Advice for First-Time Online Startups

I've been developing websites and apps online for over a decade now. I've built them for myself and others and so I've learned a lot from the success and failures of both myself and others.

The successes in these groups were always the ones that could afford to waste time and money. They had enough bank roll to outlast their mistakes and wasted efforts. This doesn't mean they had any greater potential than the smaller bootstrapped startups I witnessed. They could just afford to make more mistakes.

I get a lot of questions from friends, family and acquaintances who want to sell a product or service online. They usually aren't large companies with money to burn and so I feel it's my duty to lay out a map to avoiding the time and money traps. It's actually not really difficult to build a successful online business when you compare it to brick and mortar stores or traditional businesses. It just feels hard because there's so much knowledge to navigate and it's easy to lose your path and waste the limited time and money available to them.

What follows is a quickstart guide that distills all of the basic knowledge needed to get your first sale online without breaking the bank or your spirit. At that point you will have enough knowledge to start scaling, which would be another guide for another time. The trick to distilling a topic for real world results is realizing that like most things in life, it follows the pareto principle, aka 80/20 rule. 80% of your results will come from only 20% of your effort. There's no need to waste time on the things that will only bring you 20% of your results anyway.

For this reason, there needs to be a resolve when starting to not go after perfection. Rather, go after the large easy wins. Do the things that will give you results fastest and iterate over them.

Before going futher check out the appendix if you're not very technical so that you're versed in some terminology.

Before You Start

It's easy to get excited about a new idea. But the hard part is being your own cynic. Do your future self a favor and look at all the ways you can fail first. If you can navigate the potential failures, all you're left with is success. But if you focus only on the success, the 100 ways you can fail will be a very unpleasant surprise, especially once you've already invested in your project. Ask yourself:

Have you researched your competition? Is the market saturated already? How hard would it be to match what your main competitors are doing? This is called competitive analysis. If you have competitors, that's great. Your idea is a working concept. You'll just have to figure out how to one-up them.

If you don't have competitors, then do you have a proven market? Are people looking for what you're offering? If they are, are they willing to spend money on it? This would be called your market research phase. Dig deeper in your market.

If you have a proven market of people willing to buy what you're offering, and an ability to get in front of your competition so you're not drowned out, the next step is to assess your business model.

How will you attract customers? If it's SEO it will take time, do you have enough of that? If it's paid advertising, how much will this cost? Is the cost of acquiring a customer more than your product's profit margins?

A formal way of doing a lot of this is to do what's called a SWOT Analysis of your business.

If you're still excited and hopeful past this initial research phase, that's great! We can move on to the next phase, which is testing. Otherwise, keep reading how to break down your competition.

Competitive Research

If you already have a competitor in your space, don't worry that's actually great news. The concept and market have already been proven! So your task is two-fold:

Copy what they do
But do something better

What's your value added over them? Is it better shipping? Customer service? Quality? Pick one and blast that in your messaging.

If it doesn't work, pick another and try again, but don't try more than one at the same time. Your messaging will get lost.

You can find your competitors by searching for your product or service online. Become the consumer and make note of how you go about finding your own offering. Write down what you search for and where. Write down the leaders in these different categories.

Who ranks #1 in Google for your product?
Who has the most followers on which social media?
Who seems to use their newsletter most actively?
Who keeps showing up in ads in search and other websites?

Once you make note of all of this, you can dig deeper by adding them to ahrefs and get a clearer picture of what their strategies are.

Test over Build

Don't set out to build the final vision. This is a big expensive mistake. Break it down into the smallest pieces you can test. And once you do test, don't be afraid to change strategies according to your test results.

It's fine to have the orienting vision of what your final goal or product is. But DON'T let this dictate the steps you should take to build it. Let the market do that, they know best.

Understand that the vision will evolve and change every step of the way and so you should act accordingly. Things you thought would matter won't, and things you thought wouldn't matter do. That's OK. What's not ok is wasting time and money, that's why it's crucial to figure out things as quickly as possible through an iterative testing process.

Build the smallest testable steps towards your vision, not the whole thing. Stop after every test to reassess what you've learned and pivot accordingly. Don't assume you know what to build for whom, let the data you get from the market make it abundantly clear to you.

