How we test cars at What Car?

What Car? has been testing cars for 50 years. And our reviews are the most comprehensive around, because we test new models back to back with their key rivals. This allows our experienced team of reviewers to make direct comparisons.

Driving and efficiency

In addition to driving the cars for hundreds of miles on public roads, we take them to our dedicated test track – HORIBA MIRA, in Warwickshire – where we carry out performance, handling, noise and braking tests.

We find out how efficient petrol, diesel and hybrid models really are by putting them through our Real MPG test; this is done in a laboratory to ensure accuracy, but is based on a real route, that takes in motorway, urban and rural driving.

And we drive electric cars until their batteries run flat, in both summer and winter conditions, to find out how far they'll really go on a charge.

Practicality and costs

We take all of our own interior measurements, so we can tell you exactly how much space a car has compared with its rivals.

But it's not just about what the tape measure says; we also consider how comfortable the seats are, how many suitcases will fit in the boot, and how classy and user-friendly the interiors are.

Then we look at how much a car is going to cost you over three years, whether you're buying outright, taking out PCP finance, leasing or choosing your next company car.

And we consider how reliable it's likely to be, based on the results of our annual What Car? Reliability Survey.

That's the brief version though, but dozens of individual tests combine to deliver our definitive verdict. Below we reveal all.

We rate cars in up to 18 different areas:

• Performance

• Ride comfort

• Handling

• Noise and vibration (refinement)

• Range (EVs and plug-in hybrids only)

• Driving position

• Visibility

• Infotainment

• Quality

• Front space

• Rear space

• Seating flexibility

• Boot space

• Costs

• Equipment

• Reliability

• Safety and security

• Charging (EVs and plug-in hybrids only)

How we test cars - driving

Performance

To measure acceleration, top speed and braking performance, we use highly sophisticated Racelogic GPS data logging equipment. This ensures far greater accuracy than would be possible if using a stopwatch or even the GPS receiver in a smartphone.

We record the following data when performance testing a car at our HORIBA MIRA proving ground:

• Standing start acceleration times: 0-10, 0-20, 0-30, 0-40, 0-50, 0-60, 0-70, 0-80, 0-90, 0-100mph

• Standing quarter-mile time (sports cars, sports SUVs, performance cars and hot hatches only) 

• Standing kilometre time (sports cars, sports SUVs, performance cars and hot hatches only) 

For cars with a manual gearbox, we also measure rolling ‘in gear’ acceleration. This involves recording the time it takes for the car to accelerate from 30-50mph, and from 50-70mph, when left in third, fourth, fifth, sixth and (on some models) even seventh gear. This test is run twice (in opposite directions) because wind direction can have a noticeable effect on a car's ability to build speed from low revs, particularly in the higher gears.

For models with an automatic gearbox that doesn’t have a fully manual mode, we instead measure ‘kickdown’ acceleration times from 30-50mph, and from 50-70mph. This is the sort of acceleration you might need to pass a cyclist or slower-moving vehicle on a country road, or when joining a motorway from a slip road.

We measure braking performance by recording the distance (in metres) a car takes to stop from both 30mph and 70mph in a simulated emergency. 

Finally, we measure the top speed (in mph) of any car that can’t officially do more than 100mph. We do this by accelerating until the vehicle will no longer gain speed, and then completing a full (two-mile) lap of our high-speed test circuit. The average speed on this lap is recorded as the top speed.

Any car taking part in a What Car? comparison test will have the above performance tests carried out – even if we’ve tested an identical model on a previous occasion. This is because weather conditions have a big impact on performance, so we always conduct comparisons on the same day and in the same conditions to ensure a level playing field.

While the objective data we record is crucial for comparing the performance of different makes and models, our testers also consider subjective factors. How smooth and predictable is the acceleration? Is there a delay between pressing the accelerator pedal and the car responding? 

Our performance rating also considers fitness for purpose. So, for example, while a high-revving engine that delivers rapid acceleration beyond 6000rpm would be appropriate in a sports car, it wouldn’t be ideal for a family SUV. For the latter, an engine with more torque (low-rev pull) that delivers smoother, more relaxing acceleration would be far more suitable.

Ride comfort

You might forgive a rock hard, uncomfortable ride if you’re in the market for a new sports car to take on occasional weekend blasts. For most car buyers, though, how well the suspension isolates you from lumps and bumps in the road is an important consideration.

