Tag Archives: ARR

Go-To-Market Troubleshooting:  Let’s Take It From The Top

Posted on July 23, 2024 | 1 comment

So, you’re missing plan and revenue growth is down.  Well, welcome to the club.  You’re certainly not alone in these times. 

In this post, I’ll discuss what you can do about it – specifically, how you can apply some of the ideas I’ve discussed in Kellblog to troubleshoot go-to-market (GTM) performance.  I’ll focus on troubleshooting new business (“newbiz”) ARR plan attainment, the area where most companies seem to be having the most trouble [1].

Don’t Knee-Jerk Blame the Plan

The immediate temptation when missing plan is to blame the plan.  “It’s not realistic.”  “It was driven by the fundraise, not the bottom-up.”  But blaming plan is a poor place to start for two reasons.                        

First, you signed up for the plan when you submitted it to the board for approval.  Next time, if you don’t believe in a proposed plan, don’t be so quick to fold in the face of internal pressure.  Remember the old Fram oil filter commercial and think, “you can fire me now or fire me later” so if you’re asking me to sign up for a plan that I don’t think I can achieve, you might as well fire me now [2].  The need to make such difficult judgments is the price of admission to the sales leadership role.  Cop out at your own peril, because they will indeed fire you later.

Second, when you follow the approach in this post, if the plan is unachievable it will emerge from the data.  So, bite your tongue, avoid any initial temptation to blame plan, and instead go look at the funnel.

The Two Questions and Two Metrics

Recall in this post, I argued that you should ask two questions when you’re missing plan.   Every quarter: 

  1. Are we giving sales the chance to hit the number?
  2. Is sales converting enough of the pipeline to hit the number?

That’s it.  Everything comes down to these two questions.  No matter the root problem, it will be revealed in answering them.  Remember, the way to make plan for twelve consecutive quarters is one at a time.  So why not focus on next quarter?  And if you’re chronically missing plan, why not make a steady-state assumption to simplify things further? [3]

Starting with the above two questions makes things simple by breaking the entire funnel in two.  Simplifying the problem is important because you can quickly and irrecoverably descend into analytical quicksand.  When I first meet them, many companies are neck-deep in such quicksand, comparing dashboard clips, reports, and spreadsheets derived from different systems, lost in an endless sea of non-footing detail, having completely lost the business forest for the salesops trees.

Note that neither of the two above questions assigns blame.  As a consultant, I have the distinct advantage of not caring where the trouble is, making me a disinterested party, de facto impartial.  I encourage CXOs to adopt a similar approach, simply stating facts, avoiding blame (e.g., inferred causes), and acting as dispassionate analysts when analyzing GTM problems.  While you will eventually need to ask why you have certain problems, it’s always best to start with simple statements of fact, get agreement on them, and build from there.  For example:

  • “We consistently start quarters with insufficient pipeline coverage” is a blameless statement of fact.  It does not say whose job it is to generate pipeline (if that’s even been detailed out across sources) or why they are failing to do so.
  •  “We are converting a below-normal percentage of our week 3 pipeline,” less obviously, is also a blameless statement of fact.  While it’s clearly the job of sales to convert pipeline, the statement makes no assertion as to why we are seeing abnormally low conversion rates (e.g., pipeline quality, change in competitive market, sales execution).

When it comes to metrics, the first of the above questions is measured by pipeline coverage, more precisely week-3 pipeline coverage [4] [5]. The second is measured by a conversion rate, specifically week-3 pipeline conversion rate.  Notably, this is not a win rate, and please read this post to ensure that you understand why.

Are We Giving Sales a Chance to Hit the Number?

Make a chart like this one to answer this question.

Here you see, for newbiz ARR for the trailing nine quarters, week 3 pipeline dollars, week 3 pipeline coverage (pipeline/plan), ARR booked, week-3 pipeline conversion, and the pipeline coverage target implied by the week-3 conversion rate (i.e., its inverse).  Pipeline conversion rates are more interesting when viewed in conjunction with plan attainment, so I’ve added ARR plan and plan attainment as well. 

