Book - Start small, stay small: a developer's guide to launching a startup

  • Metadata:
    • author: Rob Walling, Mike Taber
    • title: Start small, stay small: a developer's guide to launching a startup

Unprocessed highlights and reading notes (turn them into separate notes when revisiting the topic):

  • Point 1: An entrepreneur is a technical visionary who creates software for a niche market. Niche markets are critical. If you want to self-fund a startup you have to choose a niche.
  • Notes: 1) These niches are too small for big competitors, so you have room to breath as a small fish
  • Point 2: An entrepreneur merges existing technical knowledge with online marketing knowledge. The key factor in an entrepreneur’s success is their ability to market their product.
  • Webmaster These days it’s imperative that your startup knows how to sell online. This means creating a website that converts visitors into customers.
  • Marketer Marketing is more important than your product. Let me say it another way: Product Last. Marketing First.
  • building something no one wants is the most common source of failure for entrepreneurs.
  • Notes: 1) This is similar to the idea of invention vs innovation
  • Without a market, a software application is just a project.
  • A study at Dominican University3 revealed that the following 3 factors substantially increased someone’s chance of following through on their goals: 1. Written Goals – “Those who wrote their goals accomplished significantly more than those who did not write their goals.” 2. Public Commitment – “…those who sent their commitments to a friend accomplished significantly more than those who wrote action commitments or did not write their goals.” 3. Accountability – “…those who sent weekly progress reports to their friend accomplished significantly more than those who had unwritten goals…”
  • I have a suggestion to help get you started: Strive to build a startup that generates $500 per month in profit.
  • Notes: 1) I might get to this point in a year or so
  • Roadblock #1: No Market This is by far the most common mistake
  • Roadblock #4: Inconsistency The main problem with inconsistency is that it makes you lose momentum and momentum is critical to staying productive.
  • Notes: 1) I have experienced this one a lot
  • How to Avoid It You’ve likely heard of the concept of an information diet. The idea is that most of the information we consume is a waste of time. Newspapers, magazines, blogs, podcasts, the news…are all enjoyable to consume, but they have a tendency to offer a constant distraction from real productivity. You can’t consume and produce at the same time –
  • These are trivial examples of what I call drip outsourcing; outsourcing small tasks as I perform my daily work.
  • Notes: 1) Similar idea to the 4 hour work week
  • Dollarizing The phrase “dollarize” is used in sales to describe the approach of showing your prospect how your price is less expensive than your competition due to the amount of money they will save in the long run.
  • Putting a value on your time is a foundational step in becoming an entrepreneur, and it’s one many entrepreneurs never take. Skipping this step can result in late nights performing menial tasks you should be outsourcing, and an effective hourly rate slightly above minimum wage.
  • While you won’t be earning anywhere near $50/hour when you begin building your product, once you launch you should aim to hit that number within 6 months.
  • At this very moment am I making progress towards crossing off a to-do, -or- am I relaxing and re-energizing?
  • Realization #4: Information Consumption is Only Good When it Produces Something
  • Notes: 1) Or when it is fun
  • Consuming and synthesizing are very different things; it’s easy to consume in mass quantity. It’s much more difficult to synthesize information.
  • Realization #2: Market Comes First, Marketing Second, Aesthetic Third, and Functionality a Distant Fourth
  • Realization #6: You Will Never Be Done Finishing a software product is a great feeling. The night you roll the new bits to the production server is indescribable. The feelings of relief, joy, and accomplishment are some of the most rewarding parts of developing software. And you’re never going to feel that way with your product.
  • Realization #8: Process is King Documenting repeatable processes for anything you will do more than once is essential to your sanity.
  • Even then, support and feature development is the easiest part to outsource; marketing is one of the most difficult.
  • that they have found a group of people willing to pay for your software.
  • Reason #1: A Niche Requires You to Narrow Your Product Focus Here’s an exercise: Pick one person you know well…your spouse, your brother, your sister, a parent, etc. How hard would it be to design a product that you’re sure this person would use…not very hard, right? That’s because you know so much about the intimate details of their life.
