Ask HN: What kind of advertising have you tried that worked?
Our startup is looking to use banner advertising to help increase sales. So I have a few questions for the HN community:
1) We have ideas of what the theme of some of these banner ads will be, but we don't know the best wording that should be used. Should we get a copywriter? If so, where? How much should we expect to pay?
2) We can do the ads ourselves from a creative (art) perspective, but it may make more sense to have someone else do them for us if they are better. Where can we find a good designer for banner ads? Again, how much should we expect to pay?
3) Is there a resource online where we can gauge the potential success of these ads? We're interested in knowing what we can expect for CTR rates and/or CPC.
4) Any additional information that we should consider?
Thank you for responses ahead of time!
(
41 comments
| 87 points
| on HN
)
The Deck - http://decknetwork.net/ We received roughly 4300 uniques across 2 weeks. We got about 9x the clicks from daringfireball than any of the other sites in the network (37signals, A List Apart, Kottke, FFFFound, etc.) If you email them they are open to negotiating.
Fusion Ads - http://fusionads.net/ Very similar model as The Deck. We ran one of the $400 one-week "bursts" that they offer. Received about 1400 uniques.
Before we ran either of these campaigns we experimented on Facebook to see what sort of graphic/copy produced the greated CTR. We found that a very simple graphic, and simple copy always did the best.
After we ran the campaigns we also realized that Facebook advertising is a great deal. You can run really targeted ads on FB for cheap and get great results. I would definitely experiment here before spending big money on one of the larger networks.
Regarding art/copy, I would see what you can come up with on your own. Unless you're running a massive campaign, not sure a designer/copywriter is worth the expense.
Finally, it should be mentioned that a clever blog entry or two can generate far more traffic than a banner ad campaign.
1. Test and track EVERYTHING. Every variation, headline,layout, domain ad is on, etc should have a unique ID, so you can see not just what gets the highest CTR but also which ad converts/backs out the best.
2. Good design does not necessarily mean good ads. I've had the highest CTR and conversions come from the simplest, ugliest ads. See Plenty of Fish banner ads for an example of ugly,yet effective advertising.
3. Regardless of the direction you take your design, have a clear, easy to see call to action ("click here!" button for example). Design the ad to draw the eye towards that button.
4. There are two main strategies in banner ad design, blend or clash. Either blend in with the design of the site your banner appears on with similar colors and visual elements, or go the other direction and make an ad that clashes with it and stands out.
More specific banner media buying advice:
Taget smaller sites your customers are likely to visit. Ask to buy remnant inventory, and pay CPM. Don't pay more than $2-$3 CPM except in specialized cases. Don't go through ad ad network unless you have a big testing budget, instead go direct to webmasters. Use your own adserver and pay based on your adserver's numbers, not theirs. 1% CTR is fantastic for banner ads, but your EPC/conversion numbers will be the way to determine if an ad is succesful.
My email is in my profile, you(or anyone else here) can email me if you have other questions.
PS I don't think you need to hire a copywriter, you're obviously good enough at writing headlines based on how many upvotes this submission has!
(Side note for everyone: Please don't withhold useful information from your readers for fear of self-promotion. But if you must, at least fill out your profile.)
1. Tradeshows. We went to industry tradeshows, and our sales guys really worked the shows. We didn't know anyone in the industry, either. They worked the shows, really got out there to know the people, and worked the people covering the shows (magazines, photographers, etc). Basically, it was all about networking.
2. Advertising in print got us coverage. We bought ad space which came with a write up in the mag and online. Is good because it gave us presence.
3. Presence is important. We'd go to trade shows all over the place, and really mingle. Tradeshows seemed to do well, and was the place for face to face meetings.
4. Incentives to get word of mouth advertising. We would pay commissions to clients who would bring in other people. It wasn't merely a link to click on. Rather, it was direct referrals. "Hey, here's a guy who could use your services..." and we'd give him a fair cut.
But really, presence was the big thing. Being out there, being seen, being apart of the environment was big, and really allowed us to get out there. If you present yourself as a successful company, people assume you are. Obviously you need to back it up when they sign on, but it's getting them to call you back that is important.
Should also stress that most of this was between 2003 and 2006 when we started this and really pushed.
2. Same as #1.
3. There are SO many variables in what you're looking for. I recommend using some sort of tracking system (I use Omniture) to track as much information as you possibly can, so you can better target future advertisements.
4. Depending upon what you're doing, you may not do very well with banner ads. I found that in my last business, being active in the "community" behind what it was based (posting on forums, showing up at events, etc.) was INCREDIBLY effective, and the banner ads didn't get me more than around 2-3K in sales (I was only making an average of 20% margin or so), which barely paid for having them in the first place.
