Very large CSV import in Rails on Heroku

Update: Our current solution for this is working pretty well. Skip to it here.

Storemapper, like many many B2B apps, allows our users to create, edit and export records in bulk using a CSV upload file. I have spent a ton of time trying to find the best method to allow the upload and asynchronous processing of very large (10,000+ rows) CSV files. This post is a documentation of our work to date on this, a plea for collective help from the internet, and hopefully will be the public home of the best solution we can come up with.

Continue reading →

Thought Experiment: Revealed Passions

Freakonomics co-author Steven Levitt often references his mentor, E.O. Wilson, the best ant scientist in the world.

He argues on several occasions that being passionate about your work is an immense competitive advantage. His mentor is the best ant scientist in the world because he absolutely freaking loves ants. He loves studying them, and he thus studies them more intently and more often that someone who’s just in the ant science game for the paycheck and the glamor.

Sounds pretty plausible to me.

Let’s leave aside for a minute the debate about whether “doing what you love” or earning a good paycheck so you can do what you love is more likely to make you happy. Being passionate about your job almost certainly will make you better at it. And finding work that you are uniquely more passionate about that anybody else is probably a good way to find work that you have an unbeatable competitive advantage in.

Here’s another little economics gem: Revealed vs stated preferences. Experiments have shown again and again that humans terrible forecasters of what makes them happy. When asked to state their preferences — purchases now or savings later, a new TV or a weekend in Paris, and so on — the choices they tell rarely match up to their actual decisions. Economics use the concept of revealed preferences, those trade-offs people express through their actual decisions, to determine what truly makes people happy… or if you prefer econonomist-speak: what maximizes their utility.

I think it’s not too far to stretch to say that humans probably are not great at forecasting what they are actually passionate about. So maybe we should use the same toolkit of revealed preferences to think about revealed passions.

So I’ve been doing this thought experiment over the last few years. What do my actions actually reveal about what I am passionate about? What do I consistently spend my idle time thinking about, tinkering with and improving without any obvious economic benefit?

The thought experiment is to think about this first as objectively as possible, then if you strike something interesting, contemplate how you could work or get paid from it.

Here’s an example for me. I freaking love travel bags. I’ve lived a large portion of my adult life out of a single carry on suitcase or backpack. I currently live out of one. I don’t have a home, just some flexible storage space with Makespace and I still personally own eight or nine different variations of backpacks and luggage. I’m constantly upgrading, constantly looking at reviews of new gear and constantly checking out other people’s bags in airports — taking notes, evaluating features and generally spending way too much time thinking about travel bags.

Maybe I should start a travel bag design company. Why not?

I don’t think this is some secret formula and I’m not dropping everything to start a travel bag company. But I never even considered it until I did this revealed passions thought experiment. So it was a useful 30 seconds of brain work.

And now I’m thinking about it! Maybe it is a good idea. That thought prompted me to start researching, learning about manufacturing and design. If anything it spurred me to learn a bit more about something I’ve never dug into before.

What are your revealed passions? Give it a shot.

Putting a Micro SaaS Business on Autopilot

Regular readers of this blog will know that I announced back in February that I’m writing a short ebook on building Micro-SaaS businesses. So, what the heck is taking me so long to finish it? Well, I wrote it with a narrative structure, starting at the beginning, and as the first draft caught up to the present day, I realized it might be a bit premature to publish the book. I had just finished a period of heavy time investment into the business to get it to the point where it generated a comfortable full-time income, but I hadn’t put much thought or work into the end game of micro-SaaS.

What was the next, more or less final, stage of Storemapper and micro-SaaS in general. I thought I should work on this in practice before putting a book out there.

Three months ago I posted about Storemapper crossing $100k in annual recurring revenue. Since then revenue has still grown fairly steadily and is approaching $11k/month.

At this point in the business there are three main options: keep focusing on growing the revenue, sell the business, or put it on autopilot. As I mentioned in my last post, I don’t want to be the King of Store Locators and I have several other projects on deck that I want to devote my attention to. So focusing a lot of time on growing the business isn’t an option I want to pursue right now. Selling Storemapper would certainly free up time, but for now I’ve decided instead to try to put the business on autopilot. The goal of autopilot is to minimize the time that I spend working on Storemapper, while keeping my take home profits at least the same.

So why not just sell it?

