Can't Buy me Love (or customers)

18 Dec 2015

This is part 7 in my series of blog posts documenting my attempt to take Saber from Side Project to Profitable Startup. If you've not yet read the previous parts, you should definitely start from part 1.
Heads Up Before August 2018 Saber was known as "BugMuncher", so you'll see the name BugMuncher instead of Saber throughout these older posts. You can read more about the name change here - BugMuncher rebrands to Saber.

Back in October of last year (2014), I was approached by a company named Inkia, asking if I’d be interested in having BugMuncher appear in an advertising campaign named “The Best Tools For Web Developers”. At this point, BugMuncher was still just my little side project, but I’d been considering trying some paid advertising, so I replied saying I was interested and asked for more details.

I was surprised to learn that the advertising campaign Inkia were running was to post a listicle along the lines of “20 Tools To Make A Web Developer’s Life Easier” to 10 high traffic sites, wherein each of the 20 tools recommended in the article had actually paid to be there.

Maybe I was naive, but I’d always assumed the rise of the listicle was simply because they’re easy to write, and highly shareable. It had never occurred to me that people were paying to be in them, after all, BugMuncher has appeared in a load of listicles without me having to pay for the privilege.

The price I would pay depended on which of the 20 positions I wanted to buy into, ranging from a couple of hundred dollars for the bottom slots, right up to a couple of thousand dollars for the top slot (I don’t recall the exact prices). The websites onto which these paid articles were being posted seemed pretty good, including davidwalsh.name, a website I’d read a fair few times when trying to solve front-end issues.

I was very tempted, as back then I thought traffic was my biggest problem. I was only getting about 30 uniques per day, with a 1.5% free trial conversion rate. Maybe this campaign could be the solution to my traffic problem.

I decided to do some due diligence first, and emailed David Walsh to see how the last Inkia sponsored post performed. David Responded saying “Inkia’s last post saw 5477 hits (so far). My blog gets 8m visitors per year.” Apparently that was good enough for me, as I went ahead and ordered the position 5 - 10 package, setting me back a cool $550. I remember thinking how based on the average lifetime value of BugMuncher’s customers, I’d need just three new paying customers to see a return on investment.

Next I had to supply Inkia with a text description of BugMuncher, and an Image. Each blog post would be written by one of Inkia’s staff writers, based on my description, ensuring that each blog post was unique. This is what I sent:

BugMuncher allows users to highlight problems on a website, and then automatically creates a screenshot and sends it to you. It's great for getting feedback and bug reports from users on a live site, as well as improving communication between team members during development.

As well as a screenshot, submissions through BugMuncher automatically include vital user information, such as: browser, operating system, a list of plugins they have installed, and even the path they took through website. Custom data such as user identifiers can be included as well.

BugMuncher is easy to set up, and can integrate with a number of third party tools, including Zendesk, Trello, ActiveCollab and GitHub. It supports all web browsers, including Internet Explorer down to version 7.

The blog posts went live around mid-November 2014:

I waited for the hits, and money, to start rolling in. Except, it didn’t. In fact, the whole thing was a bust. Time to geek out over some stats, these are the total figures from the posts going live to today:

Website Unique users referred Sign up conversions Conversion Rate
David Walsh Blog 945 11 1.16%
Design your way 32 1 3.13%
Design3edge 3 0 0%
Designm.ag 50 0 0%
Designrfix 48 0 0%
Graphic Design Junction 27 0 0%
Inspiration Hut 10 0 0%
InstantShift 49 1 2.04%
Queness 77 1 1.3%
Web Resources Depot 188 0 0%
Total 1429 14 0.98%

Of course that table is missing the most important detail, how many of those 14 new free trials converted to paying customers… Zero, Zilch, Nada, which was actually par for the course, as back then my average free trail to paying conversion rate was about 5%, so I’d have needed 60 new free trials to get those three new paying customers I needed to cover the cost of the ads.

David Walsh’s blog was by far the stand out performer. Based on David’s own data (~5,000 hits to an Inkia sponsored blog post), around 20% of views clicked through to BugMuncher, which is actually pretty damn good.

I could have easily predicted this failure, had I been a little more savvy. I knew that I’d need three new paying customers to see a return on investment. I also knew my sign up conversion rate was 1.5%, and my free trial to paying conversion rate was about 5%. David Walsh had told me the last Inkia sponsored post on his site received around 5,000 hits.

I should have then asked the other sites what kind of traffic their sponsored posts got, or just used Alexa to estimate the traffic, eg:

Website Alexa Rank Estimated Hits
David Walsh Blog 7,208 5,477
Design your way 28,444 1,388
Design3edge 79,277 498
Designm.ag 92,896 425
Designrfix 30,505 1294
Graphic Design Junction 20,061 1968
Inspiration Hut 59,960 658
InstantShift 22,777 1733
Queness 48,623 812
Web Resources Depot 45,487 868
Total - 15,121

So I could have estimated that in total the articles would receive around 15,000 hits, then working backwards from my conversion rates:

Goal of 3 new paying subscriptions

60 new free trials needed (5% conversion rate)

4,000 unique users needed (1.5% conversion rate)

26.45% click though needed (4,000 out of 15,000 hits)


Based on my estimate of 15k hits, I’d need at least 26.45% of all visitors to the blog posts to click through to BugMuncher, which is quite a big ask. But I didn’t do any of these calculations, and so I wrote it off as an expensive lesson, and gave it no more thought.