Analyzing this chart, we can see a few things:

  • From 1Q22 through 1Q23 we converted about 33% of the pipeline
  • We were also consistently hitting plan in that timeframe
  • Starting 2Q23 we started with only 2.3x coverage, converted a healthy 40% of it, but still came up short, at 91% of plan. 
  • That rough pattern continued in 3Q23 and 4Q23
  • 1Q24 started with the weakest coverage in the past nine quarters (1.9x)
  • While sales is forecasting record conversion of that pipeline (45%), we are nevertheless forecasting to land at only 86% of plan
  • I’m not sure I believe the forecast because 45% conversion is borderline unrealistic and could simply be the CRO trying their best to hold the line

I conclude that this company is starting with insufficient pipeline.  That is, they’re not giving sales a chance to hit the number.  How do I conclude that?

  • By comparison to pipeline coverage benchmarks.  3.0x is the typical pipeline coverage goal and you’ll note that in the good times (1Q22 through 1Q23) we consistently started with 3.0x+ and we consistently made plan. 
  • By comparison to pipeline conversion benchmarks.  33% is a standard conversion rate.  Here we are running at 40%+, which is best-in-class conversion.  Pipeline conversion is not the problem.
  • More importantly, by comparison to ourselves.  In our recent history, we consistently made plan when we started with 3.0x+ coverage and missed it when we started with 2.3 to 2.4x.  This quarter (1Q23) we’re starting with 1.9x, forecasting record conversion, and still only 86% of plan.

The solution to the insufficient pipeline problem is, unsurprisingly, to make a plan to generate more pipeline.

Here are some of the high-level steps in making that plan:

  • Define pipeline generation targets across the four major pipeline sources.  It’s surprising how many companies don’t start with this basic step.  For bonus points, over-allocate the goals to target 110% of what you need. [6]
  • I prefer to set these targets by opportunity count, not pipeline dollars, because I think it’s more visceral and less easily gamed [7].
  • Do a cost/oppty analysis across your pipeline sources to get an idea of how much money any given pipeline source (e.g., alliances, demandgen) would need to create, for example, 20 more oppties next quarter.  Remember to focus on variable, not average, cost [8].
  • Be sure to check with the leader of each pipeline source on their ability to absorb extra money to generate more pipeline.  If you have 12 SDRs reporting to one manager, they may need to bring in another manager before hiring 3 more SDRs.  Alternatively, sellers may have extra time on their hands and the ability to put more time into outbound.  Alliances may have a hot candidate they want to hire, but no open headcount, and could execute quickly if one were opened.  It’s not just about money; it’s about the ability to productively spend it.
  • Accept that you may be overallocated to sales versus pipeline generation.  In this case, the best solution might well be to terminate the bottom N sellers and convert the newly liberated budget to pipeline generation — so that everyone else has a chance at success.  This is painful, but sometimes necessary, and after you’ve had to do it once, you’ll be more careful to plan holistically in the future.

This all goes without saying that no pipeline analytics will work if you lack basic pipeline discipline – i.e., if you don’t have clear definitions for stages, close dates, oppty values, and forecast categories, and if you don’t regularly enforce them via periodic pipeline scrubs.

The Floating Bar Problem

Before diving into pipeline conversion, let’s address a special case of insufficient pipeline:  one where the pipeline initially looks sufficient but burns off at an above-average rate across the quarter.  You can see this by looking weekly at to-go pipeline coverage.

What’s usually happening in these cases is that some material percentage of your week-3 pipeline is effectively fake.  This happens because, when pipeline is scarce and if sellers are under pressure to each carry 3x coverage [9], they will take lower-quality opportunities into their pipeline.  For example, long-shot oppties that appear rigged for the competition, immature oppties where sellers hope to create a buying timeline, or self-nurture leads that may only become real oppies in the future.

I call the tendency to work on lower quality oppties in tough times, the “floating bar problem” because sales silently lowers (or in good times, raises) the bar for admission into the pipeline.  This is insidious because the result is fake pipeline that creates an illusion of coverage which disappears as the quarter progresses.

The solution to his problem is simple in theory, but hard in practice.

  • Sales management needs to hold the line on what gets into the pipeline, applying the same standards in tough times as good ones.
  • If sales management wants to allow sellers to work on low probability “oppties,” that’s fine but, well, get them out of the opportunity management system.  Use tasks to track work.  But only promote a lead to an oppty when it meets the standard for being an oppty.