That’s a good question. I’m not opposed to selling Storemapper to a good owner, but at the moment it’s not something I am actively pursuing for a couple of reasons:

1) Price The primary reason I haven’t already sold Storemapper is price. I’ve had a lot inbound inquiries and spoken to a few brokers and my sense is that SaaS businesses of this size can pretty readily be sold for between 3-4x profits or something in the $300-450k range. That’s a nice chunk of change, but when I really look at the numbers I don’t think it’s a good trade.

The financials of the business are public and they are really strong. It grows without any marketing budget and customers almost never churn. If I’m able to automate the operations of the business, the revenues alone will massively out-perform even the top end of that range. Particularly in the current financial environment where other investments are yielding such low returns.

What do you think? Am I making the right call there?

2) Tax: Storemapper is currently a single-member LLC. As I understand it, if I sold it right now, the proceeds from the sale would all be ordinary income which means I’d lose a lot to Uncle Sam right off the bat. If any tax gurus have some suggestions on a better structure I’m all ears please.

3) Synergies: Operating a growing profitable business is kinda fun. It’s nice to have material to blog about. I’ve met some really fun people through my customers and there are also lots of ways to use the customer base as an audience. I’ve been experimenting a little with cross-selling partnerships too. I’d give all that up in selling the business just for the cashflows.

Every time I considered selling Storemapper over the years I have come to the same conclusion: better to hold it. I think this will generally be the case, if the business is still growing at all, for most micro-SaaS businesses. Unless you need the money, or just have no time at all to work on it, most micro-SaaS businesses will be better off held on to unless market multiples rise further. I know myself I always kept thinking that the end game for Storemapper would be to sell it. But at the moment I’m fairly convinced that putting it on autopilot is the way to go. I also think that should be the dominate end game model for entrepreneurs just starting out. The idea of a big pay day is appealing, but autopiloting the business is both more likely and probably the better financial outcome.

So what does autopilot mean?

Putting Storemapper on autopilot does not mean abandoning it or letting it fall into disarray. It means maintaining and growing the product, but with the key metric being reducing my total time spent working on it, rather than MRR growth. The classic “managing myself out of a job” approach.

The actual work of Storemapper falls into three categories: software development, customer support, business operations. For the first two I have hired two folks part-time to handle the work and for the second I’m leaning on apps and services.

I was fortunate to recently meet Nico Appel, who runs Tight Operations (TightOps) a consultancy that focuses intensely on business process automation. We’ve been working together on strategies for how to put Storemapper on autopilot and you’ll see some stuff documenting the process coming from us soon.

Software development is all very straightforward stuff. Catching and fixing bugs, updating versions on different gems and plugins, releasing a few new features and tuning performance. To that end I’ve hired a developer part-time on oDesk (or I guess now it’s called Upwork) and things are going really well. For the hiring process, this post from WPCurve was helpful. I didn’t follow it exactly but I did hire several candidates for paid test jobs before selecting one. It was a time-consuming process but I’m really happy with how it worked out. Our workflow right now is pretty hands-on with me delegating tasks in Asana and reviewing code in pull requests from Github, but as the developer gets to know the code better I’ll start to get more of that process out of my hands.

A great suggestion from Nico at TightOps was to build in redundant layers of alerts for bugs and other issues. Making sure that the support “team” can directly register bug reports for the dev “team” (Each “team” is only one person for now). I’m using Airbrake for error notifications and New Relic for performance monitoring. We’ll be taking a look at setting those up with a series of escalations so that I can take more and more time away from checking them every day, but know that if something goes seriously wrong everybody on the team gets an alert, an SMS goes to me, and so on.

Customer support is the other workload that really needs a human. I’ve hired someone part-time to handle customer support inquiries. It’s been an interesting process since I built every aspect of the business and until now I answered every support question. I saw a lot of the same questions and became super efficient at solving problems quickly, typically on less than 5 hours per week. Not knowing any other way, I took an apprentice model of onboarding my first support hire. Essentially just throwing him in to the stream of tickets and seeing how he sorted them out, asking questions of me along the way. This has worked well so far, but it’s quite time-consuming, which is of course the opposite of what I’m trying to achieve with the autopilot process. However, after the first few months it will start to pay time dividends where fewer and fewer tickets require any involvement from me.

One of the things the TightOps team and I are working on is relentlessly moving all of the information in my head into searchable and updatable internal documentation.

