Article reading time: 19 min
Book reading time: 3 hours and 52 min
What follows is a summary of the book “Traction: How Any Startup Can Achieve Explosive Customer Growth” written by Gabriel Weinberg and Justin Mares. The book is an exhaustive guide on the nineteen traditional marketing channels startups can use to start selling their product and grow.
I give the book a 9/10. If you plan on building a company and don’t know yet how to get customers, this book is a must.
Find a summary below.
Chapter 1: Introduction to Traction
Traction is a sign that your company is taking off. Traction is growth.
Traction for a free product means that your user base is growing. On a paid product, it means that your customers are buying.
Traction trumps everything. Without traction, you don’t have a business.
There are nineteen different channels through which startups can grow. Many startups tried several channels until they found the right one.
- Most founders only consider using traction channels they already know. That means that far too many people focus on the same channels instead of focusing on the right ones. They are biased.
- It’s impossible to predict which channel will work best. You have to run tests.
Get one channel working that your competitors dismiss, and you will grow.
Here are the nineteen channels:
- Viral marketing: encouraging your users to get other users to sign up.
- PR: appearing in newspapers, TV, traditional media.
- Unconventional PR: Richard Branson in his balloon, celebrity marketing, etc.
- SEM (search engine marketing)
- Social and display ads: ads on social media
- Offline ads (TV, radio, billboards, newspaper, flyers, etc)
- SEO: Search Engine Optimization
- Content marketing (blogs)
- Email marketing
- Engineering as marketing: fixing problems for yourself, then selling/giving out for free that solution to other people that have the same problem.
- Targeting blogs: writing content in blogs where your potential customers hang out.
- Business development: partnering with other companies.
- Affiliate programs
- Using existing platforms: Facebook, Twitter, Reddit, TikTok, IG, YT, the Apple App store, etc
- Trade shows
- Offline events: small meetups, large conferences, webinars, festivals, etc
- Speaking engagement: speaking at conferences
- Community building: building communities around your product and services.
Chapter 2: the Bullseye Framework
The Bullseye framework helps you choose among one of these nineteen channels. The process is in five steps:
- Focus on what works.
Come up with a way to advertise for all of the nineteen channels. This step helps you beat the bias you may have for one channel or another. To do so, build a spreadsheet with nineteen columns, each with a different acquisition channel. Afterward, write in each row ideas of how to advertise using that channel specifically. Add for each acquisition channel:
- the probability that it works
- the expected cost
- how many customers can you expect to acquire at that cost
- the timeframe needed to run the test.
Get another spreadsheet and rank each of the channels in three different columns:
- First column: the most promising right now.
- Second column: some that are less promising
- Third column: the ones that are the least promising
If you have more than three channels in the first column, then get rid of enough channels until you only have three to rest.
Test to find out which channel is worth pursuing.
To know that, you should pay attention to:
- How much it will cost to acquire customers
- How many customers do you think are available through this channel
- Are the customers you’re getting the ones you need right now?
Find the channel that works best and use it. Keep on experimenting to find what works and what doesn’t. At some point, an ideal channel will stop working as it should, and so you’ll have to choose a new one suiting the situation where you are then.
Research how startups like yours usually get traction. Go talk to founders who failed (or succeeded) at what you are trying to do.
Chapter 3: Traction Thinking
All startups have a product. What differentiates successful from failing startups is the number of customers. The number one cause of startup failure in startups is a lack of customers.
This is why you should focus on getting customers as much as you focus on developing the product.
A startup with a crappy product and a lot of customers can survive and improve the product. A startup with a great product and zero customers will die.
Most startups think they only need to build something people want. It’s not enough as you need to make money from. Sometimes, there is no market for your product, the market is too small, too hard to reach, or too competitive market.
The product trap: fallacy expressed when a startup thinks the product alone will attract customers and fail to build traction as they build their product.
This justifies the 50% rule: spend 50% of the time on getting traction, and 50% on product development. Get them to feed each other (adapt the product according to market feedback).
