Hitting the snooze button? Trying to understand what the COVID lockdown has done to our sleep patterns

Differences of opinion

I wonder how many of you have had this experience. On a mid-morning Zoom call the other day one of my colleagues yawned very visibly and audibly. There were some jokey comments asking whether we were keeping the individual in question up and he, somewhat sheepishly, apologised saying he hadn’t been sleeping too well lately. This prompted a few minutes of quite animated conversation with some agreeing with the yawner and others saying (rather smugly) that actually they had been sleeping better than ever in lockdown.

Intrigued, I started to dig. A quick Google around the usual news and commentary sites suggested that the balance of opinion was with Mr Sleepy: there is, indeed, lots of discussion (such as this piece from KCL) around the idea that COVID is negatively impacting our sleep patterns. But, again, you can find the contrary view easily enough.

Understanding the “truth”: our survey says……………..

Much of the commentary about COVID-driven sleeplessness derives from surveys of the population. So, for example:

  • A SleepCouncil survey in April found that close to half of the respondents (43%) were finding it harder to fall asleep, with unease around the current situation affecting sleep for three quarters of people (75%). Over three quarters (77%) said that lack of sleep was interfering with their ability to function in the day (daytime fatigue, concentration, mood)
  • A new study of the UK public by King’s College London and Ipsos MORI has found that many have experienced changes to their sleep patterns since lockdown, with nearly two-thirds (63%) overall saying their sleep has been worse, half (50%) that their sleep has been more disturbed and two in five (39%) that they’ve been sleeping fewer hours a night on average.
  • The lockdown triggered a sharp increase in anxiety-related sleeping problems, with mothers, key workers and people from minority ethnic backgrounds the worst affected, according to a study by the Economic and Social Research Council-funded Centre for Population Change.
  • An international group of researchers led by Professor Colin Espie is has set up the International COVID-19 Sleep Study (ICOSS) to investigate the impact of the COVID-19 pandemic on sleep and daily rhythms in adults.

Surveys can give us great insights into the attitudes and behaviours of the population. But, we’d be remiss not to recognise their potential flaws: people may or may not have an accurate self-perception of how they are sleeping; they may be able to sense the “agenda” behind the questioning and so determine that they are happy to play to that agenda, or to undermine it, and they may be undertaken by those (such as mattress manufacturers) with a vested interest in proving the argument one way or the other. Hence it is important to try to validate survey answers with inputs of other types.


Everybody lies, so what’s the truth?

Understanding what the Government was/is asking us to do is not enough. For many, they need to also to understand why the specific policy guidelines have been designed as they have, though there is also, of course, a danger of over-communication that can cause exhaustion amongst the audience and lead to messages being disregarded.

“People’s search for information is, in itself, information. When and where they search for facts, quote,  jokes, places, persons, things, or help, it turns out, can tell us a lot more about what they really think, really desire, really fear and really do than anyone might have guessed”

Seth Stephens-Davidowitz in the introduction to “Everybody Lies: What the internet can tell us about who we really are”

So, I’ve looked at some of the Google searches around sleep. There is a clear downward trend from the start of the year which was suddenly reversed as we went into lockdown. Searches then fluctuated, though do seem to oscillate around a higher average than pre-lockdown, with three noticeable peaks:

  • March 27th, just four days into lockdown, no doubt as the seriousness and likely duration of the situation was staring to hit home for many, although the release of a collaborative studio album with this name by British rappers Skepta, Chip and Young Adz is probably also a contributor to this spike (and, by the way, showcases some of the difficulties in interpreting search data)!
  • June 26th – the likely cause for this being the hot weather of that time (Meteoguru). We were in the midst of a mini heatwave with several days of temperatures in the high twenties/low thirties and steamy, sticky nights
  • September 7th – a lower peak, most likely attributed to worries about the return of pupils to UK schools

My take on this is that there are indeed more sleep searches in lockdown than there were immediately preceding it, though it is too simplistic to attribute sleeplessness to COVID alone. In fact, there are likely to be multiple reasons for insomnia, including some or all of the following, which have all been affected in some way by lockdown:

  • COVID worries on behalf of ourselves, our nearest and dearest (particularly elderly or otherwise vulnerable relatives) and society as a whole
  • Economic worries, such as short-term concerns like job losses, paying the bills and longer-term anxiety over the future
  • Worries about the “return to normality” (to work, school or commuting, etc)
  • Disrupted daily patterns caused by the shift to WFH
  • The weather, particularly here in the UK where few homes have AC
  • Lack of exercise
  • Poor diet and drinking too much alcohol


Looking through the Twitter-shaped window

Another way of investigating this is via social media listening. Services like Meltwater, Crimson Hexagon and Brandwatch allow users to see what people are posting and commenting on via platforms such as Twitter, Instagram and Facebook.

The chart below shows that the daily average volume of mentions is not hugely different before and after lockdown, though it is up by nearly 10%. In addition, post-lockdown did see bigger spikes in the conversation around insomnia suggesting perhaps that people were more highly sensitised to the issue of sleep and when problems did occur, they were very quick to shout about it online.


