A Quick Primer On What Marketing Does

Most technology people (and many marketers) don’t really know what the Marketing department does, so I’ll summarise the big parts of what they’re for.


Firstly, they should do, and commission, research. You do research so that you can prove or disprove hypotheses about what your company should do.

Hypotheses include questions like: What is the industry structure we are in? How large is the market we are in? What do our customers want to buy? From us, and from other people? Do people like our company and its products? Why? Why not? How much are people willing to pay for our product? What would happen if we raised the price 2%? Dropped it by 3%? Changed the colour? Etc. Etc.

If you don’t do research, you can’t answer these questions, and that means you can’t do any of the other things marketers are supposed to do.

Research is fundamental. Without the data, you don’t know anything; you’re just making up stories and hoping you’re lucky.


Segmentation is splitting the overall market into segments. Well duh, but what is a segment?

A segment is a subset of the market where members of the segment have something (or several things) in common. Segments can be defined in various ways.

The best way to segment is psychographic which is a fancy term for the beliefs, attitudes, values, and behaviours of a person or group of people. People with similar attitudes (“likes going fast”, “hates cilantro”) can be grouped together into a segment.

The kind of segmentation you’re probably familiar with is a demographic segment, which is generally something you can tell by looking at someone, like age, gender, or where they live. Demographics are useful in being able to find someone (which is used in targeting, below) but aren’t that great for segmentation for what should be obvious reasons, but those reasons are nonetheless ignored on a regular basis.

Being 20 years old doesn’t mean you will definitely like Coke, or football, any more than being female means you’ll like the colour pink. Those are broad stereotypes, and while there may be some correlation between being female and liking the colour pink, it’s far from universal. Demographics are too broad to be useful on their own for good segmentation and targeting and if that’s what you’re doing, stop.

Why do we segment? Because the ideal segment size is one. You want to buy a thing that is exactly what you need or want, not what your neighbour needs or wants. The reason companies group people together is because it’s easier (and often more efficient) to build a product that suits more than one person. You can get some shoes custom-made for you by Theo the hipster shoe-maker, and you can buy one pair of the several million pairs of shoes built in the same way by a multi-national shoe company. (Yes, mass-customisation is a thing as well, which exists because the ideal segment size is one.)

I may well accept something that isn’t exactly what I want if it’s close enough. My feet are slightly different to my neighbour’s feet, but these shoes fit well enough that we both buy a pair anyway. The difference isn’t large enough that I’m willing to spend the substantial amount of money it would cost to have some running shoes custom-made for me.

While it’d be great (for companies) if all men (or women) universally loved Lynx/Axe deodorant or football, they don’t (thank the gods). Which is why an overly broad segment based on lazy stereotypes is stupid and wrong.

But demographics are a lot easier to find out, so lazy marketers segment based on demographics and then wonder why women get mad at them for making everything pink.


Targeting is about making choices. Once you’ve identified a bunch of customer segments, you need to decide which ones you want to sell products or services to. Saying “all of them” is stupid and wrong because you’ve just missed the whole point of segmentation. Go back and re-read the last section.

You can decide which segment to target before you make a product in the first place, which is a great idea, because then you don’t build something no one wants. Or, you can figure out which segment most closely matches the characteristics of the product/service you already have, and then sell to that one.

Every segment you target gets a different product or service. If your product matches more than one segment, then your product is either too broad, or your segments too narrow. Segments are different in some way, which might be as simple as “segment A is willing to pay more than segment B for the same product” in which case you can sell the same item for two different prices (price is a part of the product!) and make more profit.

Or you can just combine both segments together and ignore the difference in willingness-to-pay in the same way that you are ignoring many of the individual differences in attitudes of specific people who make up your segment. I like cilantro, and my neighbour doesn’t, but that has no bearing on our desire to purchase your shoes unless you make them cilantro flavoured for some reason.

Note how we’re not trying to sell something to people it isn’t a match for? That’s the kind of thing that makes people hate marketing and advertising. If you discover that your product isn’t a close match to any segments, either your research is lacking and you need to try again to find the right segment, or your product is wrong and needs to be redesigned.

Usually the product is wrong, but plenty of marketers invent a segment that matches their product full of people who don’t exist and thus don’t buy the product. This goes about as well as you’d expect.

When you have a product/service that closely matches the desires of a well defined segment, the people in that segment like your thing and want to buy it from you. When you sell your product to these people at a price they are willing to pay, and that makes you enough money to at least cover the cost of capital of making it, that’s called profit.

Tada! You have a successful business.

This post was inspired by watching this video of my old business school Brand professor, Mark Ritson, giving his “social media is over-rated” presentation. This is a pretty good rendition, but seeing him present in person is a treat.

Bookmark the permalink.

Comments are closed.