Round 2, fight!

In September of this year, Inkia emailed me again, asking if I’d like to participate in another similar campaign, on 9 of the same websites as last time. This time there would only be 10 tools featured in each post, and the pricing structure was:

  • Position 1. $2,200
  • Position 2. $1,600
  • Position 3. $1,100
  • Position 4. $800
  • Positions 5 - 10. $550

I responded saying based on the cost and performance of the last campaign I would not be interested. But then I gave it some more thought, the timing was interesting, as I’d just started out on my mission to make BugMuncher my full time job. I had budgeted £500 (~$750) / month for paid marketing, and had already made a lot of improvements to both the landing page and BugMuncher itself. Maybe this one would be more successful.

Having not learned form last time, I still didn’t try to estimate my potential return on investment. The next day I sent Inkia another email saying I’d reconsidered, and asked if they could do one of the position 5 - 10 package for $500. They agreed, and once again I sent them an image and description:

BugMuncher is an amazing feedback tool that allows your users, team and clients to easily send detailed feedback and bug reports. BugMuncher automatically includes a screenshot of what the user actually sees, without requiring any browser plugins. This means you can easily understand the feedback, recreate the conditions of bug reports, and even catch cross-browser rendering issues.

BugMuncher also automatically capture the following information, and includes it with every feedback report:

  • Browser and Operating System Details
  • Details of any Javascript errors that occurred, including a stack trace.
  • Screen size and resolution
  • Which Browser Plugins they have installed
  • The user’s language and geo-location
  • The path they took through your site

Looking at them side-by-side, I think I actually prefer the image I used for the first campaign, but the new image matched the updated branding and tag-line I was using at the time. The blog posts went live on the 22nd of October:

Time for some more stat-geeking, I realise this campaign hasn’t had as long to run as the first, but the vast majority of the traffic had tailed off after one month on the first campaign, and the second campaign showed a similar spike and drop off in traffic, so I think after nearly 2 months it’ll be a fair analysis:

Website Unique users referred Sign up conversions Conversion Rate
David Walsh Blog 358 2 0.56%
Design your way 15 0 0%
Design3edge 2 0 0%
Designm.ag 29 0 0%
Designrfix 6 0 0%
Inspiration Hut 9 0 0%
InstantShift 23 0 0%
Queness 12 0 0%
Web Design Ledger 111 0 0%
Web Resources Depot 16 1 6.25%
Total 581 3 0.52%

Oh dear, my second foray into paid advertising had faired even worse, bringing me around 1/3 as many uniques, and an even worse conversion rate. I’m sure it will come as no surprise that not one of those free trials converted to paying customers. Once again David Walsh’s blog was by a wide margin the best performer, with newcomer Web Design Ledger doing pretty well too.

I’m sure my parents have warned me about throwing good money after bad. I should really listen to them more.

I was understandably disappointed by the performance of this second campaign, so I decided to try and figure out why it had been such a flop. Was buying coverage always destined to fail, or was my image and copy the problem?

I went through the other nine tools listed in the articles, emailing each one, asking how the Inkia campaign had performed for them. I received two replies:

“The results were pretty good for us and we are happy with this campaign. We tested many similar campaigns in the past and we can tell you that top 3 usually converts. You may have had such results because of the position in that campaign.”

and

“We have used that promotion as an SEO promotion tool so we were quite satisfied with it.”

Lessons learned

So having wasted over $1,000 on these campaigns, this is my advice to anyone considering buying into these kinds of posts, based on the mistakes I made:

  1. Only bother with the top three positions.
  2. Ask about the other advertisers who’ve already signed up, to make sure you’re a good fit.
  3. Don’t go DIY on the image and copy - if you don’t have professionals at your company, pay freelancers.
  4. Really research the quality of the websites, there were a few duds in these campaigns (I’m looking at you design3edge.com)
  5. Contact each of the websites, find out how these posts usually perform, and use your own conversion figures to predict return on investment.
  6. Negotiate the asking price.

Personally I won’t be trying this again, as buying a top 3 position and then hiring a designer and copywriter would end up costing me around $2,000. The cost isn’t the only problem - some of the sites onto which these listicles were posted have mediocre traffic, and a few of them seem to be using the same WordPress theme, posting very similar content. Not exactly quality links.

If I were to try paid placement again, I’d approach the sites directly. Clearly all these sites are open to sponsored posts, and that way I’d have more control over the content, and where it was posted, all without the added cost of a middle man.

So that’s how I wasted $1,050 trying (and failing) to buy customers. I realise there’s a chance that by posting all this, I could potentially piss someone off at Inkia, or any of the host sites. After all, most of the sites don’t give any indication that those are sponsored posts, and even when they do it’s not particularly obvious. However, there’s nothing in the agreement I have with Inkia about confidentiality or disclosure, so as far as I’m concerned it’s fair game.

This may well be my last post before Christmas, so Merry Christmas to all, and to all a good night!

- Matt

Continue to Part 8 - December's Details

With the end of the year looming, I had predicted December to be a quiet month. I was pretty sure I wouldn't maintain my 10% month on month growth, as businesses would be winding down for the year. Sometimes it's good to be wrong.

Keep Reading


comments powered by Disqus

Saber - Quick and Easy Website Feedback