We test cars on a variety of surfaces and at different speeds – on the public road and at our private testing centre in MIRA, Warwickshire – so that we can tell you how comfortable a car is compared to its rivals.

You might imagine that soft and spongy suspension is all you need for an agreeable ride – but this isn’t always the case. Cars that exhibit exaggerated vertical or side-to-side body movements may not rattle your teeth out, but they can make you (or more likely your passengers) feel queasy. 

On the other hand, overly stiff suspension can make your daily commute to work really quite unpleasant. 

Our testers compare a car’s ride comfort not only with that of its direct rivals, but also different versions of the same make and model. This is because different suspension setups (sometimes fitted to sportier trim levels or available as an option) can dramatically affect a car’s ride. 

Likewise, the size of wheels you choose can also influence comfort. It’s usually the case that the smallest available rims help deliver the smoothest ride, although this isn’t always true. 

Our score for ride comfort is based largely on what we consider the best wheel and suspension combination available on a car. However, if this combo pushes the price up considerably, the score is reduced.

Handling

Handling concerns a car’s ability to change direction. However, it isn’t purely about how quickly it can go around a corner; it’s about how stable it feels when doing so, and how much confidence and enjoyment it gives the driver.

Our testers consider the amount of body roll, or lean, a car exhibits when being asked to change direction, how well the tyres grip the road in different conditions, and how the car responds when the limits of grip are reached. For example, a car that suddenly and unpredictably loses grip at the rear is usually harder to control, and potentially more dangerous, than one with a more neutral handling balance and no inherent vices.

Steering is another important factor we consider when rating a car’s handling. While some steering characteristics, such as a good sense of connection with the front wheels and accurate, predictable responses to inputs, are universally positive, others depend on the type of car we’re reviewing. For instance, a quick, sharp steering setup is often appropriate for a sports car or hot hatch, but wouldn’t necessarily be suitable for a luxury car with softer suspension.

We also consider how manoeuvrable a car is in tight confines. A tight turning circle and light steering can make a car much easier to park and less stressful to drive in urban areas. This attribute obviously carries more weight in a small hatchback designed for the city than it does in a sports car.

For sports cars, sports SUVs, performance cars and hot hatches, we also record lap times around our twisty 0.9-mile test track. This provides us with objective data on handling ability to go with our testers’ subjective impressions.

Noise and vibration (refinement)

How good a job your car does at isolating you from noise and vibration can dramatically improve your experience of living with it.

Noise and vibration can come from a variety of sources, including the engine (or electric motor in an EV), the wind as it rushes over the bodywork, and the tyres as they slap away at the surface of the road.

Using a sound meter, we measure the sound intensity in decibels that a car generates at a steady 30mph and 70mph. We do this at our testing centre so that other traffic doesn’t interfere with the readings.

However, our tests can still be heavily influenced by the weather (particularly wind direction and speed), so for all cars featuring in a What Car? comparison test, we take decibel readings on the same day in the same conditions, to ensure fairness. 

We don’t take noise readings when it’s raining, because the sound of water droplets hitting a car’s windscreen drowns out other noises.

As well as objective sound measurements, our testers also carry out subjective tests to isolate where any noise or vibration is coming from, and under what conditions it might present itself (i.e. when accelerating from low revs).

How smooth and easy a car is to drive is also factored into our noise and vibration score. In models with a manual gearbox, we consider how feelsome the clutch action is, and how positive and slick the gearbox is. For automatics, we rate how smooth (or clunky) the shifts are.

We also consider how easy it is to judge how much pressure you need to apply to the brake pedal to slow your progress smoothly. Some electric cars, for example, have overly grabby brakes, which counts against their overall noise and vibration score.

Range (electric cars and plug-in hybrids only)

Our research shows that how many miles an electric car can travel between charges is an important consideration for buyers. That’s no wonder when you consider that the UK’s public charging infrastructure is patchy in its coverage, and that charging an EV takes a lot longer than filling a petrol car with fuel.

Sadly, the official (WLTP) range figures published by electric car makers can’t be relied upon. They represent a ‘best case’ scenario, and indeed in many cases simply can’t be achieved in real-world driving – even in optimum conditions.