  • Reason #2: Niche Advertising is more Cost Effective
  • Odds are high that if a publisher started a magazine aimed at a niche, and full-page ads are less than $5,000, it’s a good niche to explore.
  • Advertising to this smaller audience is going to garner more “actions per view.”
  • Notes: 1) This got a lot harder with the new privacy changes to ios
  • Reason #3: Niches Have Less Competition
  • Notes: 1) Operating in small nieches is not profotable for big players
  • Reason #4: Niches Have Higher Profit Margins
  • Reason #5: Niche Markets Are Not Used to Good Marketing
  • Notes: 1) The same should go for aesthetics and features
  • Exercise A “warm” niche is a niche where you have some kind of association. Perhaps you worked for a credit card company for a few years, your wife is a lawyer, you collect comic books, or your brother is a plumber. Each of these would be considered a warm niche,
  • Now for each row write the name of someone you know, including yourself, friends, relatives or colleagues, and write their work experience or hobby in the right column. These are your warm niches.
  • Pamela Slim, of the Escape from Cubicle Nation blog19 and podcast20, has a six-minute look at choosing a business idea21 that I highly recommend. Her process takes some time over the course of a month, but it’s worth the effort to gather information about your interests.
  • in recent years as consumers have become more and more immune to interruption marketing (that is, marketing that tries to interrupt someone during their daily life rather than speak to them at the exact moment when they are searching for a solution to a problem).
  • Notes: 1) Studies?
  • The idea is that if a market has a magazine devoted to it, it’s large enough to provide enough customers and if a full-page ad is less than $5,000, the market is small enough that you’ll be able to effectively market to it.
  • Your “market size” research should involve the following: • A web search for related magazines and websites • Tracking down their advertising rate cards, which tell you how much it will cost to advertise in their magazine or on their website • Reviewing their circulation (for magazines), visitor statistics (for websites), and rates • Entering websites without rate cards into Compete.com22 to check traffic levels • Checking in with your friendly neighborhood bureau of labor statistics for a sanity check on your findings For
  • Be wary of a market if you are unable to find a handful of websites or a magazine/journal dedicated to it, and the labor statistics claim less than 10,000 members.
  • “If I could get only on the front page of [small-but-very-focused-niche-website].com.”
  • The Top Shelf approaches focus on two key areas: building an audience and search engine optimization.
  • Conversion Rates - The ratio of people who buy vs. visit your website • Traffic Levels - A look at the level of traffic you might need to succeed in a particular niche • Traffic Breakdown – A look at common percentages that different traffic sources might provide to your site Everything You Need to Know About Sales You only need to master two skills to sell online: human behavior and math.
  • If your price point is in what I refer to as the consumer range of $1 to $50 and your product is priced appropriately for your market, your conversion rate should be between 1% and 4%.
  • If you’re priced between $50 and $1,000 or offer recurring pricing and your product is priced appropriately for your market, you’ll most likely convert between 0.5% and 2%.
  • Traffic Breakdown You can divide website traffic into four categories: • Search Engines – Organic searches on Google, Yahoo!, etc… • Incoming Links – Organic links from blogs, directories, or any other website. • Direct Traffic – Someone types your URL into the address bar of their browser. • Advertising – AdWords, banner ads, etc…
  • I’ve been involved with have settled down to receiving around 1/3rd of their traffic from search engines, 1/3rd from incoming links, and 1/3rd from direct traffic.
  • With this in mind, if we need 1,000 visitors at a price point of $200 to gross $1,000 per month, we need to bring in around 333 visitors per month via organic search. For safety let’s round up to 500 visitors. If we can rank high enough for enough keywords to garner 500 organic search engine visitors each month, it’s a pretty good bet the other two categories (incoming links and direct traffic) will fall in line.
  • the best place to start is with your contact for the niche (the person in the right-hand column). Give this person a call and grill her on what pains her in the job or hobby that could be solved with software. You’ll probably find at least a few ideas; everyone loves to talk about their daily woes, and especially about how awful their software is.