If, for example, SDRs are passing low quality stage-1 oppties to sales that should not show up in the numbers as a reduced pipeline conversion rate.  Instead, it should show up in a higher stage-2 rejection rate.  This point is completely lost on most sales managers so please make sure you understand it.  If you maintain pipeline discipline, lower quality oppties should show up not as a reduced stage-2 to close rate, but as an increased stage-2 rejection rate.  And pipeline discipline starts at stage 2 – where sales decides to accept or reject oppties.  It’s wrong to accept sub-standard oppties, pollute the oppty management system with fake pipeline, convert little of it, miss plan, and wreck the company’s pipeline analytics in the process.

I’m not trying to prevent sales from working on whatever sales management wants them to work on.  But I am saying one thing:  whatever they are, don’t call them oppties in “my” oppty management system if they don’t meet the defined standards for oppties [10].

Is Sales Converting Enough of the Pipeline?

While it’s the job of sales to convert pipeline into ARR, that doesn’t mean sales execution is the only factor that drives conversion rates.

Here you see conversion rates plummeting, dropping by 11 percentage points between 1Q23 and 2Q23 and then by another 5 percentage points by 4Q23.  By the 1Q24 forecast, the pipeline conversion rate has been effectively cut in half from ~32% to ~16%.  Note that during the recent dark times (from 2Q23 to 1Q24) we have been starting with ~3.0x pipeline coverage, but converting so little that we’re landing in the dismal range of 47% to 65% of plan.

Let’s assume we have the operational basics covered, so this is real pipeline, validated and scrubbed by sales management, and held to consistent standards over time.  But we’re converting a lot less of it than we used to.  Thus, I conclude that the company’s problem is pipeline conversion, not pipeline coverage.

What possible factors could be driving reduce pipeline conversion rates?  Well, there are a lot of them, so we’ll talk about each.

  • Changes in averages (i.e., ceteris non paribus).  Most productivity models assume a constant average sales price (ASP) and average sales cycle length (ASC).  If ASPs go down, you will hit your count-based targets, but miss your dollar-based ones.  If ASCs increase you may preserve your eventual close rates, but stretch them out over time, reducing quarterly conversion rates and plan attainment. 
  • ASP decreases.  Typically, due to budgetary pressure and increased price competition, but also can be due to an overreliance on discounting.  Some of this is inevitable in a downturn.  You can mitigate it through pricing and packaging changes (e.g., new add-ons to preserve price and/or offset churn at renewal).
  • Slip rate increases.  When ASCs lengthen, more deals slip to the following quarter(s).  Pipeline scrubs can provide early detection and deals reviews can offer re-acceleration strategies.  The biggest risk is that these deals never close at all and simply hit no-decision or derail.
  • Win rate decreases.  Win rates usually decrease when a new competitor enters the market or when an existing competitor leapfrogs your product or your market position (e.g., passes you in market share).  Competitive research, sales training, and selling the roadmap are the usual responses.  
  • An absence of big deals.  Some CROs run their business as a mix of baseline deals to hit say 60-80% of plan, topped up by big deals that provide the rest.  During a downturn those big deals may evaporate leaving only the run-rate business.  The usual response is a strategic accounts program to focus on generating big deals and a focus on pipeline generation in the run-rate business to cover the gap.
  • Pipeline substitution.  This is a subtle problem due to a change in pipeline mix, with low-converting pipeline substituting for high-converting pipeline.  This is dangerous because you “look covered” at the start of the quarter but end up below plan at the end.  Let’s drill in a bit here.

Pipeline Substitution

Not all pipeline is created equal.  Pipeline for certain products often converts at a higher rate than others.  Pipeline conversion rates typically vary by source, e.g., with outbound SDRs typically converting at a low rate and alliances converting at a high rate.  Pipeline conversion might also vary by geography, with established geographies delivering high conversion rates than emerging ones. 

See this chart for an example:

In this example, we start every quarter with $10M in pipeline.  In 1Q23 through 3Q23 we convert 25% of it, but in 4Q23 we convert only 20%.  What happened?  The pipeline mix changed.  Starting in 4Q23, we substituted $2M in high-converting pipeline (from sales/outbound and alliances) with $2M in low-converting pipeline (from SDR/outbound).  Blended pipeline conversion thus dropped from 25% to 20% as a result of this change, effectively substituting nutrient-rich pipeline for nutrient-poor pipeline while keeping the overall amount the same. 

Identifying these problems is a lot of work because you’ll need to segment pipeline by multiple variables — such as pipeline source, product, geography, business segment (e.g., enterprise vs. corporate accounts) – to get historical average conversion rates and percent mix, and then see if changes in pipeline composition are driving reductions in conversion rates.  If so, the usual solution is to re-aim your pipeline generation back to the high-converting segments.