Here's some strategies for testing. The main one being:

What's Your MVP?

An MVP is a minimum viable product. It's the thing you can build with the least amount of effort and expense in order to prove your concept and collect some data.

You may have tons of ideas for features and cool things to offer your customers, but all of that takes time and money to develop and may not bear fruit. So you need to distill your offering to the 1 thing you think brings the most value to your customers.

In the case of an e-commerce shop, you may want to offer tons of products with different variations and options, with information and sizing guides, with support channels and account managment and... well you get the point. There's a lot that goes into an e-commerce shop. But if you've already sourced out a lot of good products you want to offer you can build a distilled version of this in one simple page. Say you sell unique widgets.

Make a spreadsheet of all your best widgets.
Add as much "meta" information to each. Add categories, tags, sizes, locations, photos, styles, etc.
Create an interactive search or quiz for your customizers to find the best widget that suits their needs. So perhaps they can input some information specific to their situation like their style or location or price range.
Present them with a match from your list of widgets you've already curated!

If you're lucky, you can work out a fulfillment deal with the suppliers so you make a few dollars by sending the lead to your suppliers website to finish the purchase. You're not interested in making money in this phase, you're interesting in proving the different assumptions you have about your product.

You're not going to make much money with this, but the information you get through a page like this is worth a whole lot more. All of this can be built yourself without any coding knowledge, but even with a competent developer you can get something along these lines for $2000 if you provide all the data needed.

However, with this $2000 investment, you will be able to find out:

Which marketing and advertising channels work the best
Which demographics respond best
The rate at which your visitors actually use and submit this quiz
How many users go through with a purchase with your supplier or at least click through to try and buy (establishing buying intent)
Which categories of widgets are most popular

With this information you can calculate your CPA. You can also get a good idea of what your average purchase price and profit would be. With these two numbers you can know with a high degree of certainty whether you have a viable, profitable business model that simply needs to scale. And if the results are negative, you can keep tweaking and trying different things until your numbers turn positive.

This whole experiment may cost a total of $5000 including advertising costs. That might seem expensive to some, but it's much better than the $20,000 and months of work that a full e-commerce can cost (not including the advertising costs).

How to Test

Make as many variations as you can and make sure you're always analysing the data. Test carefully by changing only one thing at a time.

For example, if you're running Facebook Ads, try the same demographics but different headlines in each to see what words and concepts your demographic best respond to. Conversely, if you're testing out your age demographics, serve the same generic ad to both groups but only vary the age you're targeting in each group. Don't try to make money from these ads, try to gather information. The money will come later when you're ready to scale.

Another example would be split testing your landing page. You would be amazed the difference in conversion rate changing the position, color and text of a "Buy Now" button can lead to. To figure this out, use split-testing software like Google Optimize to create variations of the same page and change specific things for a certain amount of time.

Marketing

At this point you should have a good idea of the potential for your online business and have gathered some numbers to point you in the right direction.It's time to focus on the one skill that will make or break you: marketing. Get good at this and everything else will be a piece of cake.

Know Your Demographic

At the core of marketing is knowing your demographic. It's essential to figure this out as early as possible because it will guide most of your decisions moving forward. To do this you must find and understand your customer profile.

A customer profile is a detailed description of your ideal customer.

Find Your Customer Profile

To find your customer profile you must first test and research the different demographics available to you. This should have already come up in testing phases with an MVP.

Don't be scared to adjust your profile early. This should be part of your early testing phases.

An Example Test

Using the same landing page you built for your MVP, run ads to different age and/or gender demographics to see which ones click on your ad most CTR. The landing page doesn't matter, but you should tweak your ads to speak to each individual demographic specifically. The key information we're looking for here is how well these groups respond. After you know which one responds best, you can tailor to them more on your landing pages, etc. Facebook specifically is great for testing your demographics because they have a lot of particular data you can target ads with.

Research Competitor Demographics

Look into who your competitors are targeting because they've probably already gone through a lot of the leg work. Use tools like ahrefs to see what keywords they target. You can also simply infer it from the messaging on their website and other marketing material. This should've been done in the competitive analysis phase.