  • Google AdWords Keyword Suggestion Tool30 • SEO Logs Keyword Difficulty Tool31 Or the following tool which costs $97 (one-time fee): • Micro Niche Finder32 – The marketing for this tool is bad, but the tool itself is extremely powerful.
  • The Steps to Measuring Demand We’re going to look at two approaches for measuring demand: the free approach and the approach using Micro Niche Finder. The Free Approach Step 1: Google AdWords Keyword Tool Visit https://adwords.google.com/select/KeywordToolExternal Enter your main phrase, without quotes, into the Google AdWords Keyword Tool. Include synonyms, and run the search. Once the search results display, change the search type to “exact match.” Run the search again. Exact match limits your search to the exact phrase someone entered, not a variation of that phrase. For example, if someone searches on “blue running shoes” they will show up on a broad match for “blue,” “running shoes,” “blue running” and “blue running shoes.” But they are an exact match only for the exact phrase they searched for: “blue running shoes.” The results you see are approximations of how many people search for this term,
  • The Micro Niche Finder Approach Using Micro Niche Finder, the approach is a one-step process. Enter your keyword and search.
  • Note: You want the strength of SEO competition (SOC) to be low; under 100 is good. Over 1,000 is tough competition.
  • depending on the cost of AdWords in your niche and how far you want to take it. The
  • The Approach We’re going to look at an approach to testing your market called the Mini Sales Site. I developed this approach a few years ago after reading the 4-Hour Workweek, realizing that the author’s testing approach could be re-purposed from information and physical product testing to software products.
  • • Product - Your product has to be good • Market - You need a group of people willing to pay money for it • Execution – You have to market, sell, and support it
  • Estimate 4-12 hours per page depending on complexity. • Add 10-20 hours for DB design depending on complexity. • Estimate back-end functionality such as credit card processing, PayPal integration, or scheduled tasks. Add 10-40 hours each, depending on your experience. • Add it up.
  • As a rule of thumb, your path to 1.0 should fall between 200 and 400 hours.
  • The graphic design and HTML will run from $500-$1,500 if you offshore. Handling it in the U.S. will cost $2,000-6,000.
  • For consumer products dealing with hobbies (i.e., products that aren’t going to make or save someone money), you’re going to have a tough time charging more than $29 fixed-price or $14/month. For consumer products that will make or save someone a tangible amount of money you’re going to top out around $49 or $19/month. For small businesses you’re going to top out around $400 or $99/month unless you solve a serious pain. For larger businesses your top end will be around $1,00037 or $199/month unless you solve a serious pain.
    1. Ask yourself. When 37Signals determines pricing for a new product38 they ask themselves the following two questions:    a. What would we pay?    b. What numbers feel right? 3. Look at your competition. If you have competition this step is easy. With little or no competition you’ll need to compare to similar markets. Writing inventory software for small businesses? Look at invoicing or accounting software for small businesses. 4. Determine your product’s value. If your software saves your prospect 5 hours per month and you can monetize their time at $25/hour you will save them $125 per month. You won’t be able to charge more than this figure, but you can use this as a top-end of your potential price. 5. Combine. Use the numbers from steps 2-4 to determine the optimal price range for your product. Ideally you want the high end of your range to be 4x your low end. 6. Lean towards higher pricing. Developers tend to undervalue their software, and think that lower prices will result in higher sales. This is typically not the case.
    1. End in 7, 8 or 9. It’s stupid, but it works. Make each of your three prices (the dollar column) end in a 7, 8 or 9 and be sure they all end with the same number.
  • For hosted applications the question is not “when to use” it’s “when not to use.”
    1. First, that person must see your URL. This may be a link in a blog, a banner ad, a write-up on TechCrunch, an AdWords ad, or a link in a forum. 2. Second, they must click the link and become a site visitor. 3. Next, they must be interested enough to stick around for more than a few seconds; long enough to read your sales copy, watch your video demo, try your product demo, and potentially provide you with their email address. At this point they have “raised their hand” as a potential customer and they become a prospect. 4. Finally, they must be convinced that your product is going to solve their problem for the right price, and they must make the purchase. Here they become a buyer.
