A Sole Trader Chooses Online Payments on Four Axes and Compares Fees Last

Whether a sole trader can set up online payments already has its answer. They can. Supporting services are numerous and comparison articles are plentiful.

The question was replaced by another the moment that answer landed. What actually stalls is not “can I” but “on what basis do I choose”. And opening a comparison article without noticing the replacement means using whatever criteria are lined up there.

What comparison articles tend to put at the centre of the criteria is the fee. Fees resemble one another closely everywhere. Take something uniform as the criterion and nothing can be chosen. Unable to choose, the decision falls to rankings or to which name is recognised.

Setting up payments is not acquiring a means of receiving money. It is standing in a position from which the terms of a transaction can be defined on your own side. When, how much, in what form the money arrives. That right of definition is nothing other than the body of what is being chosen.

The question splits in two. Only the far side is answered here. What is happening such that fees settle at similar levels. How many axes of selection there are, and which. And what settles the order in which the four are examined.

On the tolerable level of cost, no figure is held here. Answer the question of what percentage is reasonable and the answer misses. The same rate carries an annual burden differing by an order of magnitude with unit price and volume, and a different deposit cycle moves how funds circulate. Since the meaning of cost depends on the other axes, a level extracted from cost alone has lost its conditions in the extracting.

The record of what was not chosen is worth setting down too. An option that fell away while no means of receipt was held does not remain as something that fell away. Trying a form of continuing provision, and shelving it for want of a payment method. That shelving does not become a record of examination and rejection; it vanishes as something that never rose for examination at all.

Vanishing, it cannot be counted afterwards. “That form does not suit me” and “that form could not be tried for want of a means” become, in memory, the same single fact. The material distinguishing them was already lost at the moment the option fell away. That is why the axes need rebuilding while the distinction still holds.

The order goes down first as well. The work of raising axes comes before the work of narrowing candidates. Start from candidates and the comparison items the candidates carry become the axes as they stand. The state in which the object of choosing also supplies the criteria of choosing is the default.

The Guidance on Choosing Is Built from a Ranking by Fee

The guidance on choosing is organised around fees and presented as a ranking on a comparison table.

Copy the guidance across as it stands first. Online payment is machinery for receiving money without meeting face to face. Credit cards, code payments, electronic money and direct debit are among the methods.

Two routes of setup are contrasted. Contracting directly with card companies, and using a payment agency. Sole traders are advised toward the latter, on the ground that contracts with several card companies are handled together and the procedure grows simpler.

Three costs are explained. The setup cost, the fixed monthly cost, and the fee per transaction. The fee is given as roughly three-point-something per cent of the transaction.

Screening is touched on. Sole traders tend to face harder screening than companies, runs the account, because years of trading and financial condition fall within the scope of examination.

The points raised for choosing are the level of fees, the payment methods supported, ease of setup, and fit with the use case. On use cases, distinctions are drawn between over-the-counter selling, distance selling and invoiced payment.

A comparison table naming specific services occupies the centre of such articles. Many are ordered from the cheapest fee.

On the factual side, nothing so far is in error. It is a fact that an agency makes the procedure simpler, and the account of screening tendencies is sound. The three-part division of costs corresponds to actual contracts.

The six accounts share one frame. Payment is treated as a means of processing a transaction already settled. What is sold at what price is settled, and only how the money is received remains at issue. It is organised as a problem of processing.

It is not that the organisation is an error. Cases of simply moving existing transactions online are common. Complete it on the processing side alone and the fact that the form of payment governs the form of the transaction drops out of view.

And comparing by fee leaves differences too small to decide on. Unable to decide, the judgement moves to the order things are listed in and to whether the name looks familiar. The ground for the selection is effectively absent.

A groundless choice carries no criterion for review either. Not knowing why it was chosen means being unable to judge, when conditions change, whether it ought to change. The ground for selection is used not only in choosing but in choosing again. Begin holding no ground and continuing to use that machinery becomes the default.

Holding Payments Means Standing Where the Terms Can Be Defined on Your Side

Without a means of payment, the terms of a transaction are subordinate to the other party’s practice.