With business operations my strategy has just been to throw small amounts of money at apps and services that provide automation. I signed up for the LegalZoom services that handle all the corporate filings. I recently started using Bench for bookkeeping. At first it seems a bit pricey at $135/month, but I hate bookkeeping (a bad trait for an entrepreneur I know) and Bench is so much better than using bookkeeping software because you have a real person that learns how your business works and does all the expense categorization and statement reconciliation so that I basically just need to login and check on it once per quarter. So far I love it. We’re starting up with ZenPayroll for payroll processing. Setting up payroll is still a big pain, but ZenPayroll is definitely the most painless option for this. And that’s pretty much all I can think of right now.

The end game is bigger better things to come

I’ve got a lot of exciting projects in the queue and I can’t wait to get started. One of the reasons I’m so excited about micro-SaaS businesses, and even feel compelled to write a little book about them, is that they are a fantastic financial platform for more ambitious projects. One of the primary reasons that my previous startup failed was that I personally ran out of money. When we exhausted our angel funding there was literally no way I could pay rent and keep working full-time on the startup. Things might have gone differently if I had a full passive income like Storemapper up and running before jumping into the startup game.

So that’s the plan, automate the business and use the time and money to work on some riskier projects. More on those to come 😉

By the way, if you have a profitable business, software or otherwise, and want to recoup some of your time with automation, you can reach out to Nico and TightOps team at nico@tightoperations.com, or shoot me an email and I’ll connect you.

Digging in to the Open Startups List

Open Startups List

I’m super honored and excited to see my business alongside six other incredible startups that have embraced financial transparency, and completely opened up their financial metrics on the Baremetrics Open Startups List. I’ve written before here about why I think financial transparency is a huge asset, particularly when paired with a transparent blog or any kind of business advice. This list of open dashboards, which I’m sure will be growing rapidly soon, is a huge assets for current and aspiring entrepreneurs. It’s a ton of data and I thought it would be fun to dive deep and see what we can learn.

Introduction to the Open Startups

First I’ll do a terrible and cursory introduction to this illustrious list for background.

Buffer is an app for finding shareable content and queuing that content up to be shared across multiple social networks at optimal times. Founded by Joel and Leo in late 2010, Buffer has been a leader in radical transparency, publishing everything from business metrics, salaries to term sheets from their VC fundraising. Their transparency dashboard is a gold mine.

Ghost is a simple blogging platform. Launched as a Kickstarter campaign in 2013 by John O’Nolan. Ghost is an elegant Markdown-based blogging platform that I use for my blog and love. Ghost itself is completely open source software, but they sell Ghost Pro, a hosted and managed version.

Baremetrics is the ringleader behind this whole list and also a provider of one-click financial metrics for Stripe. After building a tool to visualize Stripe data for his own app, founder Josh Pigford, decided the quickest way to demo the app was to make Baremetrics’ (is that the right apostrophe move there?) metrics totally open.

Hubstaff is a time-tracking app for managing remote teams. Founded by Dave and Jared in 2012. it’s a classic “scratch your own itch” story. The founders were trying to manage a remote team of freelancers and decided to build a better tool to do it.

Convertkit is an email marketing app for authors. It was founded by Nathan Barry who is also known for writing ebooks, and then writing an even better ebook about how to write ebooks called Authority. If you’re writing an ebook (like me) you should read Nathan’s book first then use Convertkit to email about it.

Promoter is a super easy way to measure Net Promoter Score in your app and get feedback from your customers. I don’t know much about these guys. Maybe we should change that guys? Say hi sometime 🙂

Storemapper is my app and it is waay less cool than all these other ones. It started as a side project and grew into a proper micro-SaaS business. I blog about it a lot here and am writing an ebook about the process if you’re in to that kind of thing.

Phew! That took forever to write. So with intros out of the way, let’s dig in to the data.

A few quick caveats

I haven’t consulted with any of the startups on this stuff. Just me, a private citizen, digging around in the public dashboards and making some guesses. I am very likely wrong about some stuff. If any of the startups here take issue with my conclusions or want to add clarification please do so.

Also there are a crap load of links in this post. I’d recommend using open in new tab for them all so you don’t get lost.

Average revenue per user (ARPU) and Pricing strategy

Below is the most recent Average Revenue Per User (ARPU) for each startup. This figure is available right on the main dashboard for each so I won’t bother linking to it.