In this post, we have shown how you can troubleshoot go-to-market problems by splitting the funnel in two and focusing on two questions:

  • Are we giving sales the change to hit the number each quarter, as measured by pipeline coverage.
  • Is sales converting enough of the pipeline to hit the number, as measured by pipeline conversion.

I’ve also provided numerous notes and links that you can use to deepen your knowledge of how to solve these problems.

# # #

Notes

[1] The same analysis approach can easily apply to expansion ARR, which should be analyzed independently via its own funnel because it typically has different conversion rates and shorter sales cycles. 

[2] Deadheads will understand that I had to resist writing, “nothing else shaking, so you might just as well.”

[3] Think:  given that we’re off rails, forget the plan for a minute and let’s analyze what do we need to do to add $4M in newbiz ARR every quarter?  This liberates you from needless, complexifying math that makes it harder to see the answer and is a great way to start in the crawl-walk-run exercise of getting back on track.

[4] More precisely, day-1, week-3, current-quarter pipeline coverage.  Snapshotting Sunday night before the start of week 3 gives you a consistent point to compare across quarters.  Waiting until the start of week 3 gives sales (more than) enough time to clean up the pipeline after the end of the prior quarter but is still early enough to be considered “starting pipeline.”  Note that you may need to apply corrections for any deals that close in the first two weeks of the quarter.  A high-class problem, at least.

[5] Or, in a monthly cadence, day-3 pipeline coverage.  See my post on the mental mapping from quarterly to monthly cadence for more on this concept.

[6] There is a cost to this type of insurance; it’s not great for your CAC ratio if you don’t end up over-performing plan (which ceteris paribus, starting with 110% of your pipeline target, you should).  But it does reduce the risk of missing plan.  To me, the correct sequence is to focus on making plan first, before focusing on efficiency — but you need to have the cash to underwrite that philosophy.

[7] For example, one big deal that masks what’s otherwise a pipeline starvation situation.  If you’re going to set targets on dollars (which typically involves using some placeholder value) then you should create the oppties with a close date far in the future (e.g., one year) that sales can pull forward once they further qualify the account.  The alternative is usually generating lots of fake pipeline that is auto-dumped into next quarter that gets pushed out in the first weeks of the quarter.  Also, see this for more on ensuring pipeline coverage by seller, and not just in aggregate.

[8] You’re not going to hire an extra CMO, an extra PR agency, and an extra product marketer to generate 20 more oppties.  Those costs are effectively fixed.

[9] And putting them under such pressure can run in diametric opposition to pipeline discipline and enforcing pipeline standards by encouraging reps to enter dubious deals as pipeline to get their manager off their backs.

[10] I say “my” oppty management system to remind people that carrying sub-standard oppties has impacts well beyond themselves and that oppty management system is the company’s property, not theirs.  For old movie fans, when speaking of the oppties in “my” oppty management system, I’m always reminded of Cool Hand Luke: “what’s your dirt doing in Boss Keen’s ditch?”

1 Comment

Posted in GTM, Pipeline, Sales, Salesops

Tagged , , , ,

What Are The Units On Your Lead SaaS Metric — And What Does That Say About Your Culture

Posted on February 22, 2021 | Leave a comment

Quick:

  • How big is the Acme deal?  $250K.
  • What’s Joe’s forecast for the quarter?  $500K
  • What’s the number this year?  Duh.  $7,500K.

Awesome.  By the way:  $250K what?  $500K what?  $7,500K what?  ARR, ACV, bookings, TCV, new ARR, net new ARR, committed ARR, contracted ARR, terminal ARR, or something else?

Defining those terms isn’t the point of this post, so see note [1] below if interested.

The point is that these ambiguous, unitless conversations happen all the time in enterprise software companies.  This isn’t a post about confusion; the vast majority of the time, everyone understands exactly what is being said.  What those implicit units really tell you about is culture.

Since there can be only one lead metric, every company, typically silently, decides what it is.  And what you pick says a lot about what you’re focused on.