Concretely: send an invoice and wait for a transfer, and the following cannot be settled on your side. When the money arrives. You follow the other party’s closing date and payment date.

Whether it can be received in instalments is likewise theirs to determine. Making it a continuing receipt requires an invoice each time and a transfer each time. And handling a late payment falls to you, individually.

Hold a means of payment and these reverse. When money is received becomes the moment you set. Continuing receipt repeats on a single configuration. Processing a failed payment is executed by the machinery.

Which is to say what setting up payments yields is not a reduction of labour but the right to define terms.

The difference reaches the form of the business. Define the terms of receipt and the form of provision can be settled first. Whether to provide continuously or once only. Without a means of receipt, the form of provision is settled by how the other party pays.

While counterparties are few, transfers suffice. At low volume, handling each individually is faster.

Where transfers stop sufficing is when the form of provision is changed. Try a continuing form and it falls out of the options at the point no means of payment exists. What fell away leaves no record. Whether it went unconsidered because it was impossible, or was considered and not chosen, cannot be distinguished afterwards.

Once terms can be defined, the work of deciding terms arises. When, how much, on what cycle. Since nothing can be configured undecided, setting up payments puts the undecided on the table.

Holding the right of definition and exercising it are separate. Where the other party is an organisation with settled payment procedures, being unable to impose your terms can happen.

Even so, holding two routes means something. Receiving by payment system and receiving by invoice. With only one, you match the terms on their side. With two, you use them differently by counterparty.

And holding two tells in the negotiation of terms itself. A side holding only one has two options: accept on those terms, or decline the transaction. One more available form changes the weight of declining.

The line for using them differently can be drawn by whether the other party’s procedure is settled. Where it is, do not force your route in; match their form. Where it is not — often with individuals — put your terms first. The same provision can carry terms of receipt settled separately for each counterparty.

Fees Resemble One Another Because Price Is Allocated Across Two Sides

The level of fees is settled by the structure of the market, not by differences of effort between operators.

Card payment is a market requiring two sides at once. Without cardholders no merchant joins, and without merchants there is no point holding a card.

The economists Jean-Charles Rochet and Jean Tirole analysed the properties of such markets in 2003 (Rochet & Tirole, 2003, Journal of the European Economic Association, 1(4), 990–1029). The proposition at the centre is that the decisive variable is not the total price but its allocation.

Where charges can be taken from both sides, how much is taken from which changes how many participate. Making one side cheap to gather people and recovering from the other is an allocation that holds.

The difference of object is worth placing first. What this analysis looks at is the structure of the market, not the merits of individual contracts. What holds is that allocation tells as an independent variable.

In card payment this allocation shows clearly. The side using the card bears no fee and receives rewards instead. The side bearing the cost is the side receiving.

And the allocation is not something individual operators can negotiate. It is settled as the design of the whole network. What agencies take on top differs somewhat, but the structure beneath is shared.

So fees settle at similar levels everywhere. They resemble one another not for want of competition but because they sit on the same structure.

Carried across, the implication for selection runs thus. Fees carry little weight as an axis of selection. Take something structurally uniform as the principal criterion and the judgement holds no precision.

It is not that fees can be ignored. As transaction values grow, one percentage point tells as a sum. The handling is a matter of order. Compare on fees after candidates have been narrowed on other axes. Order by fee first and the other axes go unexamined while the candidates are fixed.

Low-fee services genuinely exist. Where it is low, it is being recovered somewhere. A long deposit cycle, a narrow range of supported transaction forms, a charge on the refund procedure.

Choose on fees alone without confirming the place of recovery and the unconfirmed items tell later. The gap between three per cent and three and a half matters less to how funds circulate than the gap between the end of next month and the month after.

The structure by which the cost is allocated to the receiving side bears on pricing as well. Whether the fee is folded into the price or charged separately. In many fields, charging separately is not accepted as practice, so it gets folded in. Folded in, part of the price is committed to fees from the outset, though that fact does not appear in the price displayed. Where prices are set by building up from cost, this is an item that readily goes missing. Setting prices as such is handled in not starting price from cost.

The Axes of Selection Are Four, and Fees Are Only One of the Four

The material for the judgement divides into four axes.