  • At each step you will lose the majority of the group:
  • Don’t Plan to Sell to a Customer on Their First Visit The first rule of sales websites is that the most common approach is wrong. That is: You shouldn’t plan to sell to a customer on their first visit
  • The Number One Goal of Your Website Your number one goal, even beyond selling your product, is turning browsers into prospects.
  • On the web, this is typically achieved by asking someone to provide their email addresses.
  • Establish Trust – Your visitor must believe that you aren’t going to spam them, sell their email address, or send offers for V1@gra.
  • Establish Relevance – Your visitor must believe that your product is relevant to their need
  • Establish a Reward – We are predictable creatures.
  • In a successful sales website, every page has a single, primary call to action.
  • Any page they land on should include a prominent call to enter their email address, and perhaps a secondary call to purchase your product (or learn more about it).
  • As a result of rule #1, every page needs a call to action. A visitor may first interact with your website through your Tour, Testimonials, or Pricing page.
  • Every page needs a single focus.
  • Rule #4: Everything should be within 2 clicks.
  • Rule #6: Make buttons look like buttons. Make your buttons so clickable that people can’t help but click them.
  • Rule #7: No One Reads. Text is a terrible selling tool; audio, video and images are always better.
  • The #1 goal of your home page is to convince your visitors to click 1 link. That’s all you have to do to convince them not to leave is click a single link.
  • One of the most difficult pieces of marketing to create is your hook. Your hook is your four-second sales pitch and it should be the headline of your home page. It’s the single sentence that grabs the reader in and makes her know she’s in the right place. • DotNetInvoice’s hook is “Save Time. Get Paid Faster.” • FogBugz’s hook is “Bring Your Project into Focus”
    1. Explain what your product does and for whom. Such as “Simple proposal software made for designers.” 2. Make a promise to the customer espousing a benefit of your product, such as “Save Time. Get Paid Faster.” 3. Describe the single most remarkable feature of your product, such as “One Thousand Songs in Your Pocket.”
  • Testimonials This page can also be titled “Buzz” or “Who uses [Your AppName]?” and it’s one of the most important pages on your site. Do not launch without testimonials.
  • Contact Us It’s best to provide both a contact form in the browser and a separate email address.
  • offer a 30-day, no questions asked, money back guarantee. It may sound like crazy talk to do this with something like software, but the number of people who try to rip you off will pale in comparison to the number of people who purchase the product because of this guarantee. I’ve had a money back guarantee at DotNetInvoice for years and I’ve never regretted it.
  • Notes: 1) Similar idea in 4 hour work week
  • If you sell beach towels, offer a free report on the Top 10 Expert Tips for Saving Money on your Beach Trip (an actual report I gave away on JustBeachTowels.com). I recommend shooting for a PDF report from 5-15 pages.
  • one provider I recommend is MailChimp58.
  • What to Send There are two types of mailing lists: Relevant Information
  • A powerful tip I’ve used extensively is to use an autoresponder series, which is a series of emails sent in a specific order to new subscribers with pre-determined gaps between each email. With an autoresponder series you can write content once and have it be re-used for years.
  • Product Updates This is the type of list we run at DotNetInvoice.
  • This is why we’ve elected to use our email list to provide product launch updates and to ask for peoples’ opinions on the future direction.
  • Idea #4: Hire a Writer Hire a writer from Elance59 and have them write an original short-form article on a subject of your choosing. It will cost $20-30, but if it contains unique information, it can make your subscribers happy, and may also get your email forwarded, which will grow your list.
  • Tuesdays, Wednesdays and Thursdays are the best days, between 7am and 10:30am.