  • New ARR means you’re focused on sales adding water to the SaaS leaky bucket — regardless of whether it’s from new or existing customers.
  • Net New ARR means you’re focused the change in water level in the SaaS leaky bucket — balancing new sales and churn — and presumably means you hold AEs accountable for both sales and renewals within their patch.
  • New Logo ARR means you’re focused on new ARR from new customers.  That is, you’re focused on “lands” [2].
  • Bookings means you’re focused on cash [3], bringing in dollars regardless of whether they’re from subscription or services, or potentially something else [4].
  • TCV, which became a four-letter word after management teams too often conflated it with ARR, is probably still best avoided in polite company.  Use RPO for a similar, if not identical, concept.
  • Committed ARR usually means somebody important is a fan of Bessemer metrics, and means the company is (as with Net New ARR) focused on new ARR net of actual and projected churn.
  • Terminal ARR means you’re focused on the final-year ARR of multi-year contracts, implying you sign contracts with built-in expansion, not a bad idea in an NDR-focused world, I might add.
  • Contracted ARR can be a synonym for either committed or terminal ARR, so I’d refer to the appropriate bullet above as the case may be.

While your choice of lead metric certainly affects the calculations of other metrics (a bookings CAC or a terminal-ARR CAC) that’s not today’s point, either.  Today’s point is simple.  What you pick says a lot about you and what you want your organization focused on.

  • What number do you celebrate at the all hands meeting?
  • What number do you tell employees is “the number” for the year?

For example, in my opinion:

  • A strong sales culture should focus on New ARR.  Yes, the CFO and CEO care about Ending ARR and thus Net New ARR, but the job of sales is to fill the bucket.  Someone else typically worries about what leaks out.
  • A shareholder value culture would focus on Ending ARR, and ergo Net New ARR.  After all, the company’s value is typically a linear function of its Ending ARR (with slope determined by growth).
  • A strong land-and-expand culture might focus on Terminal ARR, thinking, regardless of precisely when they come in, we have contracts that converge to a given total ARR value over time [5].
  • Conversely, a strong land and expand-through-usage culture might focus on New Logo ARR (i.e., “land”), especially if the downstream, usage-based expansion is seen as somewhat automatic [6].
  • A cash-focused culture (and I hope you’re bootstrapped) would focus on bookings.  Think:  we eat what we kill.

This isn’t about a right or wrong answer [7].  It’s about a choice for your organization, and one that likely changes as you scale.  It’s about mindfulness in making a subtle choice that actually makes a big statement about what you value.

# # #

Notes
[1] For clarity’s sake, ARR is annual recurring revenue, the annual subscription value.  ACV is annual contract value which, while some treat as identical to ARR, others treat as first-year total contract value, i.e., first-year ARR plus year-one services.  Bookings is usually used as a proxy for cash and ergo would include any effects of multi-year prepayments, e.g., a two-year, prepaid, $100K/year ARR contract would be $200K in bookings.  TCV is total contract value which is typically the total (subscription) value of the contract, e.g., a 3-year deal with an ARR stream of $100K, $200K, $300K would have a $600K, regardless of when the cash payments occurred.  New ARR is new ARR from either new customers (often called New Logo ARR) or existing customers (often called Upsell ARR).  Net New ARR is new ARR minus churn ARR, e.g., if a regional manager starts with $10,000K in their region, adds $2,000K in new ARR and churns $500K, then net new ARR is $1,500K.  Committed ARR (as defined by Bessemer who defined the term) is “contracted, but not yet live ARR, plus live ARR netted against known projected ARR churn” (e.g., if a regional manager starts with $10,000K in their region, has signed contracts that start within an acceptable time period of $2,000K, takes $200K of expected churn in the period, and knows of $500K of new projected churn upcoming, then their ending committed is ARR is $11,500K.  (Why not $11,300K?  Because the $200K of expected churn was presumably already in the starting figure.)  Terminal ARR the ARR in the last year of the contract, e.g., say a contract has an ARR stream of $100K, $200K, $300K, the terminal ARR is $300K [1A].  Contracted ARR is for companies that have hybrid models (e.g., annual subscription plus usage fee) and includes only the contractually committed recurring revenues and not usage fees.

[1A] Note that it’s not yet clear to me how far Bessemer goes out with “contracted” ARR in their committed ARR definition, but I’m currently guessing they don’t mean three years.  Watch this space as I get clarification from them on this issue.

[2] In the sense of land-and-expand.

[3] On the assumptions that bookings is being used as a proxy for cash, which I recommend, but which is not always the case.

[4] e.g., non-recurring engineering; a bad thing to be focused on.

