Marc Hughes

I am a developer from a bit west of Boston.

Geek to Marketer in 30 Days

14 Sep 2015

alt text

I'm a geek with a semi-successful online project management tool ( For a long time I've felt that the biggest thing holding us back was our marketing & sales side of things.

For obvious reasons, I really want to change that. Last week I attended Inbound15 in Boston and am psyched to step up our marketing game.

My 30 day challenge

To help focus my efforts, I'm challenging myself to do one new thing in this area every day for a month.

I'm planning on writing about what I'm doing, and when there is data, to share some results of those efforts on this blog. If you're interested in hearing about that journey follow me on twitter @mhughes

Day 1

Today is the first day, and I've taken a few baby steps.

First, I figured out how to post on this blog again and have it running. I'm quickly realizing how woefully out of date the software for it is and will be looking at some alternatives. But this isn't the main product so I don't want to spend much time on it.

I AM spending some time looking at our current publishing platform (jekyll) for ScrumDo. We're considering HubSpot or perhaps a less expensive service.

One small lesson I learned at inbound was the value of having an image on every piece of content, even if it's just a re-hash of the title. The main goal being to have something visual to post if someone shares on facebook or twitter. So I've signed up for Canva and made that nifty graphic at the top of this post. I'm skeptical if something like that which is mostly devoid of useful content will actually help, so maybe an A/B test at some point.

Marketing Automation

I'm investigating some marketing automation solutions, but am really disappointed by them so far. Maybe it's just my engineering background being disappointed by something sounding so cool "Marketing Automation" but turning out to be so dumb. They all seem to fall into one of two overly simplistic modes:

  1. A pre-defined drip campaign, perhaps with some simple if/then branching to choose the message.

  2. A completely event based targeted message based on criteria while more or less ignoring the overall message flow being sent to the person. (ie. when they do this, send them this message a day later)

I would really like a more intelligent mechanism that is a mix between those two that works something like this...

  1. Every N days, send the prospect an email.
  2. Choose the email to send from this large pool of potential messages.

Each message in the pool would have some sort of weighting algorithm. For instance there might be a default priority on each plus several behavior based rules to modify the priority.

Then, when it's time to send a message to a person, the system calculates all the priorities for that user, chooses the one with the lowest priority score that the user hasn't seen yet, and sends it.

Message one - about pricing
Priority 100
If the user has looked at the pricing page already, add 300 to priority.

Message two - Tell about a premium feature A
Priority 200
If the user has used feature B, subtract 100 priority (for instance, if feature B and A go well together)

Message three - Tell user how to invite other people
Priority 50
If the user has already invited someone, add 200 to priority.

And maybe we have 50 different messages to choose from (each user would only get X messages total)

From there, we could track what messages lead to sales most often and go back and adjust priorities from time to time based on that.

There is a greater-than-zero chance that I'll wake up from a caffeine induced coma at some point having made a prototype of such a system.


I'm also looking at our new user onboarding flow. Already I've identified a couple changes I know I want to make.

  1. The first email we send the user is a "verify email address" message. I'm going to change that to a Welcome style message with links to resources to help them be successful and include the verification there. If we don't get them to verify their email, we can always try again later.
  2. The actual signup process had one step that was just taking too long to process. That's already fixed.
  3. We have a getting started guide, but a. It's hard to find and not on the first page people see b. Is not formatted nicely c. Could have more useful information on it
  4. The initial toolbar on our main project page is too complex.
  5. Creating a new card (a very basic operation) needs to be a lot more obvious.

Talking to people

I am going to talk to more people. It's been way too easy to hide in my cave and ignore the outside world. Today, I've already had a call with an agile coach, and later today I have a call with an ex-customer who recently left us.


So, there's day #1... I feel like it's a good start.

Follow me on twitter or subscribe to the RSS feed of this blog if you want to see how it goes.

You can see all the posts in this marketing series here.