If you first focus on the product then only traction, you may need to go through yet another cycle of product development.
Working on a product is done in three phases:
- Phase 1: make something people want.
- Phase 2: market something people want.
- Phase 3: scale your business.
In the Phase 1, you should work 50% of the time on your product, and 50% of the time on getting traction. You will most likely NOT get traction in a scalable way.
As Paul Graham explains: “in the beginning, you are forced to do things that don’t scale and recruit your users manually: calls, send emails, do speaking engagement, write guest posts…”
Startups that take off, take off because founders make them take off.
Must-do of Phase 1: get the first few customers.
Phase 2 is reached when you reach market fit. Your product works and your customers like it. You can then position yourself in the market and boost your marketing. Must-do of Phase 2: getting enough customers so you can be cash-flow positive.
Phase 3 is when you have an established business model, a significant position in the market, and a focus on scaling and dominating. Must-do of Phase 3: increase your earning, scale your marketing, build real sustainability.
Traction channels will evolve along the different phases. Growth is like an exponential curve. The beginning is slow, but as time goes and traction channels open, the number of users explode, then the channel becomes saturated, and you need to find a new channel to unlock.
The needed traction to get financed depends on your market and the competitiveness. Furthermore, you should talk to investors that understand what you are doing, so they’ll be more lenient if you don’t have too much traction.
A startup can be awesome if you believe in it. If not, it can get old pretty quickly.
Adopters VS outlines: The first likes your product, the second likes you as a person. Both of these people will be your first customers.
How to choose whether to give up or not?
Find bright spots (people that really engage with your product) and find out why they like your product. See if you can expand them. If you can’t, it’s time to pivot. Also, find out if you are too early or not, and if yes, why you would be.
Chapter 4: Traction Testing
Step 1: test channels in your inner circle.
Step 2: when you find one, tweak it until you find something that works.
The law of crappy click-through rate: at some point, any channel will be oversaturated. To combat that, you should be continually experimenting with testing different marketing channels and strategies, find one that works before it becomes saturated, then hop onto the next one.
Focusing on one channel and optimizing takes time and resources. You should only do it AFTER you tested and got confirmation there was potential.
You optimize a marketing channel by running split tests. Making split test a habit will improve your efficiency in a traction channel by 2-3x.
Embrace online tools to understand the efficiency of your ad strategy.
As you are running with your three different channels, you should add the numbers in a spreadsheet to compare.
Metrics to keep in mind:
- Cost per customer acquisition
- The lifetime value of a customer
- Cost per click
- Buy/click ratio
Chapter 5: Critical Path: Deciding What To Work On: The One Thing
What you choose to do should directly be related to getting traction.
You should always have a traction goal to work towards. It can be 1000 free users, 100 paying users, or 10% more users.
The critical path is the path that defines reaching traction goals with the fewest and minimum steps. Work on the first step and nothing else until you reach it. Btw, your original plan to get traction will likely end up wrong.
Define the critical path of your company as much as you define the critical path of your employees.
Three reasons why founders ignore some potential traction channels:
- They are out of their field of vision.
- They refuse to consider some channels they see negatively, like sales or affiliate marketing.
- Bias against schlep (annoying things that seem to be time-consuming).
Which traction channels are you currently bias for or against? Make sure you take each channel seriously.
When you consider weird or untried channels, you often end up competition-free, and can grow quickly.
We will now have a look at the nineteen traction channels.
First Traction Channel: Viral Marketing
Getting existing users to get their friends to use your product. What fueled Twitter, Facebook, and WhatsApp. However, as good as your product may be, it is unlikely that true viral marketing only works by itself.
To use this channel, you have to create viral loops:
- A user uses a product.
- That user tells their friends about it
- They become users and tell their friends about it.
- It repeats.
There are several ways to achieve viral marketing.
- Word of mouth
- Inherent virality: network effects
- Collaboration: valuable on its own but even more so as you encourage adoption (google docs).