[5] Although if they all do so in different timeframes it becomes less meaningful.  Also unless the company has a track record of actually achieving the contractually committed growth figures, it becomes less credible.

[6] Which it never actually is in my experience, but it is a matter of degree.

[7] Though your investors will definitely like some of these choices better than others.

 

Leave a comment

Posted in Culture, FP&A, SaaS, Sales, Startups

Tagged ,

Does Enterprise SaaS Need a Same-Store Sales Metric?

Posted on February 18, 2020 | 3 comments

Enterprise SaaS and retailers have more in common than you might think.

Let’s think about retailers for a minute. Retailers drive growth in two ways:

  • They open new stores
  • They increase sales at existing stores

Opening new stores is great, but it’s an expensive way to drive new sales and requires a lot of up-front investment. It’s also risky because, despite having a small army of MBAs working to determine the right locations, sometimes new locations just don’t work out. Blending the results of these two different activities can blur what’s really happening. For example, consider this company:

Things look reasonable overall, the company is growing at 17%. But when you dig deeper you see that virtually all of the growth is coming from new stores. Revenue from existing stores is virtually flat at 2%.

It’s for this reason that retailers routinely publish same-store sales in their financial results. So you can see not only overall, blended growth but also understand how much of that growth is coming from new store openings vs. increasing sales at existing stores.

Now, let’s think about enterprise software.

Enterprise software vendors drive growth in two ways:

  • They hire new salesreps
  • They increase productivity of existing salesreps

Hiring new salesreps is great, but it’s an expensive way to drive new sales and requires a lot of up-front investment. It’s also risky because, despite having a small army of MBAs working to determine the right territories, hiring profiles and interviewing process, sometimes new salesreps just don’t work out. Blending the results of these two different activities can blur what’s really happening. For example, consider this company:

If you’re feeling a certain déjà vu, you’re right. I simply copy-and-pasted the text, substituting “enterprise software vendor” for “retailer” and “salesrep” for “store.” It’s exactly the same concept.

The problem is that we, as an industry, have basically no metric that addresses it.

  • Revenue, bookings, and billings growth are all blended metrics that mix results from existing and new salespeople [1]
  • Retention and expansion rates are about cohorts, but cohorts of customers, not cohorts of salespeople [2]
  • Sales productivity is typically measured as ARR/salesrep which blends new and existing salesreps [3]
  • Sales per ramped rep, measured as ARR/ramped-rep, starts to get close, but it’s not cohort-based, few companies measure it, and those that do often calculate it wrong [4]

So what we need is a cohort-based metric that compares the productivity of reps here today with those here a year ago [5]. Unlike retail, where stores don’t really ramp [6], we need to consider ramping in defining the cohort, and thus define the year-ago cohort to include only fully-ramped reps [6].

So here’s how I define same-rep sales: sales from reps who were fully ramped a year ago and still here.

Here’s an example of presenting it:

The above table shows same-rep sales via an example where overall sales growth is good at 48%, driven by a 17% increase in same-rep sales and an 89% increase in new-rep sales. Note that enterprise software is a business largely built on the back of sales force expansion so — absent an acquisition or new product launch to put something new in sale’s proverbial bag — I view a 17% increase in same-rep sales as pretty good.

Let’s conclude by sharing a table of sales productivity metrics discussed in this post that I think provides a nice view of sales productivity as related to hiring and ramping.

The spreadsheet I used for this post is available for download, here.

# # #

Notes

[1] Billings is a public company SaaS metric and typically a proxy for bookings.

[2] See here for my thoughts on churn

[3] Public companies never release this but most public and private companies track it.

[4] By taking overall new ARR (i.e., from all reps) and dividing it by the number of ramped reps, thus blending contribution from both new and existing reps in the numerator. Plus, these are usually calculated on a snapshot (not a cohort) basis.

[5] This is not survivor-biased in my mind because I am trying to get a productivity metric. By analogy, I believe stores that closed in the interim are not included in same-store sales calculations.

[6] Or to the extent they do, it takes weeks or months, not quarters. Thus you can simply include all stores open in the year-ago cohort, even if they just opened.

[6] I am trying to avoid seeing an increase in same-rep sales due to ramping — e.g., someone who just started in the year-ago cohort will have year sales, but should increase to full productivity simply by virtue of ramping.

