This article was originally published on Patrick McKenzie's blog and is reprinted here with permission. The original article is here. It covers generic post-launch steps for a uISV to increase software sales.
Here is a question which comes up all the time on the Business of Software forums.
Hello, my name is XXX and I created an application N months ago which sells for $Y. I have gotten Z sales so far. What can I do to increase sales?
First off, congratulations on finishing an application that real people think was worth spending money on. That’s an accomplishment. It is also the easiest part about running a software business, and you’ve got a long road ahead of you. Lets get cracking.
Actually, first, the obligatory disclaimer: everyone’s markets are different. Everyone’s goals are different. Everyone’s strengths are different. I sell to technically disinclined B2C customers and make enough money to nicely supplement my dayjob every month but could not go full time on Bingo Card Creator (a good thing, too, as 1-2 hours of work a week would leave me with far too much free time). I’ve also only been doing this for, hmm, coming on 10 months now. This isn’t the Bible of Internet Software Marketing, its just things that I have found effective and advice passed along from people I trust.
1. Install Analytics. It is critical that you have enough data on how your business is working to make informed decisions. Analytics software (I like Google Analytics and CrazyEgg) lets you know how many people are visiting your website, how they got there (in particular, what search engine terms got them there), how many of them go on to download your free trial or purchase, what pages are most of interest to them, etc. If you can afford to invest more in analytics, tools like Omniture may be worthwhile.
2. Start Search Engine Optimization. (Wiki link: SEO) Ideally, you’ll have had your website domain up for a significant portion of the time you were developing, packed full of keyword-rich content gradually aging like a fine cheese and convincing Google that you’re not some fly-by-night spam site operator. Didn’t do that? Don’t worry, I didn’t either — its a great thing to have done but for first projects it is a great idea in hindsight for almost anyone. Anyhow, age is one factor which gets you out of the Google sandbox, which is where sites languish without getting headway on competitive search terms.
However, even if your website went up 2 months ago, you can still start SEOing actively. Concentrate first on making your website very useful for people who land on it. This involves sharpening your pencils (or WYSIWYG editors) and writing some compelling content. Its shocking the number of folks who come in for advice on the forums who have less than 100 words of content on their website in total. Google can’t read minds, folks — if it isn’t on the page or in referring links they have no clue your web page is about that. So start writing.
Write about what your customers care about in language similar to what they use. You know that plastic marketing speak that large companies seem to produce far too much of? Introducing a new paradigm in best-of-breed B2C customer empowerment synergies? Nobody writes “new paradigm in best-of-breed B2C customer empowerment synergies” in Google! They write things like “How do I deal with abusive customers?” (natural language search is very scarily common among non-technical users, incidentally). If you write your page like they write their search queries, you win. (Simple example: My original title for this post was “So You’ve Got Sales. What Now?”. That is how I talk, but its certainly not a natural search string. “increase software sales” is, however, and this post will probably be on the front page of Google for that query within a week.)
Don’t neglect the technical end of onsite search optimization. There are stupidly simple five minute fixes which will improve your rankings dramatically. Use your title tags. Use h1 and bold to call out the important bits of your page (that helps Web readability, too). Add descriptive alt text to your images. Use this really easy trick I shamelessly stole from Nick Hebb (who makes flowchart software, which he handily describes in a terminology box) and include a sidebar box listing synonyms for your key search terms. Its not obvious to Google that I sell a bingo card maker without the bold callout on my frontpage saying so.
3. Start getting links.
This is the other side of the SEO puzzle. I have tried buying links and, well, that was a crushing failure. (I ended up paying $40 to get mentioned on a Chinese forum and a spam site… thank you sir, may I have another.) Its the links I didn’t lift a finger to get that are actually worthwhile to me in terms of traffic and SEO juice.
Well, “didn’t lift a finger” understates the efforts a bit. Sites don’t attract links. Content attracts links. A person who tosses you a link from their site, blog, livejournal, email to Mom, newsletter, whatever, has taken a bit of time out of their day to promote something you have done to people whose trust they have built up. They really value that trust, and they don’t waste it by wasting people’s times with links to useless pages (and God knows there are enough of those on the Internet). Rather, they send links to pages which are interesting, topical, useful, etc.
So how do you get links? Write content which is designed to be linked to, sometimes called linkbait. Sure, you’ve got your software to sell, but unless you’re exceptionally lucky people won’t wake up one morning and decide their blog readers need to hear about your product. However, folks in your niche have a variety of common interests, and they’re always eager to hear about that. For example, elementary school English teachers are some of my best customers. One thing they really like is having lists of Dolch sight words, which form the basis of early English instruction. They pass them around to colleagues, print them out and hand them to parents, include them in the classroom’s weekly newsletter, link them from the Early Readers Homepage, etc etc. Writing that one page, which does genuinely provide value to people in my niche, took a few hours but pays off every single day of the year.
By the way, notice the instructions at the top on how to link to the page? This is a fairly important thing for non-technical customers, who might not know what a URL is. Blogging software, etc, makes it easier than every for folks to provide links to things. For folks not using it yet, I try to make it as easy as possible for them to help their friends out while in the process helping me. Putting up a simple HTML sample helps quite a bit for that.
In the category of providing useful, easily linkable content, blogging has few equals. If you talk about what your customers care about, people toss around links like candy. The culture and technical nature of blogging strongly encourages links. You can capitalize this by having a customer-focused blog on your site. (This blog is neither customer focused nor on my site, yet it sends me 10% of my traffic and a few hundred dollars worth of sales. Please, do better at following my advice than I do. ) I strongly prefer http://www.mysite.com/blog over http://blog.mysite.com and http://myblog.myblogprovider.com in terms of SEO benefits. Oh, a word on software — Wordpress just works. I have heard good things about MovableType, too, but Wordpress is good enough for me.