ARPU is an interesting metric because it’s not inherently good or bad on it’s own. Raising your prices, or the mix of your pricing options can increase ARPU but you don’t know if it’s a good change on balance until you look at how it affects signup conversions and trials. However SaaS gurus argue that SaaS startups consistently underprice their products and should look to raise ARPU in most instances.

  1. Promotor $97 This is really interesting to me. Promoter is by far the most aggressive on pricing with plans ranging from $49 – $499/mo. The only feature that changes by plan is you go from a limit of 500 surveys/month to 30,000/month. Businesses must really find the NPS data valuable. If I had built this product I definitely would have made the mistake of pricing it much lower than this. Lesson learned for SaaS startups to always check: http://shouldichargemore.com/
  2. Baremetrics $80 This one I completely get. Baremetrics gives you a ton of stuff right away that’s directly related to your bottom line and probably makes you more money. You’re probably paying an accountant or someone to do this crap already. On top of that it would be a huge pain in the ass to build it yourself. Prices range from $29 – $249/mo primarily scaling with the number of paying customers you have; no free option; requires credit card upfront.
  3. Convertkit $54 is also priced fairly aggressively. Particularly when customers might be switching from a free Mailchimp account. $54 is just a hair above their lowest tier at $49/mo. I would imagine that writing Authority generates a lot of first-time ebook authors, like me, who sign up for the lower tier, like me. As the app matures, one would expect a lot of their customers to progress up the plans which scale with the number of subscribers up to $359/mo @ 45,000 subscribers.
  4. Hubstaff $29 is still a B2B product like the previous three but is priced noticeably lower with a free option for 1 person and the highest tier at $99/mo (though it does go up further for very large teams). The ARPU is just a bit higher than the $25/mo plan for 5 person teams. The bulk of their revenue comes from plans for teams of 5-15 (more on contribution by plans later). I wonder if they might take a cue from the other startups here and consider raising prices at those plan levels.
  5. Storemapper $16 as I’ve chronicled on my blog, Storemapper was very much underpriced and raising ARPU has been a big priority for me. When it launched I only charged $5/mo. Over time I increased that price several times, then added Premium plans at a much higher price point and focused heavily on upgrades. For a B2B product we’re still very much on the low end. For example: We have 60% more customers than Baremetrics but on 1/3 the revenue 🙁 I think it represents a big opportunity for Storemapper though but something we’re not doing well yet.
  6. Buffer $13 Buffer and Ghost are the only two B2C startups here so it makes sense that their ARPU would be lower. The $13 number surprised me since I know Buffer’s 1-person paid plan is $10/mo and you would expect other plans to be less than that per person. Here you can see the challenge of one-click metrics. It turns out Buffer does have group plans that range from $50-250/mo for 25-150 users. But because it’s only one charged account the ARPU figure is skewed high.
  7. Ghost $9 It’s interesting that’s Ghost’s current ARPU is $9, lower than their current lowest paid plan of $10/mo. They currently list plans ranging from $10-250/month allowing you to have more pageviews (Ghost’s product is a hosted instance of its open source platform) and more blogs. It’s pretty clear that the bulk up Ghost’s current paying customers (including me) are solo bloggers on the lowest tier. This makes sense as the platform is still very minimalist. As it matures it will likely appeal more to teams, companies and agencies who will opt for the higher plans. The Ghost team wrote recently about the realization that they initially weren’t charging enough and the decision to raise prices.

MRR contribution by plans, Number of plans and price discrimination

One of my favorite features on Baremetrics is the fact that for most metrics you can segment by plan to get a sense of how each plan is contributing to your key metrics. This can help you change pricing, add or retire plans that just aren’t pulling their weight.

Here are some interesting nuggets from looking into that feature. Btw, I’m not trying to pick on any of these businesses, just showing what valuable things you can learn from transparent metrics.

Almost all Buffer’s business customers are on their lowest Small Business plan of $50/mo for 5 team members. Though they have $100 and $250/mo options for bigger teams, the $50 plan (and it’s annual billing equivalent) dwarf the other two in contribution to monthly recurring revenue (MRR). Check out the table at the bottom of this page. This tells them maybe they could do a better job appealing to larger companies, or maybe that they should just increase their focus on the Awesome Plan (the 1-person $10/mo plan) and Small Business and leave the agencies to others.

