funnel.

This working-backwards presumes the unit price is fixed. Change the unit price and the count of transactions required changes, and so does the headcount. That is, the number required is an output of the design of the business, not an input settled in advance. Where to cut the stages belongs to raising the amount a customer spends.

Once it is understood that the number required is small, the urgency returns in another form. It should only take a few hundred, and even those few hundred are not gathering. Lowering the target for the number does not change the absence of a route. Working backwards moves the location of the problem from numbers to routes; it does not make the problem disappear.

The Conditions a Replacement Figure Must Satisfy

Enter the prescription. Rather than specifying what should be measured, however, let me place the conditions. Naming an indicator would fail to apply to anyone whose business has a different shape.

The first condition is that you can explain the rises and falls yourself. When that figure moves, can you say what you did that moved it. A figure you cannot explain cannot be reproduced when it improves.

The second condition is that the value does not change when the place changes its specification. A figure that changes through decisions made outside, while you have done nothing, does not express your state.

The third condition is that the figure connects directly to a quantity of the business. The number of contactable parties, the proportion read on arrival, the count of enquiries, the count of orders. Every one of these is a quantity on the side of the business.

Figures satisfying all three are generally unglamorous. They move slowly, cannot be compared with other people’s, and attract no interest when published. The property of not being comparable is in fact an advantage. Other people’s values are not displayed in the same unit, so comparison does not arise at all.

Replacing does not mean you stop looking at the figures on the side of the place. So long as you use it as an entrance, impressions carry meaning as information about how well the entrance is working. What changes is only the order — which of them is placed as the target.

Looking only at unglamorous figures leaves you without the felt sense of growing, one could object. The felt sense does diminish. And its diminishing is a large cost of this replacement. The daily highs and lows disappear, so the motive for continuing stops being supplied from outside. In exchange, whatever moves accumulates.

Figures satisfying the three conditions suffer the same distortion when treated as targets. Place the count of enquiries as a target and operations that increase the count — loosening the conditions, widening the audience — start being selected. The property by which an indicator ceases to function as a measure once it becomes a target works identically on every figure. What the three conditions guarantee is not freedom from distortion but the ability to explain the distortion yourself when it occurs.

Perform this replacement and you stop being able to talk with other people. In places where publishing is discussed, the common subject is the figures on the side of the place. Looking at different figures is frequently read as an excuse for not producing results. The replacement includes shouldering that reading.

The range to be shouldered can be bounded. The reading only matters with people who discuss publishing in the same place. The people who commission work are not watching what you count. They are watching only whether you can be asked. The cost of replacing falls on conversation with peers, then, and not on relations with the people your business serves. Without separating the two, the cost gets estimated larger than it is.

Reading Whether the Replacement Took Hold, in Three Figures

The change shows somewhere other than the rise and fall of the follower count.

First, look at the behaviour that follows looking at a figure.

If an unplanned post used to follow checking a figure and that chain has stopped occurring, the target has moved. The material for the determination is not how often you look at the figures but what happens after you look.

Second, look at whether you can state a reason for a week in which nothing grew.

If before the replacement you could say no more than perhaps the machinery of display changed, being able to say something now is evidence of looking at an explainable figure. Being able to explain also means that what to change next gets settled.

Third, put that same question — the place unusable from tomorrow — and count the figures that remain.

If the remaining figures are zero, the replacement has not progressed. If even one figure remains, that is the portion accumulating on your own side.

None of the three puts the rise or fall of the follower count to work. What was placed instead is the chain of behaviour, the possibility of explanation, and the presence of remaining values. The follower count was not a quantity of the business, so whether it grew is not material for the determination.

The figure stopped in three digits is still in three digits after the reading changes. What changed is what quantity that figure was. It was never placed there to measure the state of your business; it was put in front, by the side of the place, to give participants a reason to come back. That its rises and falls could not be read was not a failure of reading. It is a value positioned where reading is not possible.

The time spent counting it still has a use. The same figure can be read again as an inspection of the entrance, rather than as an indicator of the business. Display fell, so the entrance narrowed; reactions are dull, so the shape of the entrance does not fit — that much is readable. What is not readable is only the next step, and therefore how the business stands. One figure was being asked to do two jobs, so both answers came out vague.

One thing has been added to what you can do the next time you look at the figure. Recall once who placed that value there, and for what. Then watch whether an unplanned post follows the looking. If it no longer follows, the target already sits somewhere else. If it does, it has not moved yet. Either way, what is being read is not aptitude.

The order of diagnosis to run before chasing that figure is treated in the diagnostic procedure for why nothing sells, and the principle that only a place able to count those who leave can be improved in how to read the drop-off rate. The design of holding a contactable state itself sits in what comes before list building, and the origin of the load of chasing figures in Being tired of social media comes from stacking on rented ground.