For every electric car that’s included in a What Car? comparison test, we measure its efficiency (in miles per kWh) when driven on a specially devised route at our testing centre. The route is designed to simulate motorway, rural and town driving – but without the variables of traffic conditions that we’d have to deal with if testing on the public road.

The car is driven repeatedly around our test route in convoy with its rivals, with driver changes and a switch in running order at the end of each lap to ensure a level playing field. Each car starts the test with an indicated battery level of 90% and with the climate control system set to 21deg C.

We then use the car’s efficiency reading in combination with its usable battery size to calculate the theoretical maximum range. This gives a far more representative and trustworthy range than the official WLTP figures suggest is possible. However, it’s important to note that our real-world range figures still aren’t necessarily what you’ll achieve. That’s because your commute might involve a smaller or greater percentage of motorway driving than our test does, and because weather conditions have a huge impact on the mileage any EV can manage. 

In cold weather, the maximum range between charges of an EV will always be significantly shorter than in the warmer summer months.

Twice a year – in the height of summer and depths of winter – we conduct an even more scientific range test on a selection of the latest fully electric cars. This involves running cars from full to completely flat, rather than extrapolating from efficiency figures. 

Our process when testing the real-world electric range of any plug-in hybrid (PHEV) that features in a What Car? comparison test is very similar. We start with a fully charged battery, select EV mode (to force the car to run on electric power), and repeat our test route until the battery runs flat and the petrol (or diesel) engine fires up. The distance achieved before this happens is recorded as the real-world range,

Any pure petrol or diesel car can be refilled in a matter of seconds, so the range between is only judged if a car has a particularly small fuel tank.

How we test cars - interior

Driving position

There are some fundamentals of a good driving position. The steering wheel, seat and pedals need to line up neatly with one another, and there should be enough adjustment in the wheel and seat to allow drivers of all shapes and sizes to get comfortable and reach the major controls without stretching.

Cars that have limited or no steering adjustment are downgraded, as are cars without adjustable lumbar support for the seat – particularly if lower back support is inadequate to begin with. Other aspects of seat support are also considered, including for the driver’s thighs, shoulder and sides, not only when going in a straight line but around corners, too.

Although we don’t award extra marks for electric seat adjustment in this area (that’s a convenience feature considered in the equipment section), we do mark cars down if adjusting the seat manually is particularly fiddly or time consuming.

Our research shows that SUV buyers like to sit high up, so we use laser technology to measure the height (in mm) of the base of the driver’s seat from the road. A higher seating position doesn’t necessarily mean we award a better score, but if an SUV has a driving position almost as low as that of an equivalent conventional hatchback, this will be factored into our rating.

We also consider the ergonomic layout of the dashboard, including how easy it is to adjust the air-conditioning system. If you need to use the touchscreen (or touch-sensitive pads) to change the interior temperature, this is generally more distracting than if there are physical controls. However, there are differences in the effectiveness of each approach, something our testers carry out subjective tests to determine.

Visibility

Being able to see out of a car, either directly with your eyes or via digital aids, is obviously crucial to your ability to drive it safely and confidently.

That’s why our testers assess the view out from behind the steering wheel in a range of different driving positions. Overly thick or sharply angled front pillars, a high window line or chunky rear pillars will reduce a car’s score in this area, as will big blindspots along the sides when the driver looks back over his or her shoulders.

If blindspot monitoring cameras are fitted as standard, or available as an option, this will be factored into our visibility star rating. We also consider the positioning and size of the door and rear-view mirrors.

Parking sensors at the front, rear and sides of the car will all increase its score for visibility – particularly if these aids are fitted as standard. Further marks are awarded for rear-view and 360deg parking cameras.

Infotainment

A car’s infotainment system is there to provide both entertainment and added convenience. Modern cars are being crammed with an ever increasing amount of tech to help, although this can bring as many challenges as it does benefits.

On one hand, more gadgets – from smartphone mirroring (Android Auto and Apple CarPlay), satellite navigation and even apps you can record video on – can make your drive easier, more enjoyable and sometimes even safer. However, interacting with these systems can also take your focus away from actually driving the car, which in turn can increase your chance of having an accident.