Hubstaff’s sweet spot is 5 – 15 person teams. As I mentioned earlier it’s clear that the plans for teams of 5-15 people make up the lion’s share of MRR: Table at bottom. They are clearly doing well with this audience and maybe that’s a good segment to focus on increasing ARPU if they think that’s a good idea.

Promoter.io could improve the value proposition of their annual plans. Even though they offer prepaid annual pricing for all their monthly options, annual/yearly plans contribute almost nothing to their MRR: Same table. Prepaid annual plans are extremely valuable to SaaS businesses. Even though they actually decrease revenue from a customer, they massively improve cashflow which you can re-invest in the business. Prepaid annual plans have been hugely helpful for my business (and even helped me pay the rent in a pinch one time). Maybe customers are hesitant to commit to a full year, thinking they only want to test NPS intermittently. Maybe a deeper discount for annual is in order. Or maybe they just need to deploy this amazing end of year pitch email from Patrick McKenzie.. Or maybe things are just the way they like them and I don’t know what I’m talking about.

Storemapper has done way too many attempts at price segmentation. I spent a bunch of time setting up with elaborate way to A/B test prices, cookie’ing users to a plan cohort, attempting to track lifecycle differences in churn and so on. We have three plans each with monthly or annual payment options so just running one A/B test requires setting up 12 different plans in Stripe. The reality is we just didn’t have enough traffic for a test to really be conclusive. Lo and behold, the vast bulk of our revenue comes from the two plans my gut told me would work $19 and $39/mo: table. This is definitely a lesson in premature optimization. Related: I love David Kadavy on why you should run an A/A test.

Churn, the SaaS entrepreneur’s worst enemy

Next let’s look at user churn. Even with the Baremetrics data right in front of you it can still be tricky to figure out how to define and chart churn.

Below are each startup ranked by an approximate range of monthly user churn over the last 12 months.

  1. Storemapper 1 – 1.5% this is the category where we’re totally kicking butt. Storemapper has very low churn. The main reason is because users more or less put our app on autopilot once they set it up and get their store locator looking right whereas all these others guys are active apps that you use as part of your workflow regularly and therefore might look into alternatives more often. Chart link
  2. Buffer 3 – 4.5% Buffer is the most mature app by far with 10x the users/revenue of anybody else on the list. This is a great example of a target churn for a fairly mature SaaS app done well. Chart
  3. Hubstaff 4 – 8% had really nice 3-4% churn for a while there a year ago but it’s been on the rise. I don’t use the app directly so haven’t been following it close enough but I wonder if they raised prices around then end of 2014. So frequently in SaaS raising prices increases churn but is still a huge net win, with the remaining customers more than making up for departures. Chart
  4. Promoter 5.5 – 8.7% most months they are in the 5-6% range, which is great for a young SaaS business, with some occasional big spikes. The user base is comparatively quite small (104 customers) so those spikes might not mean much. Keeping churn as low as they have is all the more impressive as they are really aggressive on pricing with the highest ARPU of all the startups on the list. Chart
  5. Baremetrics 5.5 – 10.0% certainly has the most erratic churn rate with as low as 3% and as high as 10%. I suspect this is partly due to the fact that it’s a pretty low commitment to actually sign up. You put in your credit card, connect your Stripe account and boom you see your metrics. Folks who are just window-shopping might still want to give this a shot even if they aren’t truly a good fit. I suspect churn smoothes out if you exclude people who cancel within the first month. Chart
  6. Convertkit 15 – 20% has a much higher churn rate than the others. This is probably due to how crowded the email marketing space is. It’s also evidence of still being a bit pre-product-market-fit (that’s weird to type). Nathan wrote recently about how initially the product was targeting too broad an audience. After refocusing the product exclusively on authors I suspect churn will fall rapidly as it has for the last six months. Chart
  7. Ghost (not enough data)

Lifetime Value: the holy grail of SaaS

Customer lifetime value is the holy grail of SaaS. The reason SaaS is such a great business model is that new customers continue paying and over time generate a big predictable stream of cashflows. It’s the reason VC-backed SaaS startups can lose money for years while growing and still be worth billions.

If each new customer was forced to keep paying forever, well SaaS would be like printing money, but customer’s leave at some point. On the churn pages, Baremetrics gives you a handy “time to churn” figure which is the length of time the average new customer is expected to stick around before canceling — based on your monthly churn rate. Combining the average amount of money you expect from each new customer and the length of time you expect them to stick around gives you your lifetime value (LTV).