- Incentives: Dropbox gives you more free space if you get other people to sign up
- Embedding content: Reddit and Youtube
There are two factors that drive viral growth: viral coefficient and viral cycle time.
Viral coefficient, k, is how many additional users you bring for each user. k=i x “conversation percentage”
i is the number of additional users that are invited, and cp is the number of users that sign up.
Eg: a new user sends an invite to three of his friend, and 2 sign up.
k= 3*(2/3)= 2.
Any viral coefficient above 1 is exponential. Any above 0,5 helps you immensely.
Viral cycle time: a measure of how long it takes a user to go through the loop. The shorter, the better.
It is important to create target number of users, and work to reach them.
Loop design: all of the ways your customers can get into the loop: ads, Facebook pages, tweets, etc. Draw a map of the entire process and cut off unnecessary steps to decrease friction.
To be successful, users need to like and repeatedly use your product.
Distribution mechanisms: the more common are email and social media platforms. Try new platforms where it is not so crowded yet. Marketing is constantly becoming more saturated.
Invitation: when users that signed up invite their friends. Invitations are short, and sometimes contain incentives.
Undertsnading why people are clicking on your link will help you improve your invites.
Things to test:
- Button VS text links
- Location of your CTA
- Size, color, and contract of your action buttons
- Page speed
- Site copy
- Signs of social proof (logo of media brand, pics of people, companies)
- Number of form fields
- Allowing users to test the product before signing up
- Ease of sign up (FB login, for example)
- Length of the sign-up process
All optimization are worthy to test: sometimes, changing one word in the headline can have a big impact.
Viral pocket: where users grow fast (Eg: growing fast in South-Africa and nowhere else).
Seeding: get new users in your viral loop.
Viral mistakes not to make:
- Adding viral features to a product that is not viral
- Bad products not adding value going viral
- Not enough split testing
- Not understanding how users share your product
- Not getting coaching from people who have done it
- Considering virality as a tactic and not as an important part of the strategy.
The best way to create a viral loop is to copy someone else’s and tweak it until it works.
Second Traction Channel: PR
Start small when targeting media outlets. Big media read small media and pick up stories from there. So basically, try to appear in the media that the big media read. Blogs compete to get stories first, newspaper compete to confirm, and tv competes to airtime it.
When the NYT writes about you, they are doing you a favor because there is a finite amount of space in the paper. When business insider writes about you, you are doing them a favor because more content = more money because it is a blog.
These events typically get reporters’ attention:
- raising money
- launching a new product
- breaking a usage barrier
- a PR stunt
- big partnership
- special industry report
When you have small announcements, bound them into one big.
When you pitch to media outlet, it is your job to present it in a way that makes a story compelling. You need to sell that story.
After pitching small blogs, go to bigger ones. Follow influencers in your industry and pitch blogs they often link to.
Once you get your story: submit it to community sites (digg, reddit, hacker news); share it on social networks; send it to influencers; ping blogs in the space.
Third Traction Channel: Unconventional PR
Do something crazy or funny that gets media attention. WePay, PayPal competitor, asked for a bloc of ice to be put at a PayPal conference because PayPal froze some customers’ accounts. Half.com renamed the town of Halfway into…half.com.
Be nice: send cookies, cakes, handwritten notes, or whatever to your customer when they buy from you, or to people that mention your product on twitter, give free swags (t-shirts, etc), send personal emails thanking users, etc.
Viral videos: a blender company started a Youtube show called “will it blend”, and they started blending pretty much anything.
Organize contests and giveaways.
Be exceptionally caring to your customers. When you do something nice, it is picked up by the media.
If you go to a city for an event or anything else, email your customers and invite them out for dinner.
Fourth Traction Channel: SEM (Search Engine Marketing)
It enables you to test ideas before building the product. You do simply by running ads on search engine. It also enables you to measure the CPA (cost per acquisition) of a customer.
