Clients obsess about defining them (no, really, we do) and agencies wait on tenterhooks for them, briefs are the documents that keep the work flowing and money moving. And yet, many are awful. Here are four clues that you’ve written a bad brief.

You’ve not specified a budget

Now come on, you’ve got a number. Unless you’re just out wasting everyone’s time, you’ve got a budget so state it.

Client-agency relationships are often treated as a one way street. The client sets the terms, the agency does the work, plus they throw in a few friendly drinks now and again.

Wrong. This is all wrong. If you want the best from your agencies, you need to invest time to understand them, share information about your company and your own preferences, you need to agree ways of working, and you need to give honest and detailed feedback.

You can only do all of that if there is trust between you and the agency. And there won’t be any trust because right at the start, you won’t even share a number with a pound sign in front of it.

You might be thinking, “That’s all well and good, but they’ll just come back with a proposal that costs that amount instead of the right thing.” And that’s not an unreasonable retort. But it’s your job to look at the proposal, question it, firm up the numbers, really query that every element is necessary. That’s your job. And if you put out a brief with no budget and someone came in with something really cheap, you’d still have to have question it.

If you don’t critically assess proposals, you’re being negligent. So put the number in the brief.

You’ve answered the question

Have you explained what you’d like to achieve or have just written out what you’d like done? For example, if it’s a digital brief to boost brand awareness, have you said you want a video? Because then you’ve answered the question and you’re going to get a video plus some distribution and a few frilly bits. So really, you’ve written a brief for a video that you’d like to use to raise brand awareness. If you want to boost awareness, just say that you want that and you want it done through digital channels.

Leave it to those who are pitching to answer the question. It puts the onus on agencies to be more creative, as a client you’ll get a more varied range of options and you’re setting the types of challenges that agencies love; things that let them show they can go beyond service delivery and really add value.

Too many objectives

Want to raise brand awareness? Great. And drive up sales? Obvs. And improve perceptions of your environmental impact? Umm, OK, I guess. And build your chief exec’s profile? Hang on. And repair your tarnished employer brand? No.

I’m not sure I need to spell this out any clearer.

Cakeism

Do you want your brand to be really well known? Do you also want to avoid any form of public criticism by not taking a defined position on any issue? Do you see the problem here?

You want your cake and you want to eat it.

If you’ve written a brief strewn with caveats, check it again. Are they all necessary? Are you being over cautious? Are you asking for things that are contradictory? If you refuse to reconcile them, have you at least had the decency to flag up the contradictions?

There is nothing worse than an undeliverable brief.

Measurement is big and will only get bigger. The user data generated by digital channels allows us to cut website visits or social media interactions in a variety of ways.

At first the abundance of data was amazing. Then it became confusing. Now, if we’re honest, it’s become hugely convenient. It’s possible to put a positive spin on performance when reporting results to people who aren’t literate in the data of digital communications and marketing. It’s even easier when you’re presenting it to people who don’t understand data at all.

There are too many metrics and it’s too tempting to highlight only the positive ones. At some point, however, someone’s going to question why the reports are great but there’s little overall progress. It won’t add up. Even the totally innumerate will see it clearly.

Remove your hiding places. A clear objective should be measured by a single metric. Other metrics can play a support role, but give one primacy. If you do that, you’ve nowhere to hide. It’s liberating. It gives you focus. Your performance will step out of the shadows.

A business has objectives and a strategy. It then needs plans for delivery, plus good quality tactical execution and sensible measures to monitor progress. What it doesn’t need is a strategy at every level.

It doesn’t need an SEO strategy, content strategy, engagement strategy and social strategy sitting on top of one another. It doesn’t need endless reassessments of its position and where ought to place itself.

Often strategy is seen as synonymous with seniority or high fees. Or worse, it’s an excuse for endless insight, fudging choices and avoiding actually delivering something.

Too many strategies is worse than no strategy at all.

You’re at a conference, you’ve taken on a ton of information. Your brain’s trying to make sense of it all and then, very occasionally, a speaker steps up and just totally blows your mind. That happened to me recently when I heard Roeland Dietvorst, a neuroscientist who works with marketers to improve campaign outcomes.

His talk ranged from football playing bees, to monkeys’ innate sense of fairness, and from Kahneman’s systems 1 and 2, to MRI scans that can be used to draw the images you’re picturing in your mind (yes, really). All of it was amazing and I it’s been circling round in head ever since.

In practical terms, for digital comms and marketing there were some very pertinent points. Some of them intuitively feel right, but it’s worth formalising them.

Keeping people in flow

The immediate priority for me to is to refocus on flow. People are busy living their lives and do many everyday things with little thought or effort; they’re using Kahneman’s System 1, the automatic, easy, low power mode. When they’re scrolling through Facebook, they’re definitely in System 1.

Traditionally, marketers interrupt people with messages. We’re trying to create some space in their minds for our brand or product or service while they’re busy doing other things. For example, we drop TV ads or billboards into their day. Online, we break into their feeds with ads. Typically, we finish our messages with a call to action – we want them to do something (usually buy from us).

The immediate lesson isn’t to not interrupt people, it’s to not interrupt them in a way that makes them think. If they’re using System 1, don’t switch them to the harder work of System 2 (the slow, deep thinking process) because it means they’re concentrating harder, are more critical and less likely to buy.

For instance, if you’re selling winter tyres, don’t evoke images of the tyres saving people from accidents because you’re making people think of accidents. Better to go with lighter messages that don’t make people think critically, that don’t raise fears.

You’ll be thinking this is obvious, but in an age where marketers increasingly seek to shock people to get attention, it’s a timely lesson to learn.

Don’t overthink it

Our brains make choices faster than we realise. They make instant judgments. When asked to explain our choices, we post-rationalise them.

A good example of this is the performance improvement made by booking.com from a small, subtle change to its website. They switched black bullet points to green ticks. It led a significant uplift in conversion.

If you asked a typical customer, what made you go ahead and make your recent purchase, no one would say, “the green ticks really swayed me.” But they made an observable impact. Those ticks changed the choices people made.

The lesson here is not to over rationalise people’s choices. It is better to observe and optimise.

Test, test, test

Everyone doing anything digital will be testing and learning constantly. Dietvorst cannot endorse this enough. Don’t overthink through which pricing option or call to action will work best, run tests and see which does. Constantly improve performance through repeat and ongoing testing. And it doesn’t have to be the big things. The composition of an image, or the type of bullet point you use can lead to statistically significant uplifts in performance.

The trap many people fall into when they learn about things like heuristics and automatic preferences, is that they take this understanding of how our brains use shortcuts and then try to apply them as a general theory. Beyond understanding that our brains automatically make choices quickly and not rationally, nothing else should be taken for granted. Don’t a single observation and apply it generally out of context. Instead, test, test, test.

A note on ethics

Keeping people in flow, optimising to increase sales, testing and learning. None of that is controversial, but it does raise ethical questions. We know people don’t always make the best choices. If you’re in the business of selling something, even something relatively benign, I think we all have a responsibility to think about customer outcomes, not just selling more. So while optimising is important, there should be a parallel stream of work to protect and properly serve more vulnerable customers.