The first axis is the period to deposit. The days from payment to the money reaching your account. And whether that cycle is fixed or selectable.

The second axis is the treatment of refunds and disputes. What procedure follows when the other party seeks a refund. What is asked of you where a challenge is raised through the card company.

The third axis is the transaction forms supported. Only one-off receipt, or continuing receipt as well. Whether amounts can be changed and payment stopped partway.

The fourth axis is cost. Fees, setup cost, monthly charge.

Of the four, the hardest to change afterwards is the third. Discover after the form of provision is settled that it is unsupported and the choice becomes changing the business or changing the payment system.

The second tends to be overlooked. Refunds and disputes have not occurred at the point of starting, so they are hard to imagine. And when they do occur, there is no leisure for checking terms.

Confirm the four in order and candidates narrow of themselves. Candidates thin on the third axis, and thin further on the second. Comparing what remains on the fourth is the order.

The falling has a direction too. Candidates falling on the third and second can be attractive on cost. Narrowing supported forms and simplifying the refund procedure is what lets the rate come down. Order by cost first and candidates that ought to fall on those two appear at the top. Change the order alone and the faces at the top are exchanged.

This order carries a further effect. In putting the first through third into words, the conditions of your own business get settled. Answering when the money is needed requires knowing when money goes out. The work of selection is shaped to prompt a grasp of the business side.

Confirming the four means reading contract clauses, since these are items absent from the comparison table. To narrow what gets read, take two places: the deposit cycle and the refund procedure. Both are written on the operator’s explanatory page or terms of use. Where they are not, that itself is information.

The judgement holds that the second and third have not happened yet and can be thought about when they do. Think about it when it happens and you are choosing again. And payments fall among the costly things to choose again. With anyone receiving continuously, all of them are asked to go through a procedure once more.

Laid out, it runs thus. Items that can be handled after the fact and items that must be chosen before it are different. Cost can be negotiated after the fact; an unsupported function cannot be added after the fact. Not lining the four up as equals, and looking in order of how hard they are to change, is the reason.

The Period to Deposit Cuts into the Design of the Business

The deposit cycle settles when funds are on hand, and that settles what can be paid first.

Run solo and this tells directly. A company has a working capital facility; for an individual, what comes in often becomes the source for the next outgoing as it stands.

Concretely: on a cycle closing at month end and paying at the end of the following month, money received early in a month reaches you roughly two months later. Two months of outgoings have to be met from another source.

Choose a service with a shorter cycle and that requirement shrinks. Weekly, or on request at any time, are among the forms. A short cycle can carry a cost attached, however. An additional fee for early deposit, or a higher base fee.

This cost has the same nature as the cost of borrowing funds. It is the price of receiving early. Seen so, the object of comparison changes. Hold a separate means of borrowing and the cycle may be long. Hold none and the shortness of the cycle is worth paying for.

While transaction values are small, the effect of the cycle is small too. The cycle is a hard item to change afterwards, though. Try to change it once transactions have grown and the payment system itself gets moved.

And moving requires rebuilding the configuration of anyone receiving continuously. Asking them to go through a procedure again means a certain proportion leave. Which is to say this item carries a particularly high cost of switching.

Choose on the cycle and you compromise on other conditions. Something best on all four axes is hard to find. The order of compromise is settled by ease of change. Cost has room to be changed later by negotiation or switching; supported transaction forms and the deposit cycle are hard to change. Prioritising what is hard to change and compromising on what is easy is the rational order.

The cycle tells most where outgoings come first. In a form where materials or subcontracting are paid ahead, a delayed deposit tells directly. Where provision is your own time alone, the effect is smaller, since little goes out ahead. Which is to say the weight of this axis varies with the cost structure of the business. “Shorter is better” cannot be said flatly.

There is one way to measure the weight. Look at how many times the total of what goes out ahead fits into the period to deposit. The smaller the multiple, the more the length of the cycle surfaces directly as a shortfall of funds. Large enough, and the cycle can yield to the other axes. What is being measured is not the terms of the payment system but the ordering of your own outgoings.

Refunds and Disputes Are Judgements Settled Beforehand

Requests for refund can arise, and if the response is not settled beforehand it gets settled on the spot.

