What I Should Be Doing as a Marketer. But Won’t Be


“Should” is a complicated – and dangerous – word in marketing. How many times have you read blogs and articles proclaiming that you “Should be doing more mobile marketing”, that you “Should have full content strategy”, that you “Should be creating personae for all of you target segments”, “Should be doing more on Twitter”? And it’s not just from marketing “thought-leaders” – as a marketer, I probably hear suggests on a daily basis of new things that we “should” be doing.

But I want to use this post to dissect the word “should” a little. When someone says “You should be doing more social media advertising”, what do they mean? I think “should” is a word with too broad a definition in this context.

As a first pass, I’d like to narrow “should” to mean “Activities which will have a positive impact on the business and affect the KPIs you worry about”. I know this is obvious but we’re not doing God’s work here – marketing activities and spend exist to generate and nurture leads to make money for your business. Unless you’re doing marketing for the church, in which case, yes – you are doing God’s work.

A second criteria that has to be applied is a constraint – you only have limited resource (either people or money). I suspect there are very few activities that would be particularly negative for your business. The problem is that many have very little impact at all. This would be fine, if it weren’t for the fact that marketing costs money – so it needs to do something positive for your business, or you’re just wasting your limited resource.

So the criteria so far is that you “should” carry out an activity if it positively affects your business and if it’s do-able with your limited time and money. So far, so obvious. But, how do you know whether these criteria are met (the first one is the hard one – it’s pretty easy to know how much something will cost to do, much harder to know if you get a return on that investment)?

I try to use four levels of qualification to assess whether we do something. I’ll describe these below, then try to apply them to the plethora of activities that our business does. Really these are measures of “How strict am I going to be on the evidence I need and level of analysis required to say that a given activity is really worthwhile?”.

The levels are:

  1. “Should” do, because thought-leaders on the Internet think it’s a good idea,
  2. “Should” do, because although there’s no data to support it at all, I believe these are having a beneficial impact,
  3. “Should” do, because there’s pretty good data showing it actually works,
  4. “Should” do, because there’s good data that would stand up to scrutiny by any scientist worth his/her salt.

And here’s my table of activities showing whether I think we “should” do these things using these criteria. NB: By “Should” I really mean “Spend a lot of time and/or money getting this right”:


As you can see – a lot.

Couple of points to make:

  1. Nothing fits the fourth criteria – having evidence good enough that wouldn’t be ripped apart by a half-decent scientist. I studied science at college, and remember ripping apart published psychology papers because of poor experimental setup (there is always an alternative explanation in psychology!). I’ve never seen any study or data in the area of marketing which reaches even the lowest standards applied by a psychology researcher – for example, running longitudinal studies or setting up proper controls. Still, if we applied this criteria, we’d never do a single thing in marketing!
  2. The list of things where there is even half-decent data supporting the work (criteria three) is also pretty short. As per my previous post, this is why we’re working with HubSpot – that list has to grow bigger, so that I can more confidently say “Yes, that works, and here’s the evidence (albeit, scientifically embarrassing) that it does”, converting criteria three items in to criteria two items.

Given all of the above, here’s what we’re planning to spend more time on going forward shown in red. As I say, the fact that there’s nothing against “SEO Optimisation”, say, doesn’t mean we don’t spend some effort on it as a company. What it means is that I don’t believe spending a lot more time and effort on it, will move any needle that I’m interested in:


Again, a few points:

  1. Everything we’re doing is something I believe works! Even if I have no evidence for it…
  2. But we’re not doing everything I believe works – this is the time/resource constraint. We have to prioritise and some activities that could have an impact just won’t happen. Even though we “should” do them.
  3. There are things like “Google Adwords” which we won’t be spending more time on, not because it’s not very important but actually because I think we already do a great job and we’re in the arena of diminishing returns.

But, we do plan to kill a lot of activity. Activity that we really “should” be doing. It’s just that I need to use a different definition of “should”, one that allows us to focus and focus on things that I think will make a difference.

0 thoughts on “What I Should Be Doing as a Marketer. But Won’t Be

  1. Really like the rigour of this approach. I’d love to hear more about the market sizing activity. My instinct is that it’s a great input to balancing investment, but do you see it adding any other value to more explicitly ‘marketing’ objectives? Is there a case for it to increase income, or do you see it mostly as reducing costs by better targeting of other activities?

    1. Sometimes it can, but not always. So, if your product supports multiple different platforms (say you have a product that works against 5 different NoSQL databases 😉 ), and your market sizing says that database platform A is 5 times more prevalent in the people you reach than database B, then that’s what goes at the top of your “supported platform” pages, and that’s what platform you use for your first case studies and reference customers.

      Also of course, the fact that something is useful doesn’t necessarily make it easy! Most of the items listed aren’t enormously difficult, they’re just time intensive, or expensive. But it’s very easy to pick the activity that you already know how to do, rather than the harder thing (like rigorous market sizing research), which might be more fruitful.

  2. Good post. Like the use of framework to put rigour into the analysis.

    I was left wondering how you feel about Content Marketing. I couldn’t decide if you’d included it with Content Strategy (doubtful); if it was spread across a number of other items, or if you left it off the list because it already had more items than can be resourced?

    1. Thanks for the comment Paul. The omission of “Content Marketing” is firstly a bit of an oversight from rushing the post (I lost the first draft to a bad connection, so re-wrote it hastily!), so apologies for that. I should certainly have been clearer in what our approach to “Content Marketing” is.

      But secondly it’s because we’re very lucky to have a set up where I work, where “Content Marketing” is almost taken for granted. We have a couple of sites we own (http://www.simple-talk.com and http://www.sqlservercentral.com if interested) run by our publishing team which produce a phenomenal amount of amazing content and have been doing so for years. I’d say it’s been one of the key things that has driven the growth of our business (the traffic to these sites outstrips to the traffic to our company site many-fold) and we have enormous email lists from these.

      So one of the reasons I “forgot” it, is that I do, to some extent take it for granted – not least because there’s a whole separate team who run it, so it’s not generally in my list of things to do. Massive apologies to our publishing team if they read this…

      The one other subtlety is that we have produced a great deal of content and activity without any real “Content Strategy”! Hence why I put this as something we don’t do. We’ve just produced tons of quality stuff, and of course we think about the subjects and the audience a lot, but the approach has been quite serendipitous, rather than planned. And it’s worked out very well!

Leave a Reply to Paul Walsh Cancel reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>