That’s why our infotainment scoring system takes into account not only the number of gadgets a car is available with, but also how easy the system is to interact with. As a general rule, systems that rely solely on a touchscreen score lower marks than those with a separate rotary or touchpad controller. Not all touchscreens score the same, though; our testers carry out objective and subjective tests to find out how distracting the systems are.

We’ve measured the time it takes to complete certain set tasks on a number of different infotainment systems, but we also consider the design and ease of use of every operating system, the quality of graphics on the screen, how effective any voice control function is and how good the audio quality from the sound system is. 

All of these aspects form part of our overall infotainment star rating which, as is the case with all What Car? ratings, is relative to a car’s direct class rivals. 

Quality

Two basic factors are considered when awarding marks for interior quality: perceived quality and build quality.

Perceived quality relates to the tactile nature of the materials inside the car, including the use of premium finishes – particularly on surfaces that occupants’ hands come into contact with regularly.

Extra marks are awarded for soft-touch plastic on the dashboard, and for leather (artificial or genuine) on the steering wheel. The action of the buttons and switches on the dashboard, including how well damped they feel when used, is also factored into our quality score.

Build quality, on the other hand, is how robust and well-assembled the interior feels. How sturdy are the fixtures and fittings; do they flex or wobble in normal use? If so, the car will receive fewer marks in this area.

We don’t award marks for interior style, on the basis that this is a subjective consideration and ultimately comes down to taste. However, the design of the interior from an ergonomic perspective is considered in our ‘driving position’ assessment.

How we test cars - space and practicality

Front space

We take our own measurements of interior space to ensure comparisons between different makes and models are direct and fair.

This starts with front space, where we first position the driver’s seat as low as it will go and adjust the seatback angle to 90deg (relative to the seat base). We then slide the seat as far back as possible and measure the distance (in millimetres) between the middle of the lumbar region (100mm above the seat base) to the centre of the brake pedal. This gives us the car’s maximum potential leg room.

Once this figure is recorded, we then slide the seat forward until the distance between the lumbar region and the centre of the brake pedal is exactly 1000mm. This provides a basis for all further measurements.

Next we measure the distance between the seat base and the ceiling (head room) and the width of the interior in front of the seatbacks (shoulder room). 

As well as the above objective measurements, our testers also carry out subjective tests to evaluate front space, considering factors such as the encroachment of any fixtures and fittings.

The amount of storage space in the front (including the size of the door pockets and glovebox and the number of cupholders) is also considered in the star rating we award for front space.

Rear space

In a four or five-seat car, our test team measures the amount of leg, head and shoulder room (in millimetres) in the seat directly behind the driver. Before doing this, we ensure the driver’s seat is in the same position as it is when we measure front space (1000mm from the brake pedal).

If the rear seats slide or recline, we first set them as far back as possible with the backrest angle as close to 90deg (to the seat base) as possible.

In seven and eight-seat cars, we repeat the above steps to measure available space for third-row passengers.

As well as the above objective measurements, our testers also carry out subjective tests to evaluate seating position comfort in all rear seats, ease of ingress and egress and the view out. The amount of storage space in the rear (including cupholders and door pockets) is also considered in the star rating we award for rear space.

Seating flexibility

One of the first things you’ll probably consider when buying a car is how many seats it has, but how flexible and configurable those seats are can make a huge difference to your ownership experience.

Even budget family hatchbacks almost always have folding rear seats, but does the seatback split 60:40 or in a more flexible 40:20:40 configuration? And is there a ‘ski hatch’ to allow you to poke long items between two rear passengers? The greater the flexibility, the higher the score.

We award extra marks for cars that have sliding or reclining second (or third) row seats, and we consider how easy it is to raise all of those seats into position and stow them away again when they aren’t needed. 

Extra marks are given for models that allow you to remove the rear seats from the car entirely, although how easy (or otherwise) this process is will be factored into the score.

Boot space

Car manufacturers quote boot capacity in litres, which can be useful when making comparisons between different makes and models. However, one number is of limited value because it doesn’t tell you anything about the space. Is the boot deep? Shallow? Long? Short? Wide? Narrow?