Ex: if you pay 1 euro each time someone clicks on your ad and it takes 5 customers to have one buying your product, then your CPA is 5 euro. If you sell a product for 100 euros, it is clearly worth it. If you sell your product for 15 euros, not so much.
The basic SEM process is to find high-potential keywords, group them into ad groups, and then test different ad copy and landing pages within each ad group.
Average CTR (click through rate, the number of people that click out of the number of people that saw the ad) is 2%. If it is less than 1.5% there is something you are not doing well.
Fifth Traction Channel: Social And Display Ads
This channel is running ads on different websites. Diplay ads enable you to reach a broader audience than SEM. While the big platforms have their own ad center, smaller websites use ad networks that run ads on millions of other websites.
Here they are:
- Google Ad Sense
- Facebook network ads
- Apple advertising
- Yahoo network
Niche ad networks: focus on smaller sites that fit certain audience demographics.
The last approach to display ads is to talk to site owners and ask them to display an ad on their websites. If you know precisely where your audience hangs out, then you can do that.
First, you need to understand which ads work in your industry, so look at what your competitors are running. MixRank and Adbeat do just that. Alexa and Quantcast can help you determine who visits the sites that feature your competitors’ ads. Then you can determine whether a site’s audience is the right fit for you.
SEM ads work for stuff people want to buy right now. Diplay ads work more to introduce new products and generate interest, or reach new buyers. The purpose is awareness, not conversion (usually happens down the line).
Use social ads to create content that teach or entertain, since that’s the reason why people go to these websites. You can also use these ads to promote organic content. If the content is good, it gets retweeted. As such, social ads are like helping a fire to start, and it should be treated this way.
Sixth Traction Channel: Offline Ads
- Print magazine and newspaper ads
- Direct mail: you can actually buy lists of people (google “direct mail lists”).
- TV ads: mostly used for branding
- Posters (in taxi, train stations, airports)
- Radio ads
A few principles to guide your use of offline ads
Demographics: most important factor to consider.
Questions to ask yourself:
- What are the basic demographics of the audience?
- What are the economic demographics?
- How well does this advertising demographic match up with my target customer demographic?
You should get these info from the company you are buying the ad from.
In general, the cost of an ad depends on the reach it has.
To run tests, look for remnant advertising (ad space which is free). Near publication, media accept any price to fill it up. Some agencies specialize in selling remnant advertising.
Tracking: tracking is hard. It usually entails to special URLs or promo code to measure the effect. You can also ask people a question “how did you hear about us” when they sign up.
Seventh Channel: SEO
SEO is at its core, having a content strategy.
In SEO, there are two high-level approaches: fat-head and long-tail.
Fat-head: rank for terms that directly describe your company, usually competitive terms, and broad.
If you find nice keywords, test them with Google Ads. If the ads convert, then SEO is interesting to pursue. No point wasting time on terms that don’t yield traction.
Once you found your keyword with Google Keyword Planner, go to Google Trends to see what is ranking over there and choose the best keywords.
Then you should look at which keywords will be harder to compete with using Open Site Explorer, which tells you the number of links competitors have for a given term.
Long-tail: rank for very specific terms in a niche with less competition, but smaller search volume too. Majority of keyword search today. About 70% of all searches.
You need to create content that attracts users to your site with long-tail keywords. This is all it boils down to. Look at your competitors, their landing page (if you can find the sitemap, or on Google), and on Alexa, how much volume they get from search alone.
SEO works well for geographically targeted services and products (pizzeria in Brussels, Belgium).
Another way is to get people to your website by sharing the data you naturally create, collect, or use.
The point to create amazing content is that people actually link to it.
Getting links: PR, creating social media pages with your products, content marketing that people want to share, widgets.
Eighth Traction Channel: Content Marketing
Unbounce began blogging a year before they had a product. Thanks to the amazing content they put out there, they got recognized in the industry, built an email list, and launched with 5000 people on it. The team relied on social media to spread their post, ask influencers to comment them, etc.
OkCupid wrote interesting stories and stats on their blogs that got pitched to mainstream media which drove users to their blog.
