Unhappy man at desk

When has my campaign failed?

I recently answered a question on Quora about failure, having run a few startups it was something that struck home. As an entrepreneur you never believe you’re going to fail. Even when the writings on the wall, you still have this vague notion that it can work. When it’s finally buried, that’s when you stop. It’s probably a good trait for a very well funded, highly motivated (almost pathological) serial entrepreneur.

Probably not so much for a regular guy, or a marketer.   We don’t have the luxury of being able to ignore all the signs. That whole concept of failing fast & frequently, the one HBR’s been writing about and the one that Silicon Valley seems to fashionably espouse. That means you need to be able to detect failure and respond to it quickly. You should be planning to fail, or if you prefer, failing gracefully by design.

Knowing when a campaign is going wrong
Detecting potential failure in the future


In the absence of a minority-report-future-seeing-psychic, I check my campaigns against metrics and try to ascertain whether we’re on track or not. It’s a questionably better than relying a more arcane-mystic methods, but it’s certainly better than not doing anything at all.

Creating Metrics

When deciding what to track success/failure against, it’s really important to think about your campaign. Personally I believe that (for small businesses) all campaigns should only be focused on client acquisition & lead generation. Which means that all my metrics are funnel metrics (views -> leads -> conversions). If any campaign isn’t performing well, rather than throw more money into it, I’d rather cut it short and redirect the resources to another campaign. There’s an overhead of creating a campaign and running it. Especially for more offline campaigns, but that’s why you should be A/B testing even when you’re doing something as offline as an event. For example:

  1. Handout two variants of flyer with different offers instead of one
  2. Have two different sign-up methods on the flyer and see what works best (QR code & a conventional web link)
  3. Try using electronic sign-up via iPads as well as clipboard sign-ups

If you started any type of marketing campaign, and you don’t have success metrics, you’re doing something wrong. There’s a good chance that you don’t actually have a plan. Check out this guide about setting up a marketing plan. If you don’t have the time to create a marketing strategy, try to think tactically and build metrics on a per campaign basis. It’s not ideal, but it’s better than nothing at all.

Failing to Plan

Action without a plan is not uncommon, it feels better to be doing something rather than nothing. Isn’t some marketing better than no marketing? Yes, you need to do something, even if you don’t have a plan. Just be smart about it, create some metrics, monitor and manage it.  At the very least you’ll have discovered if something works or not. Don’t use planning as an excuse to sit on the fence forever.

Cake or Death

Lego Cake or Death - Eddie Izzard

If things are on-track, and everything’s looking peachy, it’s time for cake! Yay. Don’t rest on your laurels or get too cheeky, start thinking about your next campaign. That’s going to begin from zero, it doesn’t get any head-start from your previous campaigns.

If things aren’t looking good, don’t despair, it’s time for the campaign to die. If you ran it properly you learned something useful that’ll make your next campaign more successful – even if it’s just how to get a consistent 2% improvement in sign-ups through better offers. You should eat the cake anyway, because it’ll make you feel better & it’s silly to waste good cake.

If you don’t have any metrics to measure against, you won’t know whether it’s time for cake or death (or both). It’s probably time to have a cup of tea and watch this YouTube video:

beta testing stamp

Beta testing is for Marketeers

Why Best Test?

I recently met a small company that’s getting ready to release a new app for sale. It’s a great idea and I wish them success, but I was really surprised to learn that they’d had almost no beta-testing. Beta testing isn’t only about bug testing. Asking prospective clients to use your service provides you lots of information that’d be really hard (or painful) to other wise get. Specifically:

  1. Actual use-cases; knowing what our clients actually do with your product is surprisingly helpful. It’s not always when, where or what you might think. How they’re using your product might have an impact on whom you market to, or how you market it to them.
  2. What they like or dislike; if your friends & family are testing your product, you might not get the same level of objectivity, or a large enough sample size. Importantly, the larger you sample size, the more valid your data. There’s no sense in building features that your audience isn’t interested in.
  3. Market testing; The best beta-testers are your prospective clients. Approaching them directly should be straight forward if you know your customer. If they’re not interested in beta testing your product, consider how you’re going to sell the product to them.

Don’t sign them up and ignore them. Remember to speak to your testers, they’re your gateway to real feedback and (actually) important feature requests. Engaging with them in a non-intrusive & un-annoying way can provide you a wealth of information. Your testers are also the most likely prospectives to convert to paying customers or, if you’re really lucky, early evangelists.

Beta Test Sites

If you’re not clear on who your target audience is, or you’re looking for more general feedback there are plenty of beta-testers available through beta test sites. Here’s a list of a few sites where you can register your product and get some impartial feedback:

  • prefinery.com
  • utest.com
  • centercode.com
  • Betali.st
  • Startupli.st
  • killerstartups.com
  • feedmyapp.com
  • listio.com
  • StartupTunes.com
  • openforbeta.com


You can also start working the Internet forums, it’s a great way to get your service in front of very targeted audiences. Some of the forums drive enough traffic to make your product successful just through their traffic alone. The caveat is that you’ll need to be a reasonably regular member at the forum before your post will be taken seriously (or even allowed to include external links). So you’ll need to start work early. There are far too many forums for me to list here, but a quick google search will show you exactly where to go.