Understand Your Customer Profile

Once you have a good idea of your customer profile, sit down and imagine the ideal customer in this demographic. Imagine yourself in their shoes and go through a typical day. Write down as much detail as you can about what they do, when they do it, who they do it with, why they do it.

This seems like a weird exercise, but once you've written all of this down you'll be able to generate countless avenues for reaching your perfect customer. If your target demographic is a middle-class stay at home mom, then realizing that everyday she probably has very busy mornings getting her kids to school is good information to know not to bother sending her an email during that time because she will be distracted. On the other hand, perhaps her quietter time is in the mid-afternoon while she catches up on her daily tasks and listens to her favorite podcast. Identifying this podcast and securing an ad in it may go a long way!

Branding

Once you know your customer profile, you'll need to create a brand that speaks to them. But don't think your brand is set in stone. It's not, it will evolve with your customer demographic and even the times. Don't try to nail this right away, just get it done. Here's the four things you'll need to get done:

A logo. A logo should be simple, memorable and readable. If you squint your eyes you should be able to recognize the brand of a good logo. That means make sure it's legible at all sizes and colors (colored on your website banner, small and grayscale on your letterhead, etc). If your name is short, you can get away with just an interesting font with some stylization of the letters. If it's longer, your best bet is an icon alongside your name in a simple readable font. Limit your colors to two at most, ideally one that's your main brand color, and make it high contrast. A logo should be delivered by your graphic artist in at least 8 combinations (2 dimensions X 3 colors) several versions to be used in different situations:

Landscape, good for website headers
Square, good for banners and stand-alone graphics
Full color
White, or light grayscale so it works on a dark background
Black, or dark grayscale so it works on a light background

A color palette. Simpler is better here. Choose a strong main color and work around that. Here's some more details on simple color schemes for web.

A styleguide. This will be essential for keeping things consistent and efficient. It's a document you can hand to any creative to orient themselves with your brand and make something that fits, whether that's a sales PDF, a Facebook banner, a video or a website. Start with this and stick to it. If the brand changes, change the styleguide first. It should include things like your logo and colors used together, fonts and examples of text with images, your graphics and example usage of everything.

Some Graphics. This can come in many forms although you don't need a huge set to start, you just need to know the type of graphics you'll be using and select a handful of good ones to reuse throughout your branding. Types of graphics include photo realistic, nature shots, textures, faces or action shots, icons, illustations, etc.

Growth Tactics

In order to make sales, you'll need to grow your exposure to customers. There are inumerable ways to do this and some will lend themselves better to your particular case. These are the ones that have proven to be most effective (in descending order) for most cases in my experience:

Freemium Model
Content Marketing
SEO
Photo / Video Production
PPC
Influencer Marketing
Social Media
Email Campaigns
PR Campaigns
Offline Advertising
Referral Programs

Note that the first 4 fall under the same general category and work together:

Make something valuable for free (lead magnet). This will attract backlinks which will...
Make it easier to rank in Google and build a lot of organic traffic from SEO...
from all of the content that you have on your website. This includes...
Articles, photos, videos, audio, and pretty much any form of media that's easiest for you to create.

This is my go-to tactic for most new websites.

Lead Magnets

You will inevitably need something to draw in customer's attention. Ideally it's your main value proposition and that's enough. But sometimes you need a bit more. The "freemium" model is a great place to start. The most resistance with your customer will always come when you try to separate them from their dollars, so sometimes it's quite difficult to attract them and introduce them to your business through this avenue alone. A great tactic is to offer something of value to them for free. You establish yourself with your customers, build an audience and provide something that is much more shareable. Simply make it known that you also sell something and the sheer volume of your audience will provide the sales.

Ask yourself if you can build something of value for your customer audience and give it away for free. If it costs to build, simply consider it a marketing expense and treat it as such. In my experience, these types of expenses have a much higher ROI than any traditional marketing means.

Next Steps

At this point you've:

Tried to destroy your business model
Have done competitive analysis and market research
Tested your marketing channels, demographics and viability of your product with a series of tests around an MVP.
Made your first sale, or established with a good amount of certainty what your CPA and profit margins would be.