Now when you say, “This is my business’s lifetime value,” you need to add like 27 footnotes to that statement. Lifetime value is a tricky to calculate and assumption-riddled figure. However at least with the Open Startups List we do have LTV per customer calculated the exact same way for 7 different businesses. Fascinating.

So what do we find?

  1. Baremetrics $2,661 What?! They rigged the game for sure. 🙂
  2. Promoter $1,391 Makes sense. Aggressive high pricing while keeping churn low makes for a nice solid LTV.
  3. Storemapper $1,279 What, how’d we get up here? This just shows how important churn is in the end. Even though we don’t extract a lot of cash from each user, they just never leave, with a time to churn of nearly 7 years! Hence we get to hang out up here with people actually good at startups, unlike me.
  4. Hubstaff $388 Both Hubstaff and Convertkit have a pretty similar LTV. What’s interesting about that is they are both going after pretty established B2B markets: time-tracking and email marketing. It may be that this is good expectation if you’re going to build an app that fits many different use cases. The upside is their target market is probably much larger including smaller businesses, individual freelancers and a wider array of target customer types.
  5. Convertkit $359 see above
  6. Buffer $278 Hahaha, sucks for Buffer all the way down here. Oh wait, they’re doing $5.75 million per year in revenue. Classic difference between B2B and B2B/B2C apps. Lower LTV but a massively bigger potential market. Now that I think about it, this actually seems very high and their churn rate quite low for a product where most of their customers are individuals rather than businesses. *slow clap*
  7. Ghost $117 Again this makes total sense. Ghost is going after the totally massive blogging market. Nearly all their customers are currently on the lowest tier plan, but the product and company are also quite young. Despite that, they are totally crushing it with a $420k annual run rate. Hopefully so many people will read this blog post I’ll be forced to upgrade and nudge this number up.

You should join us!

Hopefully by now you’re convinced what an invaluable asset Open Startups can be for the business world. It’s a little scary, yes, but it’s also really fun. At first you going to be so scared that everybody out there is going to steal your idea or crush by knowing a little bit about your business. But after you do it you’re going to realize that world is totally not that zero-sum and that being helpful to other businesses and entrepreneurs pays itself back hugely.

If you’re already a Baremetrics customer, join us, Rip off that kimono!

If you’re not using Baremetrics… Wait you’re not? How do you know anything about your business? I can’t even.

The Best Networking is Not Networking

My first job was at a research consultancy. I was learning and writing about the booming wind energy industry. A big part of my job, and that of and many of my colleagues, was to attend an ever increasing number of industry conferences. The goal was to be part salesperson, convincing potential customers that the analysts at our firm were smart people with interesting things to say, and to play journalist, getting nuggets of data and an informal consensus on topics for which we lacked data. The fact that I reverse-engineered the financial model for a complicated financing structure1 almost entirely from buying beers for CFOs at conferences was probably the only reason anyone in the industry thought I had anything valuable to say, for the first few years at least.

These conferences were like any other — people milling around in uncomfortable suits at random convention centers, and sometimes fancy resorts. The startup I worked for was still young so my business card carried little cachet, that meant I would have to do some networking… ugh.

I’m a slim 5’6” so nearly everybody, already packed into tight circles, was physically either taller or wider than me. Wind energy was already an old energy industry so nearly everybody was two decades or more my senior. The battlefield was intimidating to say the least.

As a natural introvert I hate this stuff. I don’t mind talking to people, but walking into a crowded room and trying to spark up conversations with strangers. Makes my palms sweat as much as anybody else. Except for extroverts who actually like that stuff, those people are lunatics.

At first I was very analytical about figuring out how to network in the least painful way possible. I observed my colleagues who were good at it, read Dale Carnegie books and that awful Never Eat Alone book, read some blogs and experimented continuously. I learned some stuff that does in fact work.

I learned if you are young you need to dress very sharply to be taken seriously. If you manage a billion dollar private equity fund, everybody takes you seriously even if you’re in an oversized mustard yellow three-piece suit, a bolo tie and cowboy hat with an eagle feather in it. If you’re in your early twenties with nothing to trade but your wit and charm, you need a dark well-tailored suit, pocket square, some strikingly neon dress socks in polished kicks.