3) Wouldn't tracking it yourself be the best method? 4) What was your last business? Ours is a SaaS product targeted to small businesses. Would be very costly to attend events.
http://petewarden.typepad.com/searchbrowser/2009/11/get-new-...
In my limited experience it's all about testing and iteration, lots of it. Or just find a feeble excuse to show a picture of a busty girl, judging from most of the ads I see on Facebook.
Also, be prepared for shockingly low CTRs on banner ads like Facebook's. Even highly targeted I was seeing 0.07% on my best ads, and from talking to others that's not uncommon. Even at that low of a rate the ads were cheap enough to be worth running.
A couple of key points from the post:
"My ad ran for 13 days, from 3/7 to 3/20. It cost $650, and I spent $50 per day. In total it had 1,288,378 impressions (282,732 uniques) and yielded 20,700 clicks (18,420 uniques)."
"Redditors actually try out your site. 3c per unique visitor is pretty good in and of itself, but it's all worthless unless they actually try out your site. For example, you can get 5c unique visitors from StumbleUpon (presumably in a similar demographic), but StumbleUpon visitors never would try out my sites. Reddit visitors did try out Duck Duck Go."
One of the best is SomethingAwful. They have a sub-forum dedicated to firearms, and a very close knit community. An ad on SA costs $30 for a month (which is insanely cheap) and I started a thread advertising my company. We've done incredibly well that way, have already sponsored a competitive shooting team and met a bunch of nice people.
What you don't get, of course, is in depth tracking. You can only see how many clicks you get (just wrap the link in your ad around bit.ly or another URL shortener). But it's just as well with me, I know my advertising is hitting it's target audience and it's worked out very well.
What I've seen from my own personal experience, is sites that sell advertising space have been tracking their advertising since they started because they need to know the value of their real estate.
Over time, this transfers to a fact that where ever you are doing your ad banners (a web site, a blog, a search engine), the people you are buying from know how to reach their audience best. Serious advertising firms have their own creative-content guys on the inside, and using them in those cases is way better than picking an external designer, freelancer, or using your own guys. This kind of comes down to this:
"If someone has a big yard, the kids that live there know how to play in it best."
So: -Split text your ads. -Use tracking software to do it. -Don't assume you know what will work. Let the data decide (not a copywriter).
I'm pseudo-famous on the back of my Rails tutorials and cheat sheets.
This is way better than any kind of advertising you can simply plunk down money and buy, for obvious reasons -- it gets people to really like you, to spend the most time looking at what you make, etc.
1. personal appearances at game conventions dressed as the mad scientist Dr. Heinrich von Hexenhammer --- NOTE: may not work for your project
2. listings on niche sites
3. astroturf forum comments on related sites
paid advertising? oh that actually hasn't worked too well in comparison. funny.
I'm sure it's pretty effective, especially if you only need to get a handful of people to notice your product. But it just feels so dirty...
This is actually illegal in the EU now under EU's Unfair Commercial Practices Directive; eg http://www.theregister.co.uk/2009/12/08/eu_consumer_protecti...
It's soooooooo easy to create fake/throwaway identities on the Internet I think you often have to be very very skeptical about the value of any gushingly positive or gushingly negative "reviews" of a commercial product or service written by a stranger, especially if it's a sort of mainstream category like books or music albums where there might me a traditional PR/advertising company involved that's now doing "social media" promotion. Giveaways: same writing tone/voice/pov used in each review; use of industry-specific insider/professional lingo that the average Joe wouldn't know; reviews that are just too complete/thorough, etc.
Which then leads back to what I think is going to be a service category in increasing demand in the future (and thus an opportunity for more startups): trusted curators and aggregators. Entities that don't just include everything. Entities that discriminate with an eye for quality, and try hard to be independent from vendor influence. For example, for me it once got to the point where the only movie reviewer I would trust was Roger Ebert. With other reviewers I couldn't tell if they were paid shills, or they were naive/ignorant, or too much a slave to their own personal genre tastes, or they just weren't going about doing the review in an intelligent and balanced manner. Ebert I could generally trust. I want a web-based Roger Ebert for doctors, for auto repair shops, for schools. Seems like there are some guys doing quite well doing it for wines, for example.
This is actually one area where I think crowdsourcing is horrible, by the way. Crowdsourcing when applied to reviews is almost the total opposite of trusted curation. What if the crowd in question is full of fools or paid shills? Garbage in, garbage out. The Idiocracy effect where quantity drowns out quality.