Taming Metrics

When I set out to start my first SaaS business, my dream was to build a product that delivers unquestionable value at scale.  The second part of that vision was to build something that grows consistently and predictably over time.

I believe that with Restaurant Engine, I achieved the first part. It has proven it’s value to a number of customers, and the business model scales nicely.

Now my focus has moved onto growing the business.  Up to this point, our customers have primarily come through organic channels:  Search, referrals, and offline word of mouth.  Now the challenge is to uncover the strategies that will push growth further (and faster).

In my last article, I detailed our system for selling, which covered how we’ve overhauled our system for capturing leads, filing them in our CRM (Trell0), and getting on the phone and following up.

Today I’m sharing the next step in this process:  Making it all measurable and data-driven.  I think I’ve finally figured out how to tame the metrics in my business.  I’ll share my system with you in a moment.  But first, how did I get here?

Shooting in the dark

I tried many new strategies over the past two years, without really knowing which ones are (or aren’t) moving the needle.

I’ve been shooting in the dark.  I know that the only way to make progress is to keep a close eye on the metrics.  For me, that’s proven to be easier said than done.

I’ve had Google Analytics installed since day one.  And I’ve tried all of the other analytics tools and sliced and diced the data.  The only thing the data tells me is that organic is our most reliable source of new customers.  Great.  But I already knew that.  I need the data to tell me something I don’t know.

So maybe I’m just measuring the wrong things.  In some cases, I haven’t found an easy mechanism for measuring certain key metrics.  For example, it’s difficult to track customer signups when they open their account over the phone (which many of them do).  These signups won’t show up as Goal conversions in Google Analytics.

But most of the time, my challenge has been about the process.  I simply haven’t learned how to be consistent about reviewing and analyzing the metrics.  It hasn’t been engrained as part of my routine. And for that reason, metrics haven’t played much of a factor when deciding on new strategy directions.

Becoming data-driven

Why has this been such a challenge?  Is it just me?  Possibly.  I’m a visual thinker who likes to find the simplest way of looking at things.  Digging deep into charts and spreadsheets isn’t something I particularly enjoy.

For a while, I assumed that was it.  I read countless blogs about conversion optimization, SaaS metrics, and the like, and still couldn’t find ways to tame the metrics in my own business.

But then I realized why all of those articles couldn’t give me more than a general, broad strokes understanding of metrics.  Just like everything else in this business, metrics is something I can only learn by doing.

Every business operates differently.  Customer acquisition funnels come in all shapes and sizes.  For example, some apps offer free trials, while some have completely free plans.  Others don’t have any free option.  Some are one-off purchases, others are recurring like SaaS.  Some have a setup fee, most do not. Some rely heavily on making the sale over the phone or in person.  Others stick to a 100% web-based checkout.

No single article can teach you how you should be measuring the metrics in your business.  And in my experience, there isn’t a single tool for the job either. It was this realization that led me to finally figure out how to make it work for my business.

In fact, yesterday, after running a metrics report, I was able to uncover a key insight, which prompted me to shift our marketing strategy in a new direction. In the past, changes like this one would have been based solely on a hunch. But this time it’s backed up with hard data.  That’s how I know my new approach to metrics has finally “clicked”.

How I track metrics

Here’s the system that seems to work fairly well for me.  Maybe some of these ideas will work for you in your situation.

Routine

I learned that it’s all about cementing a consistent routine.  In the early days, I just logged into Analytics whenever the mood should strike (which frankly, wasn’t very often).  Obviously that’s not a very effective approach.  But to be fair, I don’t think metrics should be a top priority in the very beginning.  In the early days, the focus should be on talking to customers, refining the product, messaging, and the value proposition.

Then I began making it a monthly task to review the numbers.  This helped me gain a sense of the big picture.  I was able to see how our customers and traffic grew from month to month. Good to know, but not very actionable.

A few weeks ago, I began a weekly routine of running my metrics report.  In fact, it’s set in my calendar as a repeating to-do, every Friday morning.  The idea came from reading Buffer’s blog. Such a simple change, I can’t believe I never thought to make it earlier.

Moving to weekly metrics reports made all the difference.  Most of the activity in our business happens weekly.  Our blog article and newsletter go out weekly.  Seven days is a good period of time to test a new headline, or gather enough conversion data to make a decision.  When my metrics reports are tightly focused on 7-day periods, it’s much easier to tie specific activity (for example, a particular blog topic, or a new ad placement) to specific, measurable results (number of new leads and new signups).