  • Your Subject is Your Headline - Perhaps the only factor that determines if your mail gets opened
  • a. The shorter the better. b. Ask a question in your subject and answer it in the emails. Example: A DotNetInvoice Super-Bowl Ad? c. Make a partial statement with “…” at the end and continue the sentence in your email. Example: A Free Copy of DotNetInvoice Every Day… d. Use the recipient’s first name in the subject line. Example: Rob, DotNetInvoice is Free for 24 hours… e. Include your product’s #1 benefit in the subject. Example: Save 5 Hours of Time This Month f. Don’t exaggerate. Deliver on anything your subject promises. g. Avoid spam filters by watching your spam score. MailChimp will parse your subject and body and tell you if your email looks spammy.
    1. Have One Goal for Each Email – Is this a relationship-building email filled with information, or do you have a call to action for it?
  • Step 1: Six Months Before Launch Six months before your launch date, use your audience on Twitter, Facebook and your blog to send traffic to your startup’s home page. The home page gives a brief description of your product and offers them a deal if they provide their email address to be notified of the launch. Step 2: Over the Next Six Months Over the next six months create buzz around your product by guest blogging, sending interesting updates to your social networks, commenting on blogs and forums and generally engaging your target audience, always sending them to your home page. Step 4: One Week Before Launch Approximately one week before launch (preferably on a Tuesday, Wednesday or Thursday), email your list of targeted email addresses. Let them know that your product will be launching next week and that since they are on your mailing list, they will receive a special price available only to those on the list, but that the price will only last for 48 hours. Tell them the day and time they will receive the email. Step 4: Launch Day On launch day, email your list. Almost immediately, sales will start rolling in. You’re going to have the best sales day you will see for a while. Conversion rates on targeted mailings can be 20%+. A few hundred sales of your $19/month SaaS application is not a bad way to kick-start your startup. Step 5: Thirty Six Hours After Launch Send your list a final email informing them that the deal will end in 12 hours. You will receive another few sales before you close down your special pricing.
  • While there is value to Twitter in terms of customer service and being active in the online community, this chapter is going to focus on tactics that will drive targeted prospects to your website that have the potential to result in sales.
  • Many approaches will not work in your market.
  • Top Shelf: Traffic Strategies that Will Sustain a Business 1. A Mailing List 2. A Blog, Podcast or Video Blog 3. Organic Search Second Shelf: Supplemental Traffic Strategies 1. Social Media / Social Networks 2. Pay-per-click Advertising 3. Forums 4. Press Releases
  • Strategy #1: A Mailing List A mailing list is the most effective marketing tool you will possess. It works in any market. It’s a marketing requirement for startups.
  • Does Every Product Need a Blog? Most products you launch should have their own blog, if for nothing else than to draw search engine traffic.
  • Guideline #3: Build Links Over Time Did I mention search engines are smart? If you build 120 links in a month, or 10 links per month for a year, the second scenario is going to play out far more in your favor.
  • Approach #1: Directories
  • DMOZ (www.dmoz.org) – Free. Although its importance has diminished over the years, it’s still worth submitting to.
  • JoeAnt (www.joeant.com) – $40 • Gimpsy (www.gimpsy.com) – Free for a 6-month turnaround, and increasing in price to $49 for a 3-day turnaround.
  • Using Yahoo! Site Explorer70, look at your competitors’ backlinks and determine if it’s appropriate to email the linking site’s owner and ask for your site to be added to the list.
  • in a bio box. • iSnare costs $2 to submit to thousands of article repositories. While you don’t get much benefit from being in thousands of repositories, the real benefit happens if your content is reprinted elsewhere on the web. At that point, contact the webmaster of that site and offer to write unique content for them. That will provide unique content and a backlink to you. • For articles, 500-700 words is best. You need 500 to be over the minimum length requirements, and over 700 words most people won’t read it.
  • Approach #6: Offer a Free Academic/Non-profit Version of Your Product in Exchange for a Link Do some good for the world and get a link out of it. The best part is that academic and many non-profit organizations have authority status in Google because they are long-standing websites and don’t link to many commercial sites. Their links can be easily worth many times that of a link from a typical website.