The frequency depends on the nature of the business, but it is not necessarily zero. And a judgement on the spot is made while the relation is under strain.

Three things are worth settling beforehand. In what cases a refund is given. Until when requests are accepted. By what method the refund is made.

All three depend on the functions the payment machinery provides. Whether partial refunds are possible, whether transactions past a deadline can be reversed, whether the fee comes back. The executable range differs by system.

So the order runs: settle the refund policy first, then choose machinery that can execute it. Proceed in the reverse order and the policy gets matched to what the machinery can do.

Disputes arise by a route separate from refunds. Where the other party raises a challenge with the card company, the procedure advances without your agreement. What is asked for then is a record showing the transaction was legitimate.

What to keep as a record can be settled beforehand. What was provided, when it was provided, the record of their agreement. Design these to be kept automatically and the response to a dispute becomes a matter of producing records.

Keeping records does not guarantee the judgement goes your way. The judgement is made by the card company, and you are not a party to it. What makes them worth keeping is that without records there is no means of reply. Not a guarantee of advantage but the securing of a means.

State the terms of refund and people will make use of them. Set the terms as “no reasons asked within the window” and they get used exactly so.

That cost is the price of stating them. And the price of not stating them is a judgement each time, and judgements that do not stay consistent. Inconsistent handling leaves a burden on the party handled and on you. Which price to take is settled by the nature of what is being provided.

Stating them also works to keep you out of the scene of judgement. With terms written, a refund request is processed against the terms. Unwritten, it becomes a negotiation with the other party each time.

Turned into a negotiation, what settles the terms is the dynamics of the relation. Accede to those who press hard and not to those who do not: that shape comes readily. Stating the terms is a device against that unfairness as well. The general rule that only what has finished being decided can be automated tells in the refund setting in the same shape.

Failing Screening Does Not Mean a Defect in the Business

What screening judges is not the quality of the business but the risk carried on the payment system’s side.

Concretely, the following are examined. The possibility that provision does not complete. The possibility that refund requests arise in number. Whether the field falls among those restricted from handling.

Sole traders are said to pass less readily because the material available for these judgements is scarce. The record of years traded is short and financial documents are limited. Scarce material is handled without being distinguished from high risk.

This shape is widely observed in credit to small businesses. The economists Allen Berger and Gregory Udell showed in 1995 that lenders to small firms rely on private information accumulated through the relationship rather than on public information (Berger & Udell, 1995, The Journal of Business, 68(3), 351–381). Borrowers with longer relationships obtain easier terms.

The object here differs as well. What this study looks at is bank lending to small firms, not payment screening as such. What holds is that the material for judgement accumulates over time.

So what to confirm first on failing is whether the material was short or a condition was met.

Where material was short, time resolves it. Years accumulate, records grow, and the judgement changes. Where a condition was met, time does not resolve it. If what is handled falls in a restricted field, another system has to be found.

The distinction may not come from enquiring, since disclosure of reasons is restricted. In that case applying to several operators can distinguish it. Failing everywhere means a condition; passing somewhere means material.

Going to another operator on failing looks like a waste of time. Screening criteria differ by operator. The same content being treated differently genuinely occurs. And this information cannot be had beforehand, since the criteria are unpublished. Applying is the main means of learning them.

Repeated applications can leave a record. That numerous applications in a short span might themselves tell against you cannot be ruled out. As a way of proceeding, going in order rather than sending many at once is safer. And moving on after obtaining some clue to the reason avoids failing repeatedly for the same reason.

Means of supplementing scarce material partly exist. A page explaining the content of the business, a list of what is provided, and where transactions already exist, their record. From the screening side, these become material for judgement. Where scarcity of material is the cause of the disadvantage, putting out material leaves room for change.

And arranging these is not work for the screening alone. Making what you provide externally legible is work a business ought to settle in advance anyway. Screening can be used as the occasion that prompts it.

Continuing Receipt Is a Different System from One-Off Receipt

One-off receipt and continuing receipt are different things, technically and contractually.