3 Comments

Posted in Metrics, Sales, Salesops

Tagged ,

New ARR and CAC in Price-Ramped vs. Auto-Expanding Deals

Posted on August 7, 2019 | Leave a comment

In this post we’re going to look at the management accounting side of multi-year SaaS deals that grow in value over time.  I’ve been asked about this a few times lately, less because people value my accounting knowledge [1] but rather because people are curious about the CAC impact of such deals and how to compensate sales on them.

Say you sign a three-year deal with a customer that ramps in payment structure:  year 1 costs $1M, year 2 costs $2M, and year 3 costs $3M.  Let’s say in this example the customer is getting the exact same value in all 3 years (e.g., the right for 1,000 people to use a SaaS service) – so the payment structure is purely financial in nature and not related to customer value.

Equal Value:  The Price-Ramped Deal
The question on my mind is how do I look at this from a new ARR bookings, ending ARR, CAC, and sales compensation perspective?

GAAP rules define precisely how to take this from a GAAP revenue perspective – and with the adoption of ASC 606 even those rules are changing.  Let’s take an example from this KPMG data sheet on ASC 606 and SaaS.

(Price-Ramped) Year 1 Year 2 Year 3
Payment structure $1M $2M $3M
GAAP revenue $1M $2M $3M
GAAP unbilled deferred revenue $5M $3M $0M
ASC 606 revenue $2M $2M $2M
ASC 606 unbilled accounts receivable $1M $1M $0M
ASC 606 revenue backlog $4M $2M $0M

When I look at this is I see:

  • GAAP is being conservative and saying “no cash, no revenue.” For an early stage startup with no history of actually making these deals come true, that is not a bad position.  I like the concept of GAAP unbilled deferred revenue, but I don’t actually know anyone who tracks it, let alone discloses it.  Folks might release backlog in some sort of unbilled total contract value (TCV) metric which I suspect is similar [2].
  • ASC 606 is being aggressive and mathematical – “hey, if it’s a 3-year, $6M deal, then that’s $2M/year, let’s just smooth it all out [3]”. While “unbilled A/R” strikes me as (another) oxymoron I see why they need it and I do like the idea of ASC 606 revenue backlog [4].  I think the ASC 606 approach makes a lot of sense for more mature companies, which have a history of making these deals work [5].

Now, from an internal, management accounting perspective, what do you want to do with this deal in terms of new ARR bookings, ending ARR balance, CAC ratio, and sales comp?  We could say:

  • It’s $2M in new ARR today
  • Ergo calculate this quarter’s CAC with it counted as $2M
  • Add $2M in ending ARR
  • Pay the salesrep on a $2M ARR deal – and let our intelligently designed compensation plan protect us in terms of the delayed cash collections [6] [6A]

And I’d be OK with that treatment.  Moreover, it jibes with my definition of ARR which is:

End-of-quarter ARR / 4 = next-quarter subscription revenue, if nothing changes [7]

That’s because ASC 606 also flattens out the uneven cash flows into a flat revenue stream.

Now, personally, I don’t want to be financing my customers when I’m at a high-burn startup, so I’m going to try and avoid deals like this.  But if I have to do one, and we’re a mature enough business to be quite sure that years 2 and 3 are really coming, then I’m OK to treat it this way.  If I’m not sure we’ll get paid in years 2 and 3 – say it’s for a brand-new product that has never been used at this scale – then I might revert to the more GAAP-oriented, 1-2-3 approach, effectively treating the deal not as a price ramp, but as an auto-expander.

Increasing Value:  The Auto-Expanding Deal
Let’s say we have a different use-case.  We sell a SaaS platform and year 1 will be exclusively focused on developing a custom SaaS app, we will roll it to 500 users day 1 of year 2, and we will roll it to 500 more users on day 1 of year 3.  Further assume that the customer gets the same value from each of these phases and each phase continues until the end of the contract [8].  Also assume the customer expects that going forward, they will be paying $3M/year plus annual inflation adjustments.

Oy veh.  Now it’s much harder.  The ramped shape of the curve is not about financing at all.  It’s about the value received by the customer and the ramped shape of the payments perfectly reflects the ramped shape of the value received.  Moreover, not all application development projects succeed and if they fall behind on building the customized application they will likely delay the planned roll-outs and try to delay the payments along with them.  Moreover, since we’re an early-stage startup we don’t have enough history to know if they’ll succeed at all.

This needs to be seen as an auto-expanding deal:  $1M of new-business ARR in year 1, $1M of pre-sold upsell ARR in year 2, and another $1M of pre-sold upsell ARR in year 3.