  • Why Should My Startup Use a Virtual Assistant? Outsourcing to a virtual assistant will dramatically reduce the time you need to spend on administrative tasks, and increase the time you can commit to growing your business.
  • If you monetize your time at $50/hour and you can pay a VA $6/hour to handle administrative tasks, this frees up time for you to create real value in your business by developing new features or expanding marketing efforts.
  • The Lesson The lesson is that before you launch your product, what are some processes you can avoid automating? How about reminder emails? How about monthly billing? Could a human being run a report once a month and send emails or charge credit cards?
  • Every hour spent writing code is wasted time if that code could be replaced by a human being doing the same task until your product proves itself.
  • Point #2: After Your Product Launch The next most important time to use a VA is once your product has launched and you need to begin supporting customers. Customers make it necessary to put processes in place for marketing, sales, support, and back-end administrative tasks. Any ongoing work that can be described in a written process can be outsourced to a VA and save incredible amounts of time for the founders.
  • I tasked my VA with finding blogs that deal with startups/microISVs and rank in the top 100k in Technorati. The deliverable was a Google spreadsheet containing the blog URL, blogger’s name and blogger’s email.
  • Easing Into a VA Outsourcing is a learned skill, just like writing code. If you rush into it too quickly, you’ll wind up disappointed with the results. This is most often due to the fact that you don’t yet know how to work with a VA.
  • Task-based – ($3-10/hour overseas, $12-50/hour in the U.S.) You assign your VA an individual task and give them a deadline and maximum time to spend on the task. Since your VA works for other clients, they are in charge of prioritizing all of the tasks they receive. Task-based VA’s are a great starting point to learn the ropes of delegating.
  • Notes: 1) Use virtual assistants to label drawing data for ml
    1. Find a VA 2. Start with a single task and gradually increase the amount of work as you gain comfort 3. If things don’t work out, find a new VA
  • but the Filipinos learn English in school, do not tend to be entrepreneurial (thus are less likely to steal ideas), and are culturally service-oriented.
  • Search ODesk73 under Admin Support -> Personal Assistant or Other.        Search Google for “virtual assistants.” Typically the best looking websites are the firms that have their act together.        Search Elance74 under Admin Support -> Admin Assistant.
    1. Part-time VAs        Search ODesk under Admin Support -> Personal Assistant or Other.        Search Google for “part-time virtual assistants”
  • If you’re looking for general help, the only noticeable difference between the 10 VA’s you are screening is their hourly rate and their ability to speak and write English.
  • The best way I’ve found to evaluate English skills is to email back and forth a few times, asking 2-3 basic interview questions. This will be a good indication of how well they will be able to understand your instructions, and their responses are a good indicator of how well you will be able to understand their questions. The best approach is to email with 3-5 VA’s at once to speed up the process.
  • Step 2: The First Task Properly utilizing a VA is a learned skill. Very few developers will do it right the first time, which leads many who try it to give up after the first attempt. To keep you from falling into this trap, we’re going to look at the best way to delegate, describe and limit tasks in the section below.
  • The following is unlikely to work: Open a command prompt and type ‘ipconfig’ But this should: In your start menu go to the Run menu, type ‘cmd’ and hit enter. Once the window opens type ipconfig and hit enter.
  • Screenshots help enormously. Screencasts are even better. I record multiple screencasts each month for my VAs. Jing75 is perfect for this.
  • Timebox your requests. As an example, let’s say you have twenty blog URLs and you want your VA to find the contact information for each one (whether it’s an email address or a contact page). Provide the list of URLs to your VA and indicate they should work for 1 hour and then update you on their progress.
  • Option 1: Grow It Most startups that achieve early success find themselves in the chasm. That place Geoffrey Moore talks about in his seminal work Crossing the Chasm76, where early adopters are using your product and you are trying to solve the puzzle to get to mass adoption.
  • Every time you receive a support request, your first thought should be “how can I make sure I never receive this question again.”