That’s why our testers measure every car’s boot in the same prescription fashion, so we can tell you exactly how it compares with its rivals’. We take the following boot measurements in millimetres:

• Width (minimum and maximum)

• Minimum height (boot floor to parcel shelf)

• Maximum height (boot floor to ceiling)

• Minimum length (from tailgate to rear seatbacks) 

• Maximum length (from tailgate to front seatbacks, with rear seats folded)

• Drop from boot entrance to boot floor

• Height of boot entrance from ground level

We don’t only rely on measurements, though. We try various everyday items in the boot to see if they will fit, including golf clubs and pushchairs. We also see how many carry-on suitcases fit into the boot below the parcel shelf (or load cover), to help you relate to how much luggage space a car has.

Other things we consider in our overall boot space star rating include:

• How easy is it to fold down the rear seats? 

• Is there any underfloor storage and how useful is it?

• Is there a front boot and how big is it? (electric cars only)

• Is a powered tailgate fitted as standard or available as an option?

• Other useful standard and optional features, including boot hooks, netting etc

How we test cars - buying and owning

Costs

Buy any new car and you’ll either pay a lump sum up front or sign up to a monthly repayment. There are many more bills that feed into the overall cost of ownership, though; that’s why we crunch the numbers for you and consider all of the following:

• Purchase price 

• Available discounts (via What Car? New Car Buying)

• Monthly PCP finance cost

• Insurance premiums

• Servicing costs

• Fuel (or electricity) costs

• Depreciation

• VED tax

We don’t forget about company car drivers or leasing customers, either. Our overall score factors in benefit-in-kind (BIK) tax costs, and monthly contract hire fees. 

Equipment

In this section, we rate cars based on how much luxury equipment they come with. This excludes visibility aids and infotainment gadgets, which are considered separately in our visibility and infotainment sections.

So, our equipment star rating factors in convenience aids, including keyless entry and adaptive cruise control, as well as more frivolous features, such as massaging seats or a sunroof.

A higher score will be awarded if a feature comes as standard across the range than if you have to pay extra for it, or if it’s available only with the most expensive trim levels. 

As always, our equipment star rating is based on how well a car compares with its direct rivals. So, a £20,000 small car would inevitably need less equipment than a £100,000 luxury SUV to gain our maximum five-star rating.

Reliability

The annual What Car? Reliability Survey rates the dependability of cars aged up to five years old, based on the experiences of around 25,000 owners. We use the data from this survey to score cars based on how reliable they are compared to direct class rivals. 

In instances where an individual model hasn’t appeared in the latest survey (either because it’s too new or not enough responses were received) our reliability star rating is based on the average performance of the manufacturer in the overall brand league table.

However, if the reliability of the manufacturer is rated as below average (one or two stars), we give the benefit of the doubt and award the car an average (three star) rating until more specific data is available.

Safety and security

Safety is arguably the most crucial attribute of any car because it could, quite literally, be the difference between life and death. And while the vast majority of new cars sold in the UK are required to meet certain basic safety standards, there’s still a huge variation between different makes and models.

The independent safety experts at Euro NCAP conduct numerous tests to determine not only how well a car is likely to protect you and your passengers in the event of a crash, but also how capable the car is at helping you avoid an accident in the first place. The results of these tests are factored into our overall safety and security score.

Euro NCAP assesses crash protection by performing various simulated crash tests, including frontal and lateral impacts. The forces exerted on dummies inside the car are measured and this data is used to determine the likely protection of adult and child occupants. The effectiveness of passive safety aids, such as airbags and seatbelts, can improve a car’s ability to protect its occupants, but the actual structure of the car and how it deforms in a collision is just as important. 

Further simulated crash tests are carried out to assess pedestrian and other vulnerable road user safety. These results are also factored into our overall safety and security score.

Meanwhile, the car’s ability to avoid or minimise the severity of an accident (also known as active safety) is scored using a combination of Euro NCAP’s tests and our own in-house tests on assisted driving technology. We also consider the amount of active safety aids available on different trim levels, awarding lower marks if you have to pay extra for the technology than if it’s fitted as standard.

In this section, we also consider how secure a car is against thieves. Thatcham Research carries out rigorous independent tests to determine how easy a car is to break into and steal, and we incorporate scores into our safety and security rating. Cars with an alarm and an immobiliser inevitably score higher marks.

Charging (electric cars and plug-in hybrids only)

How quickly you can charge up your electric car or plug-in hybrid (PHEV) can be just as important as its range. That’s why we rate cars based on how quickly and easily they can draw power from the grid. The higher the charging rate, the faster you’ll be able to increase your potential range.