Whether it is supported differs by operator. What continuing receipt requires is the following. Retaining the other party’s payment details. Billing automatically on a settled cycle. Being able to change amount and cycle. Being able to stop from the other party’s side.

The last item is frequently missing. In a design where stopping cannot be done from their side, requests to stop arrive as individual messages.

And a design that is hard to stop raises the continuation figure. The raised figure looks like evidence that the content is supported.

This distinction has the same shape as the structure handled in what settles retention. A figure held up by difficulty of stopping and a figure with reasons to continue are recorded as the same figure.

So being able to stop from the other party’s side is not one function among others but the condition that settles what the figure means. Without it, what the retention rate indicates becomes unknowable.

Make stopping easy and the number stopping rises. This is the movement the design intends. What rises are people who wanted to stop and could not. They were paying while lowering their estimate of you. It is lost as a figure, but what was lost had been lost already.

Machinery supporting continuing receipt is heavier in cost and configuration than machinery for one-off alone. Handling only one-off at present, it is an excessive choice. The judgement is made not for now but for the near future. Where continuing provision might be tried, choose machinery that supports it. Where there is no prospect, one-off alone suffices. What tells here is that the supported transaction forms are hard to change later.

Adding a separate system when continuing provision begins is one move as well. Running two systems together is possible.

In that case records split across two places. Seeing who paid how much means opening two control panels. Split records generate reconciliation work every time. The judgement is made on the frequency of that work. At a monthly aggregation, split records are a small burden. Where reference is made on every transaction, consolidating is worth it.

One further problem of order attaches to running two. The side added later ends up carrying the continuing receipt. So the longest-lasting relations ride on the system chosen last. The side deserving the most time in selection becomes the side chosen in the most haste.

Can You Write One Line on Each of the Four Axes

Whether the criteria are in place can be read through three confirmations.

The first confirmation is whether, for each of the four axes, you can write your own requirement in a line. Within how many days the money is needed. In what cases refunds are accepted. Whether continuing receipt is required. Where the ceiling on cost sits. Any axis you cannot write is not yet settled.

An unsettled axis cannot be chosen on by comparing. Unable to choose, the decision falls to the ranking on a comparison table. Which is to say finishing this confirmation first is what makes selection possible.

The second confirmation is actually looking for the answers on the four axes for candidate operators. Whether the search finds them is itself information. The deposit cycle and the refund procedure are items operators often publish. Where they are hard to find, see whether enquiring produces an answer.

The third confirmation is counting how many transactions in your present trading failed to be collected. A large number means payments will resolve part of it. Zero means the reason for setting up is not collection but changing the form of provision. Which of the two it is changes which axis takes priority.

None of the three looks at fees. The criteria for judgement are not properties of cost, so what deserves checking is the side of terms.

The three differ in how far they tell. The first matters, and it is also hard. Writing requirements presupposes the form of provision being settled. At a stage where it is not, look only at the third of the four axes. Whether receipt ends at one occasion or recurs. Secure that alone and options remain when the form of provision is settled.

With the form of provision unsettled, setting up payments can simply be deferred. Screening can take time, though. And where screening fails, time is needed to find an alternative. If payments are unusable at the point provision begins, transactions in that span get processed by other means. The value of applying early can exceed the difference in fees.

Neither operator names nor a tolerable rate nor a number of days for the deposit cycle is written here. Not because they cannot be settled, but because settling them misleads. Each is one value on one of the four axes, and takes on meaning only once the other three are settled. Fix one value first and the remaining three get chosen to suit it, standing the order on its head.

Whether it can be done, and on what basis to choose. While the former, already answered, was doing the pulling, the latter did not so much as have the shape of a criterion. That holding payments is holding the right to define terms. Why fees converge on similar levels structurally. That the axes are four, examined in order of how hard they are to change. Only the third of these carries an order, so the actual work of choosing begins there.

Of the four, separate now the ones you can write in a line from the ones you cannot. Where an axis can be written, candidates can be narrowed from it. An axis that cannot is homework left not on the payment side but on the business side.

The actual procedure of setting up, and how far can be run without hands, is handled in payment setup for a sole trader. The thinking on the machinery of receipt as a whole lies above, in systemising a solo business.