Metrics first.  Tools later.

I always try to keep things simple.  But simplicity is hard to come by in today’s analytics tools.  So I learned that it’s best to focus on metrics first, tools later.

Before you decide which tool to use for tracking metrics, first figure out what are the key metrics you need to watch.  And don’t just list the ones that every marketing blog talks about.

Think about your business.  Think about the things you’re doing right now.  Think about the marketing strategies you’ve been trying or might try in the future.

Now think about which data would help you do those things better (or make better decisions).  What are the key pieces of info that might prompt you to change course in some way?  These are the metrics that matter to you.

Here are the ones that I came up with for my business, and why:

  • New customers – My primary goal is to increase the number of new customers we’re adding each month.  I’m not just looking at what this number is, but how consistent it is month to month.  If it’s very up and down, then my strategies aren’t working yet.  If it’s consistent or growing, then I’ve hit my stride.
  • Unique Visitors – Since our traffic is largely organic, I’m looking for this number to grow consistently from month to month.  On a weekly basis, this number helps me determine how successful this week’s blog and newsletter topic were.
  • Leads – Our leads are broken into a couple types:  Consultation requests, live chats, and lead generation tools like free courses, etc.  These are tallied up on a weekly basis.  I want to see a consistent or growing level of new leads each week.  If not, then something isn’t working.
  • Lead Gen Performance – We experiment with different lead generation initiatives at different times of the year.  For whichever one is active, I focus on:  Landing page conversion rate (tells me how compelling the offer is to my audience), and the email open / click rate (tells me how engaged those leads are with the free content we’re giving away).
  • PPC Performance – On each of our PPC campaigns, I look at click-through-rate (this tells me how effective the ad copy is), conversion rate (how relevant this traffic source is to the lead gen offer), cost per click, cost per lead, cost per customer (am I seeing a decent ROI on this campaign).
  • Weekly Blog Performance – Here I’m looking at traffic to this week’s blog post, how many social shares it received, and the email newsletter (which promotes the blog post) stats like open and click rate.  This tells me how well my audience received this particular topic, and helps us tailor our upcoming editorial calendar accordingly.

Google Spreadsheets

I learned that I can’t rely on a single metrics tool to report and organize all of the things I need to track.  So I gather data from several places, and organize it all in Google Spreadsheets.  I went through several iterations of my spreadsheet before I figure out the most useful way to organize it.

My goals for the perfect spreadsheet were:

  • It has to be simple and easy to get my key information at a quick glance.
  • It has to make it easy for me to compare both week-to-week and month-to-month trends.
  • It must incorporate free-form notes, in-line with the data, so I can identify actionable insights that help me make decisions.

At first, I tried building a single, massive spreadsheet.  That didn’t last long.  It was just a wall of numbers, with no way to compare weekly and monthly trends.  Plus, as our marketing changes (and it changes often), it throws off the whole spreadsheet.  I need something that’s fluid and can easily change along with our changing strategies throughout the year.

So I decided on a multi-sheet approach.  I started by creating a folder in Google Drive called “Metrics Journal”.

Within my “Metrics Journal” folder is a spreadsheet for the current month, which I named “April 2014”.  The benefit of having a different spreadsheet for each month is I can keep the rows of metrics relevant to whatever’s going on in our business at the current moment.  But I still have ability to open up previous months to compare notes.

The monthly spreadsheet has 5 tabs: One for each week of the month, and a fifth for the whole month recap.  The same set of columns and rows are copied to each of the weekly tabs.  This makes it easy to flip between consecutive weeks and see how the numbers change.

I organized the spreadsheet with the following columns:

  • Category – I group sets of metrics into categories, to make it easy to find what I’m looking for at a quick glance.
  • Metric – The specific metric being reported.
  • Stats – The data is entered (or calculated) here.
  • Notes – Perhaps the most important column.  This is where I make a comment on the notable pieces of data.  For example, yesterday, in the row for landing page conversion rate, I noted “Wow!  Organic traffic converts 5x better than PPC traffic.”  These notes help me make decisions.
  • Instructions – Here I add a link to another Google Doc, which details the procedure for collecting this specific piece of data (which app to open, where to find the data, etc.).  It’s a tedious process to gather all of it up, so documenting makes it much easier (plus it will be very useful when I soon outsource the preparation of my weekly marketing report).
My "Metrics Journal" spreadsheet.  It combines key data points, with notes about progress.