We also consider the type of charging socket the car is fitted with. CCS has become the European standard for rapid charging, and the vast majority of new EVs come with one of these as standard. However, some are fitted with the less common CHAdeMO socket, while others don’t have a rapid charging port at all. Fully electric cars without a CCS point will receive a lower score for charging.

Some PHEVs have CCS rapid charging ports, too, but most make do with a regular (Type 2) connector. Even those with the latter offer different charging speeds, though, with many able to pull only 3.7kW – barely more than you’d get if plugging into a domestic three-point socket.

Tesla is currently the only manufacturer with a proprietary charging network which (apart from a handful of exceptions) isn’t available to drivers of other electric cars. Called the Supercharger network, it’s fast, reliable and easy to use, and the relatively high number of charging points at most locations means you rarely have to wait to plug in. 

The Supercharger network is one of the most compelling reasons to choose a Tesla over another brand of electric car and, for this reason, we give all Tesla models our top five-star rating for charging.

How we reach our verdicts

The overall star rating we award a car isn’t simply an average of its individual category scores. That’s because our research tells us that buyers of different types of car have different priorities.

For example, the most important considerations for the average sports car buyer are performance and handling, with boot space and seating flexibility much less important attributes. Meanwhile, the average small car buyer rates costs and equipment as the most crucial attributes, with infotainment and visibility way down the list of priorities.

That’s why we weight the individual category scores differently depending on the class of car. Cars that achieve high scores in the key areas are far more likely to achieve a high overall star rating.

What our star ratings mean

You’ll see a What Car? star rating, from one to five, at the bottom of every first drive and review, as well as for all cars featured in our group tests.

These ratings show our verdicts at a glance, and we make use of each level of our five-point scale to highlight the differences between the best and worst cars in any class. The summaries below tell you how to interpret our star ratings.

Rated 5 out of 5 – Outstanding

Cars that receive What Car?'s five-star rating must be outstanding in a number of key areas. This means they beat their rivals at the things that matter most (ie. practicality and comfort for an MPV, performance and handling for a sports car), without disappointing in the less crucial areas. These class-leading cars shouldn't just be on your shortlist – they should be right at the top.

Rated 4 out of 5 – Above average

Four-star cars are still excellent choices, and should be high on your shortlist of models to test drive. They may not be quite as good as the class leaders in some key areas, or may be slightly more expensive than the best rivals in a particular class. However, they won’t have any major shortcomings.

Rated 3 out of 5 – Average

Our three-star rating is far from being a mark of condemnation. These cars may either be par for the course in most key areas, or may impress in some and disappoint in others. Models that receive this rating are rated average in their class – rather than above average or outstanding – but are certainly worth considering if you’ve discounted rivals with higher star ratings.

Rated 2 out of 5 – Below average

Cars that get a two-star rating are a long way off the standard set by the class leaders. These disappoint in too many key areas to be considered average. They may cost significantly more than rivals to run over three years, or have compromises in interior quality, comfort or refinement. We think these cars are best left off your shortlist.

Rated 1 out of 5 – Poor

One-star cars are rare. To be rated as poor in its class, a car will have serious shortcomings in a number of key areas. It may be very poor to drive, cramped inside, be poor value for money, or depreciate far more quickly than its rivals – or all of these things. One-star cars fall well short of the standards we expect, and as a result are impossible to recommend.

Re-evaluating verdicts

Once we've awarded a car its star ratings, the process doesn't stop there. Each time a new car joins a class, we revisit the ratings of every rival to make sure they're still valid.

Plus, it isn't just new car reviews you can find on What Car?; our used buying guides include everything from how much you should pay, to what goes wrong.

In short, our verdicts are the fairest and toughest in the business, which is why more than two million people visit whatcar.com every month.

Our new car test team

Reviews editor Will Nightingale
Deputy reviews editor Neil Winn
New cars editor Lawrence Cheung
Head of video Doug Revolta
Reviewer Dan Jones

Our used car test team

Used cars editor Mark Pearson
Used cars reporter Oliver Young

For all the latest reviews, advice and new car deals, sign up to the What Car? newsletter here

See our latest reviews >>