With this information, you're ready to reinvest and scale! This is the point where you see the profits from all the foundational work you've done in this article. Luckily, you shouldn't have wasted much money or time and you'll be much more certain how much you have to invest to make more in the next more profitable phase of scaling your business.

Appendix

Vocabulary To Know

Marketing / Business Terms

CPA: cost per acquisition, ie how much you need to spend to make acquire a customer.
ARPC: average revenue per customer, ie how much does your average customer spend per month or year
CTR: click through rate, ie how many people click on something for every 100 people that see it.
SEO: search engine optimization, ie optimizing your site to appear higher in Google search results. This type of traffic is called "organic" as opposed to "paid" traffic you can receive from PPC/SEM.
Keywords: refers to the words that are used when you search for things in Google. These "keywords" are essential to know and build your website around because it's what your customers are looking for. When someone looks for a keyword in Google, you want to rank 1st to get the most traffic from these intersted customers.
Backlinks: are links from other websites to yours. This is a strong indicator to search engines that you have a good quality website and so they're more likely to rank you higher the more backlinks you have.
PPC/SEM: pay per click adversiting like Google adsense. This falls under the advertising category of "Search engine marketing".
Analytics: data about your customers and business. You'll need to track and / or survey your customers to make informed decisions.
Freemium: a business model where you give something for free in order to build the audience that will eventually purchase something else from you.
Bounce rate: how many people out of 100 leave your site without exploring (lower is better). This is a good metric to know whether you're targeting the right interested people.
ROI: return on investment. This is the king of metrics. If you're not making more than what you're investing in your business, you need to make a change.
Content Marketing: is a great growth strategy if you want to focus on organic traffic from search engines. Write and produce a lot of valuable content through articles, videos and any other media.
Split Testing: is the process of testing variations of the same page to see which factors perform the best.
Conversion Rate: how many people out of 100 perform an action you want them to.

Website Terms

URL: the address of a webpage. This should be unique.
Webpage: a single page on a website, represented uniquely by a URL used to access it.
Website: a collection of webpages that usually all live on the same domain.
Domain: the address of your website. Buy this through a Domain Registrar
DNS: this is more or less the address book for your domain. It's where you map the different names for your domain to different servers like your web or email hosting.
Performance/Pagespeed: is an important metric of how fast your webpage appears to load. For every second your page takes to load, expect to lose exponentially more customers.
Header: the portion of your website that is more or less always the same at the top of the page. It almost always includes your logo and navigation. Put important things you want people to see here.
Footer: this is like the header but at the bottom of the page and for less important things that didn't have any context in the rest of the page. Often your less important pages go hear with legal babel.
CMS: content management systems like WordPress, Shopify and any platform that allows you to make updates to your product or website content.
Landing Page: a landing page is the page your visitor first encounters. This is often the homepage but not always the case. It's useful to make a specific page at yourcompany.com/your-demographic and cater the content towards a particular demographic that you may be driving traffic to with ads.
CTA: is your call to action. This is the thing that's getting your customer's attention and trying to get them to perform an action. For example, a "buy now" button. There should only ever be 1 per page and it should draw attention to itself.

Tools to Know

Domain Registrar: This is where you register your domain. I recommend: cloudflare.com, google.domains and gandi.net in that order.
ahrefs: this is an indispensable tool that's well worth the price. You can use it to analyze: your competition, your own market, your rankings and progress. If you're planning on using SEO as your marketing stratey it's pretty much mandatory.
Google Search: is an invaluable research tool. Look up your product or service and pay attention to what every one else on the first page of results are doing (and not doing). This includes the sponsored ad results as well.
Google Analytics:
Google Search Console: another free tool from Google that tells you what to do for better rankings. It will also tell you what keywords you rank for.
Google Optimize: for split testing
Google Fonts: great place to explore high quality open source fonts to use in your brand. Use the text preview to experiment with your logo text. All fonts are free to use in your media.
Brand Monitoring: is an important category of tools you can use. With these you can monitor mentions on the internet of your brand or products so you can reply to people directly in key moments. This includes managing your reputation with reviews. Do your own research for which ones to use for your case.