To write valuable content is to help your customers solve problems they have, with actionable results.
Infographics are shared more often than the blog post itself.
If what you are writing isn’t useful to people, your content won’t spread.
To build yourself a reputation, engage in online communities from your industry.
Do guest posting.
Ninth Traction Channel: Email Marketing
Works best when it is personalized.
Don’t buy email lists. Build it yourself by having a newsletter, a free product (course, book), or premium content accessible once people register with their email addresses.
If you don’t have an email list, then advertise on other people’s newsletters.
Use email to help people have the perfect experience when they use your product. When you see they’re not using a feature, tell them. Remind them to use your product when they drop off (don’t finish to sign up, leave purchases in their carts, etc).
Tenth Traction Channel: Engineering As Marketing
Hubspot reached growth by creating a marketing grader, a free tool that enables you to get a report of your website marketing on your email address. They can collect email addresses at the same time.
You need to look at engineering as marketing as an asset that brings in leads forever. Building a small online tool is done once, forever. Paying for ads is repetitive.
This traction channel is good as so few companies use it.
Eleventh Traction Channel: Targeting Blogs
One of the most effective ways to get a first wave of customers.
Here’s how Mint acquired customers so fast:
- Create a quant-based marketing spreadsheet
- List marketing channels
- Ran numbers in terms of traffic, click-through rates, and conversion
- Calculate the number of expected users for each channel
- Run tests
They found 600 blogs about money-management and asked them to write about Mint, which further exploded their SEO.
VIP access: get the people on your waiting list to get their friends to sign up for the waiting list in exchange for VIP access.
Another way to use this channel is to sponsor blogs (put an ad on the blog). Then track the blogs to see which ones (and which ad) are most effective and what kind of people sign up.
You can also share links under videos, etc in communities where people you target usually hang out.
Twelve Traction Channel: Business Development
Sales = exchanging value for money.
Business development = exchanging value for partnerships. Eg: when Google partners with Netscape to be their main search engine.
There are different types of business development:
- Standard partnerships: Apple + Nike that make a connected shoe.
- Joint Ventures: a separate company made by two other companies.
- Licensing: Virgin Mobile in the UK, which rents the Virgin brand
- Distribution deals: when one company is paid to distribute the product of someone else that can then reach customers.
- Supply partnership: self-explanatory
For this channel to work, you must have a clear understanding of your company objectives.
Good BD strategy aligns with your company and product strategy and are focused on critical product and distribution milestone.
Think about why would a company partner with you to select the right partner. Don’t be too much focused on yourself.
Make a list of potential business partners in a spreadsheet with columns like company, partner type, ease of use, size, relevance, then a subjective priority score.
Thirteen Traction Channel: Sales
Most of the time, your first customer is someone you know.
A first sales pitch is a list with ten bullet points of what the product/company can do for the client + a deep understanding of the client’s problem. Eg: The founder of an insurance software company went to talk to insurances to find out about their problems, then gave a talk about how their software could help them out (or not).
How to structure a sales conversation:
- Situation questions: learn about your prospect’s situation. Can be “how many employees do you have”, for example.
- Don’t ask too many of these questions as the more you ask, the less likely you will close the sale. Ask just enough.
- Problem questions: those that clarify the problems of your prospect.
- Implication questions: making the prospect understand that the problem he has is causing…well, problems. Use to make them feel the problem is bigger than what they had thought.
- Need-payoff questions: these questions focus attention on your solution and get buyers to think about the benefits of addressing the problem. How do you feel about this solution?
This framework is called SPIN:
Cold-calling: Try to get on the phone people that have power, but not too much.
The structure of a cold-call is as follow:
- Process: how does the company buy solutions like our?
- Need: how badly do they need our product?
- Authority: who in the company has the authority to make this purchase?
- Money: Can they afford it?
- Estimated timing: how fast can they buy?
After the call, send them an email with a recap of the call.
