Father of advertising David Olgivy said every advertiser should have to read Claude C. Hopkin’s book, Scientific Advertising, seven times before being allowed to work in the field. I would say the same holds true for developers. It takes a lot more than beautiful code to be successful. Scientific advertising is also key.

Hopkin’s classic was published in 1923. In his chapter on mail-order advertising, Hopkins states:

“The severest test of an advertising man is in selling goods by mail.Their cost and result are immediately apparent. False theories melt away like snowflakes in the sun. The advertising is profitable or it is not, clearly on the face of returns. Figures, which do not lie, tell one at once the merits of an ad.”

Enter 2015. The App Store is now larger than Hollywood, and every few hours Apple ships more transistors than were in all the PCs on earth in 1995.

Now imagine Hopkin’s quotation updated for today:

“The severest test of advertising, is in selling goods by in-app purchase….False theories melt away like snowflakes in the sun.”

I’m guilty of only reading the book three times so far, but I’ve applied Hopkins gems to mobile marketing at Robots and Pencils to help us become the 34th fastest growing technology company in North America (and yes, I do think if I read it a few more times we could have been like 20th).

Here’s a glimpse at what has helped us over the years.

Gem 1: Easier than horse and buggy

Hopkins said: “We see countless ads running year after year which we know to be unprofitable.The money is spent blindly, merely to satisfy some advertising whim.”

While Hopkins was limited to using key coded coupons and travelling by horse and buggy to collect the results of his headlines, offers, and proposition tests, we have the ability to track just about anything on a variety of platforms. Yet, time and time again, I meet developers who are either not willing to spend some money up front to implement testing or think it’s too much effort (no need to clean a stable in 2015). Instead, money is spent on user acquisition that developers think will work.  But, if you aren’t continuously testing hypotheses, you are likely just wasting your money and morale.

Gem 2: The tyranny of segmentation

Hopkins said “A double-page ad on women’s dresses will not gain a glance from a man. Nor will a shaving cream ad from a woman.”

While Hopkins is speaking to two different products, the perceived benefit for your product can vary. Find out what types of people are using your app and how they’re using it. Then create marketing messages and in-app offers that speak directly to each group. Products like Flurry, MixPanel and Tune are your friends.

Gem 3: Be specific not elastic

Hopkins said: “If a claim is worth making, make it in the most impressive way.”

People respond much better to specific examples. For Hopkins, it was changing an ad from an elastic claim of “Used the world over” to a specific claim “Used by peoples of 52 nations.”

As developers, we need to find our specific claims. It’s easy to be general, but it’s the social proof that stands out to potential customers — who uses the app, how many people use the app, what are the results of using the app? Scrap the “best in the world” claims. Be specific.

Gem 4: The $10,000 rule

Not Hopkins this time, but rather Mr. Mead from Mead Cycle Company. The company obsessively tested its ads to arrive at a point where Mr. Mead said he would not change a single thing on his ad, not even for $10,000.

Today, $10,000 would be equivalent to approximately $138,000. A lot of capital, or 16M Elixir in Clash of Clans, and a seemingly tempting offer. But if you’ve continuously tested and optimized your ads to the point where your ROI is something to depend on, $138,000 just won’t be worth it.

Michael Sikorsky (@mjsikorsky) is the CEO and co-founder of Robots and Pencils Inc, a mobile strategy and app development company. Launched in 2009, Robots and Pencils has since created more than 250 apps used by 77 million people world wide. Most recently, the company reimagined mobile development with the launch of PencilCase.