Gathering real-time data

If only we had some way of knowing how long people actually sleep, rather than just how long they say they do. Well, it turns out that we do, for some of the population at least, in the form of wearables like Apple Watch, Fitbits and Garmins, which are owned by some 12% of the UK population.  Clearly, getting hold of all this data is virtually impossible, except for the owners of those devices and the companies that collate all the data they collect.

Still, I’ve had a go, using my own Fitbit data as input.  I’ve taken the March 23rd lockdown as the origin and divided the periods before and after that into “months” – not calendar months but going from the 23rd of one month to the 22nd of the next. The chart below shows minutes of REM, light, and deep sleep and the total of these three. While there are, of course month to month fluctuations, my data suggests I’m getting 20 minutes more sleep a night – comprised of roughly 40 more minutes per night of light sleep and roughly 10 minutes less per night of both REM/dream sleep and deep sleep.

Actually, Fitbit themselves reckon that my data is fairly typical and matches the data they have showing that the duration and quality of many people’s sleep has improved during the coronavirus crisis, something they attribute to the fact that many of us no longer need to commute.

They found that Fitbit users in six U.S. cities—San Francisco, Los Angeles, Houston, Chicago, New York, and Phoenix—were getting 17 minutes more sleep per night in April than they were in January, with 36% of them people getting an additional 30 minutes of sleep or more compared to before lockdown.


Some observations

Not one answer but many

Clearly, the limited investigations summarised here show that the perspective that you gain when looking into an issue can vary greatly depending on what data sources you use to investigate: survey responses are different from sales figures; social media posts are different from search histories. Make sure you choose a data source appropriate to the investigation you are making and ideally assess the insights across multiple data sources.

Most people are sleeping worse

As with so many aspects of our life, COVID is causing different impacts on sleep for different people. Some, who may be enjoying their commute-free WFH existence may be gabbing more ZZZZs, while others, beset by health and financial worries are finding just the opposite. Make sure you understand which of these boats your typical consumers find themselves in. While the Fitbit data is compelling, we need to recognise that Fitbit wearers are likely to be more exercise-conscious and likely more middle-class than average. It is quite possible, then, that their findings speak only to a relatively small niche.

Perception versus reality

Even if people are not losing sleep to COVID, if they feel as if they are, then they will act to improve their sleep.

Opportunities for brands

A few years back, futurist Faith Popcorn described sleep as “the new sex”. In her view, shared by many, we were suffering a major decline in sleep quantity and quality. ds that rely on in-person engagement with their customers. No matter how severely following the rules impacts on the bottom-line there can be no justification for compromising the wellbeing of customers or staff. McKinsey & Co. have estimated the sleep-health industry to be worth between $30 billion and $40 billion and that the sector has grown by more than 8% per year. With so much money at play, it is not surprising that innovation in this space is rife and has been accelerated by COVID. Here are just some examples:

Eating right

  • Dream Water SleepStat Natural Blend is a combination of three active ingredients: Melatonin, 5-HTP, and GABA to aid in sleep
  • Nightfood “sleep-firendly ice-cream”
  • Pepsi have announced the launch of Driftwell, a calorie- and sugar-free noncarbonated water that sprouted from an employee incubator program. Driftwell is flavoured with blackberry and lavender and contains 200 milligrams of L-theanine and 10% of the daily value of magnesium to aid  de-stressing and relaxation.

Turn off that device

Leverage technology

  • Nightingale Sleep Acoustics from Cambridge Sound Management plugs into the mains power supply to create a “sound blanket” that masks disrupting night-time noises
  • Beddr Sleep Tuner: feather light forehead-worn tracker that provides clinical grade sleep data using an optical sensor and accelerometer.
  • Apple acquired Beddit, a Finnish company that makes sleep-tracking devices as a precursor to significant focus on the sleep opportunity.

The importance of what you sleep on or in

  • Bearaby make weighted blankets that discourage the sleeper from tossing and turning too much
  • Buffy have developed a comforter made out of highly breathable eucalyptus fibres, which means this bedding allows heat from the body can escape
  • There are though signs that the mattress startup frenzy is starting to cool, because there are so many entrants competing for our sleep cash

Applying this to your challenges: Meet our Human Understanding Lab

At UNLIMITED we know that in order to deliver business advantage for clients, we must truly understand human decision making and behaviour. We do this through our deep understanding of human emotion, motivation, and action, and the wider contextual meaning that impacts internal processing.

The Human Understanding Lab informs, validates and fine-tunes effective solutions for our clients and consists of a tight-knit community of expert neuroscientists, research practitioners, trends analysts, data scientists and behavioural scientists. By blending data with modern research methods, we evaluate how humans interpret, perceive, and feel the world around them, to deliver actionable and measurable insights rooted in human behaviour.

This article was written by Nick Chiarelli, Head of Trends at Unlimited with input from Elaine Farrell of TMW UNLIMITED