Your first customer
It is someone DESPERATE for the solution you provide.
- Generating leads: done through other channels (ads, email, blogs, etc). Giving something for free in exchange for an email address is very powerful.
- Qualify lead: determine how ready your prospect is by putting leads in buckets. A: within 3 months. B: ready within 12 months. C: ready after 12 months.
Marketing should help sales to arm them with reasons to sell, and aim at people likely to buy.
Lay out exactly what you are going to do, and in which timeframe. Ask your customer to commit (with a yes or no) whether they will buy.
Design your sales funnel from customers’ perspective. Find out which FUD keep them up at night regarding whether they want to buy from you or not, and design your sales funnel around that. Answer all of their questions, then give them a trigger that will make them buy what you want to sell them.
Removing obstacles to sales
- Removing needs to install programs
- Offering free trials
- Channel partners
- Demo video
- Reference customer
- Email campaigns
- Webinars, or personal demo
- Easy installation and ease of use
- Low introductory prices
- Eliminating committee decision-making
Fourteenth Traction Channel: Affiliate Programs
Your success in using affiliate programs depends on how much you are willing to pay to acquire a customer. The authors recommend to use an affiliate network to establish the program.
The alternative is to create your own program, selecting people yourself, or approaching blogs. The good thing is that you do not necessarily have to pay people with money – you can pay them with access to your product if it’s a freemium model, for example.
Fifteenth Traction Channel: Existing Platforms
Existing platforms are not only something you can leverage as a marketing channel. It can also help you develop products. Many early sites dont have all the features they need. By building them, you leverage both the platform and the product.
Eg: Youtube was an easy way to share videos on MySpace. IMGUR was an easy way to share pics on reddit.
You can do the same thing by building an extension for chrome, an app for wordpress, for shopify, etc.
Tip: evernote builds a version of Evernote to be present in every store and platform that exists.
Sixteenth Traction Channel: Trade Shows
Trade shows are good for sales, but they are also good when you are in product development. If you come with some prototypes at the show, you can speak to people about your product and get feedback.
Make sure it makes sense financially to go. Also make sure to visit the show a year before going, so you know what to expect.
Check out if your target market goes to the trade show. It is useless otherwise.
How to prepare for the trade show:
- Make a list of key attendees you will want to meet at the trade show
- Schedule meeting with them before attending the event
- People you could meet are: editors of magazines, bloggers, customers, vendors, competitors, potential partners
To get people to your booth, send them discount codes on your product to redeem at the booth, give free goodies (t-shirts, pens), etc.
Whatever you give should have a compelling CTA (a coupon on the business card, for example).
One step beyond taking people to dinner is to throw a party.
Seventeenth Traction Channel: Offline Events
This comes down to sponsoring or going to meetups or conferences, or organizing some yourself. Particularly good for startups that have long sales cycles, like enterprise software.
The strategies aren’t so different than going to trade shows. You need to talk to people, find the people that need the solution you are developing, and ask them questions.
Conclusion: when you are trying to get your first 1000 customers, you need to do stuff that don’t scale.
Eighteenth Traction Channel: Speaking Engagement
It’s easy to start. Start by giving talks to small groups of potential customers or partners. Once you are comfortable, you can look for bigger events and email the organizers to see if you can speak there.
You have to get the attention of event organizers to land a speaking engagement. These people need public speaker, so if you have a good idea for an event in your area of expertise, pitch it. Another way is to contact event organizers and ask them what idea they want to be talked about. Then make your pitch.
To become a speaker, you have to speak once. If you’re good, people will ask you to speak again. It will be like a snowball.
Ninetheenth Traction Channel: Community Building
How to build a community:
- Build an initial audience
People want to feel like they are part of something bigger than themselves, which is why you need a mission for your community.
- Foster relationships in-between the members of your community
- Talk to your audience
- Be transparent
- Ensure quality (create guidelines about the content discussed)
Communities enable many advantages, among which hiring, feedback, etc.
For more content, head to auresnotes.com.