Andy Brice (he makes software that does table plans for weddings and also has a very interesting series on marketing methods on his blog) is of the opinion that blogging rapidly diminishes in relevance, so it is a constant time commitment. I agree for blogging as practiced by many technically inclined folks, where you are perpetually identifying the New and Shiny or the controversies of the day and commenting on them. TechCrunch, for example, has archives which are stale mere days or weeks after the posts are written.
So don’t write like TechCrunch.
I like to think of blogging in terms of producing resources for readers. The best resources are evergreens — they’re good today, they’ll be good tomorrow, they’ll probably be good in years. Some of my more popular posts here, for example on software registration systems, would have been topical ten years ago and will probably be topical ten years from now. That post picks up links, visits, and comments six months after being written. Writing evergreens is like investing in yourself — it is a way for today’s labor to pay dividends tomorrow and every day thereafter.
Blogs also foster a sense of community. Having communities of your customers online is nice. It allows you to hear useful feedback on how to improve your product, gives you a built-in base of passionate folks who spread the word for you, and folks eventually get to know you personally and are nice to you because of that. For example, there is a vibrant little uISV community on the BoS boards and in a wee little circle of blogs, and within that community there are both passionate users (I have been described as the local sales rep for e-junkie before, and this blog has probably sold more copies of Direct Access than it has Bingo Card Creator) and lots of folks who help each other out. One example: I’m not sure exactly who started it but Ian Landsman and a couple folks noticed when I was writing about Free Bingo Cards and decided to spread the word.
5. Eliminate barriers to checkout.
Presumably if you’ve got sales you’re already capable of processing credit card payments through at least one processor. Good. Can you offer another one, for example if folks don’t trust Paypal? One of my favorite features of e-junkie (watch me sell them again — I swear, I really don’t make any money doing this ) is that you can get Paypal and Google Checkout working for the same amount of work (i.e. not much). Some folks already have Paypal accounts, some have heard horror stories and will never trust their credit cards with them, and never the twain shall meet. Checkout is a useful (and cheap) safety valve for those prospects.
Also, make sure your prospective customers know you can process credit cards and checks. “Pay here through Paypal” doesn’t provide useful information to customers who don’t know what Paypal is (they exist, trust me). Mention that “Paypal is a trustworthy company used by millions (including eBayers) which processes your Visa, Mastercard, or checking account so that you can buy things online” and watch your conversion rate go up. There are a variety of possible checkout logos available or you can roll your own, but for goodness sake put the credit card logos on or near the button. Its one of those no-brainer “having logos beats not having them by 3-1″ type decisions. Make sure, you're accepting the right types of payments. In europe, credit cards aren't as widespread as in the US. Money transfers are more common here, as you can see for example on the checkout-page of the European company ITSTH.
Is your checkout process instant? No? Fix that. None of this “You’ll get your registration key in 24 hours” nonsense, particularly not for B2C apps which may be impulse or time-sensitive purchases (I get LOTS of customers who need to make cards for a bridal shower tonight). Also remember, you don’t get the benefits of having your process be instant if you don’t mention them to your customers before checkout! Make sure they know they’re one simple 30 second form away from having shiny new software!
6. Offer a money-back guarantee prominently.
7. Work on your AdWords campaign.
I have poured dozens of hours and hundreds of dollars into my AdWords campaign (my #1 business expense by a factor of lots), and had my fair share of months where it cost me more money than it made and where Google had me pulling my hair out. The bottom line though? You can make it work, and when you figure out how to its free money. (My cost per trial is trending down from the profitable $.30 to the noticeably wealth-producing $.20-$.25, and its a nice hedge against fluctuations in organic search rankings.)
SharewarePromotions has some nice articles about AdWords optimization, and my archives have quite a few as well. Many of the improvements you make to your site at large, like improving landing pages and putting important keywords in your content, will help AdWords out. Other than that, use AdGroups well (focus them by theme), keep search and content networks in separate campaigns for ease of use, prune nonperforming or overly expensive keywords religiously, keep your eye on the cost per conversion number while keeping CTR at the back of your mind (keep it above about 1% on search ads or you’ll get penalized harshly by minimum bids which take MONTHS to work back down, and keep trying different ad texts/landing pages until you find copy that sells.
You also want to make sure you have the ability to get closed-loop results on your campaigns so you can track clicks / visitors / leads through to purchase, returns, add-on sales, etc. Without revenue and post-sale data, you do not have a truly accurate picture of what marketing efforts work and do not work. To accomplish this, you need to integrate your ecommerce, CRM, order management system, and accounting systems or get an integrated business management system like NetSuite that holds all of this data in one shared database.
8. Take a break.
Really, don’t knock yourself out. Improving sales (warning: overused cliche alert) is more of a journey than a destination and the test, observe, retest, observe, cycle can take months or years. Don’t burn yourself out by trying to do it all in one day, and don’t get discouraged if you can’t make a $10,000 a month in 6 months after starting. Consistent sustained improvement is the key to long-term success. Set some goals for yourself, measure progress towards them, and have fund enjoying your hard-earned successes. (I’m blowing my revenue goals for this year, sadly, but my recent round of website improvements has my conversion rates and level of understanding of my customers up nicely.)