My “Metrics Journal” spreadsheet. It combines key data points, with notes about progress.

Here’s a public version of this spreadsheet, if you’d like to try it.  To use it, click “File” > “Make a copy…”, then adjust the rows and columns as you like.

Running the weekly report

As I said, I run my weekly metrics report every Friday morning.  It’s a very time consuming process to gather up each of the data points.  It takes me about two hours (but soon a lot of this will be outsourced.  More about outsourcing in a moment).

Let me describe some of the ways I gather up the data:

  • Stripe – The customer counts are found here.
  • Google Analytics – Many of my key metrics are gathered here.  Unique visitors is easy to find.  But some of the other metrics took me a while to figure out where they’re buried.  For example, conversions from a particular PPC campaign.  Or number of views of a single landing page.  I consolidated all of my key metrics into a custom Dashboard in Google Analytics, which pulls them into view in a one click (learn more about creating a custom Google Analytics dashboard).
  • Adwords / Facebook Ads / Perfect Audience / Adroll – Depending on which ad networks I’m using at any given time.  This is where I find the number of ad impressions, click-through-rates, and amount spent.
  • Mailchimp – Weekly newsletter open and click-through rate.  Autoresponder open rates.
  • Gmail – I actually track lots of key metrics using Gmail labels.  New customer signups are most accurately tracked this way (because not all signups show up in Google Analytics, like those who signed up over the phone, but all of them show up in my email).  Another thing I track this way is cancellations.  Every cancellation request comes to my email and gets a label.  Every week, I run a search query for the gmail label and date-range to quickly tally up the count.

I gather up the hard data from the above sources, and manually enter it into the spreadsheet.  Then there are the calculations, like conversion rates and cost per customer.  These are done automatically in the spreadsheet (you’ll see that I marked these cells in light yellow, which indicates they are calculations, not hard entry).

Outsourcing

This is my next step.  Soon, I’ll be bringing on a new teammate to help with running these weekly reports.  Their job will be to follow the instructions (which are linked right from the spreadsheet) to gather and enter all of the hard data each week.

With the data prepared each Friday, all that’s left for me to do is log in, study the numbers, and draw meaningful conclusions about the trends.  I’ll note these in the notes column, and keep them in mind as I set my to-do list for the next week.

Improvements

As always, much can be improved.  Here are a few things I’m thinking about for the coming months:

  • Cost to acquire a customer.  The goal for any marketing plan is to ensure that the cost to acquire a customer remains less than the customer lifetime value.  I’m still working to reconcile the costs of organic and paid channels, which has proven to be very tricky for my business.  Luckily, we have a very low churn rate, which means we have a healthy lifetime value to work with.
  • Automation.  As I said, I plan to outsource the data-gathering piece, which is the most time-consuming part.  But I’m still holding out hope that more of this can be automated using some combination of tools out there.  For now, I’m sticking with my spreadsheet 🙂

Have you tamed your metrics?

Tell me how you go about watching your numbers from week to week.  I want to compare notes and figure this metrics thing out, once and for all.

  • Hmm, I like this idea, but I think Weekly reports for some of this data might be overkill.

    Some of it feels like data collection for the sake of — I can certainly understand metrics that impact the bottom line — visitors, uniques etc — but social shares and the e-mail campaigns seem something that would be easier to manage on a 30-day.

    Hoping you can prove me wrong! 🙂

  • Don Osborne

    Nice job, Cas. I also track my metrics weekly. Three thoughts: 1) I capture my impression and backlink data from Webmaster tools each week; since Webmaster tools only keeps a 90-day history, this data is perishable unless I record it. 2) When the time comes, consider adding in a row for referrals; you’ll need to build a referral (or affiliate) system in time, and this approach may surprise you for how effective it can be to grow your customer base. And (also much later) 3) build in a way to measure larger joint venture opportunities — e.g., when you speak in front of a large Restaurant Association, you’ll get a big spike in your traffic. This will create noise in your analytics, unless you’re able to track the referrer. A specific landing page works a treat!

  • Brett Thoreson

    I read Lean Analytics a few months ago. It included some great concepts, one of which was finding your “One Metric That Matters” to reduce data overload. I would definitely recommend it to anyone with a saas/web app. Here’s a link – http://www.amazon.com/gp/product/1449335675/ref=as_li_tf_tl?ie=UTF8&camp=1789&creative=9325&creativeASIN=1449335675&linkCode=as2&tag=leananalytics-20