I learned about a “whatzit” or a conversation piece. Sometimes the tiniest ice-breaker can kick off a great conversation so it’s good to dangle something, anything really, that someone can make an off-hand comment about to break the silence. In my case it was a metal and tortoise-shell business card holder that I deftly whipped out to distribute cards while everybody fumbled in their wallets for crinkled pieces of paper. “Oh nice card holder,” they’d say. I’d respond, “Thanks, what are you investing in these days?” I’m aware of the American Pyscho business card analog here but you gotta admit, Bateman was a smooth operator.

I learned to get to the conference really early because everybody is off their guard and bleary-eyed at the coffee line. I learned later to spend hours before the conference emailing attendees because hard to reach people were 10x more likely to agree to a 15 minute side meeting at a conference than a 30 minute phone call on their normal calendar.

These are just a few snippets, there’s mountains of advice out there on how to network better. And a lot of this stuff does technically work, it will improve what you get out of a networking event and it totally worked in some ways for me.

But then I discovered a networking tactic that worked 100x better. All the other marginal improvements I’d made were worthless by comparison.

The secret to power networking is: Make something awesome and tell people about it.

The way to crush a conference is to give a compelling, data-rich, slightly funny and slightly controversial talk. It’s as simple as that.

I learned this when the CEO of our company was supposed to give a presentation at a reasonably important conference and couldn’t make it at the last second. So my boss thought, “Tyler’s been working for us for like six months now, why not have him do it instead?” (Yea, everybody there was awesome like that).

So I worked my ass off, put together a really interesting deck, nervously told a joke or two and pulled it off. I made something and told people about it. And then an amazing thing happened, everybody wanted to network with me! My talk was on the second day, so people who had actively ignored me the first day were now introducing themselves, or sending their lackeys to arrange a meeting. I got invited to the cool nerdy after parties where everyone gets too drunk and tells you all the juicy data.

This isn’t even an 80/20 effort. Just giving a 15 minute presentation yielded 99% of the value of the conference with 1% of the effort. Unless you have a deathly fear of public speaking, sitting on stage and saying “look at this cool thing I have” is a lot less scary than walking up to a stranger and saying “Hi I’m Tyler.” At least to me.

After that my conference trips were both more effective and much more pleasant. I focused almost all my efforts on getting a good speaking slot and giving a kick ass talk and mostly didn’t worry about the rest.

The general pattern holds well too. Since I put my suits in storage and quit my conference jockey gig, I’ve spent the last four years making stuff and blogging about it. Blog posts have been by far the best source of any kind of networking for me.

Inbound requests from folks who want to partner or help with my businesses all come from blog posts. Interesting people who just want to have a chat, come from blog posts. Fellow entrepreneurs who want to swap stories, find me through my blog posts. The blog-first startup is getting a lot of traction as a business tactic because it allows you to cheaply make something useful and show it to people before you ever pitch an investor. If done well, investors read what you’re putting out there and reach out to you!

I’m sure I’ll think of other good distribution methods for the make and tell people about it but for now it’s all about the blog game for me. Or rather the full pattern is: make something awesome, blog about it and then Buffer it, and then later repost it to Medium.

People always think the end-game to actually get value out of your blog is to monetize it with ads or something. But I think the biggest benefit by far is it allows you to do “pull networking.” You write down “these are the things I’m thinking about and like talking about” and others can read them and reach out to you if they like thinking and talking about those things too. That is infinitely more valuable than the $0.00005 per page view for one weird trick ads in your sidebar.

Making things, by the way, is also the only way to network upwards. What do I mean by that? Well let’s say you want to meet someone you look up to and admire: a writer, film star or entrepreneur. If you deploy all the traditional networking tactics incredibly well, they might be impressed and maybe you would get to meet them and gram a selfie with them. But let’s say you actually want to become friends, confidants and colleagues with these people. No amount of cold-calling, leveraging alumni networks or stunts can get you there. But write a best-selling novel, Kickstart a film in Sundance, or build a fast-growing startup and now you get invited over for drinks and conversation. The only way to build a real relationship with someone who has hugely asymmetric demands on their time is to make something they admire.

Sadly there’s no one weird trick here. The making of awesome things is massively harder than becoming a LinkedIn Group power-user, but it’s massively more rewarding too.

  1. “tax equity” for the energy geeks ↩