Six checks on one specific account. The headline rate is the one everybody runs and the only one that can be withdrawn without warning - so it belongs fourth, not first.
12 min read · Level 1
Every comparison tool in this market sorts by rate, and sorting by rate is close to the worst possible default for a product whose rate is the only revisable term. The tools are not being dishonest - rate is the one field that is numeric, universal and easy to rank, and cover, access, early-exit cost and conditions are all text. So the interface optimises for what it can sort, and the customer inherits a decision procedure that was chosen by a database schema. What I would put in its place is unglamorous: eliminate before you rank. Most accounts on any given list are unsuitable for a specific pot of money for reasons that have nothing to do with price, and once the unsuitable ones are gone the price comparison is between two or three products and takes a minute. The version of this that I think is genuinely undervalued is accepting that the answer is often more than one account. There is a strong pull towards finding the best account and putting everything in it, and it produces worse outcomes than splitting money by the date it is needed and taking the best available rate within each group. That is not sophistication. It is just refusing to let a single number stand in for a decision that has four other dimensions, all of which are written down, and none of which anybody sorts by.
Six checks on one account, in the order that determines the outcome rather than the order the marketing presents them in.
One. Is it covered, and by what. Which licensed institution holds the money, which scheme covers it, what the limit is, and what you already hold with that licence. A rate is meaningless if the balance sits above the guarantee and you did not intend it to.
Two. What access you are actually agreeing to. Instant, a notice period of a stated number of days, a fixed term, or a limited number of withdrawals a year before the rate drops. These are four different products and they are frequently presented on one page under one heading.
Three. What it costs to get the money early. A number of days of interest forfeited, a fixed charge, or refusal. This is written in the terms and appears on no comparison table.
Four. The rate, and whether any of it expires. Here is where the arithmetic earns its place.
Account A 4.50% for twelve months, of which 2.00 points is an introductory bonus, reverting to 2.50% thereafter Account B 3.60%, no bonus, no expiry
On 20,000 over two years, with interest added annually:
Account A Year 1 20,000 x 0.0450 = 900.00 balance 20,900.00 Year 2 20,900 x 0.0250 = 522.50 balance 21,422.50 Interest over two years 1,422.50 Equivalent annual rate 3.4952%
Account B Year 1 20,000 x 0.0360 = 720.00 balance 20,720.00 Year 2 20,720 x 0.0360 = 745.92 balance 21,465.92 Interest over two years 1,465.92 Equivalent annual rate 3.6000%
Account B ends 43.42 ahead, on a headline rate 0.90 percentage points lower.
Five. The basis the rate is quoted on. The like-for-like annual equivalent figure, so two accounts compounding differently can be compared.
Six. What it takes to keep the rate. Some accounts require a minimum balance, a monthly deposit, a maximum number of withdrawals, or the account to be opened through a particular route. A condition you will not meet is a rate you will not receive.
Because the rate is the only one of those six that can change after you commit, and it is the one every comparison is organised around.
Run the order backwards and see what happens. Somebody who starts with the rate finds Account A at 4.50%, opens it, and discovers three things over the following year: the bonus expires, the reverted rate is poor, and moving again requires the same work they have already done once. Somebody who starts with cover and access has narrowed the list to accounts that suit the money before any rate is involved, and then picks the best rate among those - which is a shorter list and a better decision.
The introductory rate deserves the arithmetic it got above because it is the most reliably misread structure in retail cash. The headline is true, the bonus is genuinely paid, nothing is concealed, and the two-year outcome is still worse than a lower headline. What makes it work is that the comparison everybody runs is between two headline numbers at one moment, and the product is designed around the moment after that.
Notice how small the gap is in the worked case: 43.42 over two years on 20,000. It is not enough to matter on its own, and that is the point worth taking rather than the number. The bonus structure is not principally a way of paying you less - it is a way of acquiring a customer who will then not move, and the cost to the holder is not the 43.42. It is the years three, four and five at 2.50% while the market pays more, which is lesson three's gap arriving through a different door.
Access deserves its place at number two for a blunter reason. A rate you cannot get to is not a rate. Notice accounts in particular are routinely selected on rate by people who then discover that ninety days means ninety days, and the discovery happens at the moment the money is needed, which is the worst possible time to learn a product's defining feature.
And cover sits first because it is the only check that can produce an outcome no rate compensates for. Everything else on this list costs money. The first one can cost the money.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Four accounts, all real-looking, all offered to the same person on the same day. The exercise is to eliminate rather than to rank.
