The Order of AARRR

Back in December 2009 when Patrick Vlaskovits and I first contemplated our Customer Development book, I was noodling around with a graphic to illustrate the integrate Dave McClure's Pirate Metrics with Customer Development activities. I posted the graphic in a blog post which garnered lots of attention, including an important comment from Dave himself:

btw, note that AARRR isn’t exactly sequential… in fact, i’d emphasize Activation & Retention *first*, then go after Acquisition & Referral, then optimize for Revenue

Recently, I have caught a couple of my Customer Development clients optimizing the metrics in the "wrong" order, so Dave's comment bears repeating and perhaps some expounding upon.

The Pirate Says "AARRR"

The anagram is ordered by user experience:

A - Acquisition - User is directed to your site;

A - Activation - User signs up or is otherwise engaged;

R - Retention - User keeps coming back, i.e., is engaged over time;

R - Referral - User invites others;

R - Revenue - User pays or is otherwise monetized;

This is not, however, the order in which you should optimize your product while building your business.  The order (and truly, the metrics themselves) are dependent upon your business model.

For B2C Free, Say RRAAR

lion roaringRetention - Nail engagement so that users want to or must come back;

Referral - Your business model requires massive user growth, so your product must require referrals to work or must be so cool it just happens;

Activation - Once your users are here to stay and inviting others, optimize their conversion funnel.

Acquisition - Now you're ready to blow up acquisition.  (Good time to get investment funding.)

Revenue - Millions of users?  Time to monetize.

For B2C or B2B Freemium, Cheer RRRAA!

crowd cheeringRetention - Nail engagement so that users want to or must come back;

Revenue - If your business model requires users or businesses to pay, you need to figure out what they'll pay for before you blow-up anything;

Referral - Since you're likely not a true network-effects business, you optimize referral after revenue.  Optimizing referral is actually part retention, part funnel optimization.  Users are so happy they will refer others to your site or provide you testimonials or speak to the press, etc.  Not much sense in blowing up your acquisition, however, until you have that level of passion.

Activation - This refers to Blank's "Sales and Marketing roadmap."  Here you understand and optimize your sales funnel.

Acquisition - OK, now you can hire that PR firm.

For Enterprise B2B, Cheer an alternative RRRAA!

R - Revenue - If you're selling to businesses the number one thing you need to prove is that someone cares enough about what you're providing to give you money for it.

R - Referral - As above, referral in this context means that your customers are willing to sing your praises publicly.

A - Activation - In this context, activation is understanding your sales and marketing roadmap.

A - Acquisition - After you nail your sales and marketing roadmap, you're ready to feed the top of the funnel.

R - Retention - In enterprise B2B, you often don't have a subscription model, but may be charging annually for maintenance and support.

Note that you will always need to do some level of "acquisition" in order to figure out and optimize the other stuff.  But the point is that you're not concentrating on or optimizing acquisition at the start.  Exact order of AARRR is debatable from one business to another, so use above as guidelines only.


  1. Tristan Kromer February 14, 2011 at 2:52 pm

    Hi Brant,

    I rarely find myself disagreeing with you so I’m tentatively assuming I’ve misunderstood your intent.

    You suggest for free of freemium B2C to go after Retention and Referral prior to Activation. But it is unclear to what extent. It seems like you’re hinting at “don’t optimize,” since most of the time you’re going to have to have *some* level of Activation before you can hope to have someone coming back again and again for Retention.

    I feel like your post overly suggests that these metrics should be optimized in a particular order. The reality is you need to switch back and forth between them.

    Before you test anything, you’re going to have to spend *some* time on customer acquisition, even if only to get the first 5 people for customer development. Clearly you don’t need to optimize that process, but you need to get started.

    As you start testing Activation and want to move from qualitative to quantitative data, it’s going to be necessary to get at least a decent sample size for testing. That means you’re going to be going back to Acquisition again and again as your funnel starts developing.

    If you’re able to iterate your prototype on a weekly basis, 100-150 customers on a web app is plenty for a sample size, so that’s all you need. If you’re running split tests daily, then you need 100-150 customers per day, and that means more time spent on the acquisition funnel to make sure you have sufficient customers to test with.

    Short iteration times with continuous deployment with zero customers is as inane as spending 6 months writing an 80 page business plan on a FB app.

    The way I understand Dave’s metrics and the whole concept is to focus on the metrics that lead to the most learning…not arbitrarily optimizing in a predefined process.

    • brantcooper February 14, 2011 at 3:53 pm

      Thanks for the comment, Tristan. IMO, you’re not “optimizing” acquisition if you’re just trying to get enough users to test your basic value prop, i.e., retention. I’d imagine you wouldn’t want to get the top of your funnel too hot either, before you know how to get them through it. You can certainly “learn” about acquisition while getting enough users to optimize other aspects of the business model.

      I suspect we’re basically in agreement.

      • Tristan Kromer February 14, 2011 at 5:44 pm

        Yes, ok…thanks for clarifying. I think I was over reading.

  2. Shaw June 6, 2014 at 4:34 am

    Hi Brant,

    I found this post interesting. Would you elaborate on why or what data made you want to reorder the funnel depending on the business type? It seems counter-intuitive to me, so I’d like to hear you elaborate on this post. And how about a two-sided market where you face consumers on one side (i.e., B2C), but businesses on the other side (i.e., B2B) such as Esty?

Leave a Reply

Your email address will not be published. Required fields are marked *

Your Name *
Your Email *