A 4.50% for 12 months then 2.50% instant access no minimum B 3.60% fixed for two years no access at all 10,000 minimum C 3.95% 90 days notice no minimum D 3.30% instant access loses the rate above 3 withdrawals a year
The money in question is 20,000, of which about 6,000 may be needed at short notice during the year and the rest has no date attached.
B is eliminated by check two, not by its rate. No access at all for two years does not suit money that might be needed in months, and no rate repairs that.
C is eliminated for the 6,000 and remains available for the rest. Ninety days notice is not instant access; it is a product for money with no near-term claim on it.
D survives only if three withdrawals a year is genuinely enough, which is a question about behaviour rather than about the account. Somebody who moves money frequently will breach it and lose the rate, which makes the advertised 3.30% a figure they will not receive.
A survives check two and fails check four, as the arithmetic showed - unless the holder will genuinely move again in twelve months. Some people will. For them the bonus is worth exactly what it says and the structure is a fair trade.
The outcome is not one account. It is 6,000 in something instant with the best rate available, and 14,000 in C at 3.95%, with a note in the calendar for A's reversion date if A was used at all.
That is the shape of a correct answer in this class, and it is worth noticing how unlike a rate comparison it looks. Nothing was ranked. Three of the four were eliminated by a property other than price, and the final decision involved two accounts rather than one because the money had two different jobs.
Starting with the rate. It is the only one of the six checks that can be revised after you commit, and organising the decision around it means the other five are answered by whichever account happened to have the best number that week.
Comparing a bonus rate with a standard rate over one year. Account A's 4.50% beat Account B's 3.60% on the headline and produced 43.42 less over two years, and the real cost was the years afterwards rather than the 43.42.
Reading notice as a formality. Ninety days notice means the money is unavailable for ninety days, and that is learned at the moment it is needed. A notice account is a fixed term with a rolling start date, and it should be chosen on the same basis.
Missing the condition attached to the rate. A minimum balance, a monthly deposit, a withdrawal cap or a specific application route. A condition you will not meet converts the advertised rate into a rate somebody else receives.
Checking cover last, or not at all. It is the only item on this list whose failure cannot be corrected by a better rate later, and it takes two minutes at the point of opening rather than an audit afterwards.
Putting all the money in one account because one account won the comparison. Money with different dates attached has different requirements, and the correct answer is frequently two accounts rather than the best one.
Which legal institution holds the money, which scheme covers it, and what you already hold under that same licence. Opening a well-rated account at a brand that shares a licence with your main bank can push a covered position over the limit without either balance changing.
Instant, notice of a stated number of days, fixed until a stated date, or conditional on a withdrawal count. Writing it out rather than skimming it is what prevents a notice account being remembered as an instant one, which is a genuinely common and entirely avoidable confusion.
A number of days of interest, a flat charge, or refusal. It differs substantially between products that look identical on a comparison page, and it is the only clause you will care about if circumstances change. If you cannot find it, that is the finding.
A 4.50% consisting of 2.50% plus a 2.00 point bonus for twelve months is two different rates wearing one number. Compute the outcome over the period you actually intend to hold rather than over the bonus period, as the two-year arithmetic above does, and diarise the reversion date on the day you open the account.
Minimum balance, monthly funding, withdrawal limits, application route. Then ask honestly whether you will meet them in a month when things are busy. An account whose rate depends on a discipline you do not have is an account paying its standard rate, and the standard rate is the one to compare.
Money with different dates has different requirements, and the right answer is often two or three accounts rather than the single best one. Splitting by job rather than chasing one rate also keeps each balance mapped to a licence, which makes the cover check in step one something you can maintain.
Before opening any savings account, when an introductory period is approaching its end, when a notice or fixed-term product is being considered for money that has no firm date, and at each scheduled rate review.
Do not run this appraisal on an account whose full terms you cannot read before applying, and do not use it to compare a deposit against a money market fund or a short-dated government instrument - those carry different protection and are a different comparison.
Introductory bonus rates expire on a known date, disclosed at the point of opening, and the expiry is almost never entered anywhere the holder will see it. The account continues, the balance does not fall, and the reduced rate arrives without any event to mark it - the correspondence, where it exists, reads as routine. Holders frequently discover the position months or years later during an unrelated review, and describe the outcome as the bank having quietly cut the rate, when the reversion had been stated in the terms they accepted and the date had been fixed from the first day. The information was complete, disclosed and correct, and it was still lost, because nothing in the arrangement prompts a person on the day it matters.
Diarise the reversion date at the moment you open the account. It is the one piece of information about a bonus rate that is certain from the start and the one that is reliably forgotten.
Create a free account to read the rest and apply it to your own investing.