When you celebrate it at the company kickoff you can say the customer has made a $6M commitment (total contract value, or TCV [9]) to the company and when you tier your customers for customer support/success purposes you might do so by TCV as opposed to ARR [10].  When you talk to investors you can say that $1M of next year’s and $1M of the subsequent year’s upsell is already under contract, ergo increasing your confidence in your three-year plan.  Or you could roll it all together into a statement about backlog or RPO [11].  That part’s relatively easy.

The hard part is figuring out sales compensation and CAC.  While your rep will surely argue this is a $2M ARR deal (if not a $3M ARR deal) and that he/she should be paid accordingly, hopefully you have an ARR-driven (and not a total bookings-driven) compensation plan and we’ve already established that we can’t see this as $2M or $3M ARR deal.  Not yet, at least.

This deal is a layer cake:  it’s a three-year $1M ARR deal [12] that has a one-year-delayed, two-year $1M ARR deal layered atop it, and a two-year-delayed, one-year $1M ARR deal atop that.  And that, in my opinion, is how you should pay it out [13].  Think:  “hey, if you wanted to get paid on a three-year $3M ARR deal, then you should have brought me one of those [14].”

Finally, what to do about the CAC?  One might argue that the full cost of sale for the eventual $3M in ARR was born up-front.  Another might argue that, no, plenty of account management will be required to ensure we actually get the pre-sold upsell.  The easiest and most consistent thing to do is to treat the ARR as we mentioned (1+1+1) and calculate the CAC, as you normally would, using the ARR that we put in the pool.

If you do a lot of these deals, then you would see a high new-business CAC ratio that is easily explained by stellar net-dollar expansion rates (173% if these were all you did).  Think:  “yes, we spend a lot up-front to get a customer, but after we hook them, they triple by year three.”

Personally, I think any investor would quickly understand (and fall in love with) those numbers.  If you disagree, then you could always calculate some supplemental CAC ratio designed to better amortize the cost of sale across the total ARR [14].  Since you can’t have your cake and eat it too, this will make the initial CAC look better but your upsell CAC and net-dollar expansion rates worse.

As always, I think the right answer is to stick with the fundamental metrics and let them tell the story, rather than invent new metrics or worse yet, new definitions for standard metrics, which can sow the seeds of complexity and potential distrust.

# # #

Notes

For more information on ASC 606 adoption, I suggest this podcast and this web page which outlines the five core principles.

[1] I am not an accountant.  I’m a former CEO and strategic marketer who’s pretty good at finance.

[2] And which I like better as “unbilled deferred revenue” is somewhat oxymoronical to me.  (Deferred revenue is revenue that you’ve billed, but you have not yet earned.)

[3] I know in some cases, e.g., prepaid, flat multi-year deals, ASC 606 can actually decide there is a material financing event and kind of separate that from the core deal.  While pure in spirit, it strikes me as complex and the last time I looked closely at it, it actually inflated revenue as opposed to deflating it.

[4] Which I define as all the future revenue over time if every contract played out until its end.

[5] Ergo, you have high empirical confidence that you are going to get all the revenue in the contract over time.

[6] Good comp plans pay only a portion of large commissions on receipt of the order and defer the balance until the collection of cash.  If you call this a $2M ARR deal, you do the comp math as if it’s $2M, but pay out the cash as dictated by the terms in your comp plan.  (That is, make it equivalent to a $2M ARR deal with crazy-delayed payment terms.)  You also retire $2M of quota, in terms of triggering accelerators and qualifying for club.

[6A] This then begs the question of how to comp the $1M in pre-sold upsell in Year 3.  As with any of the cases of pre-sold upsell in this post, my inclination is to pay the rep on it when we get the cash but not on the terms/rates of the Year 1 comp plan, but to “build it in” into their comp plan in year 3, either directly into the structure (which I don’t like because I want reps primarily focused on new ARR) or as a bonus on top of a normal OTE.  You get a reward for pre-sold upsell, but you need to stay here to get it and you don’t year 1 comp plan rates.

[7] That is, if all your contracts are signed on the last day of the quarter, and you don’t sign any new ones, or churn any existing ones until the last day of the quarter, and no one does a mid-quarter expansion, and you don’t have to worry about any effects due to delayed start dates, then the ARR balance on the last day of the quarter / 4 = next quarter’s subscription revenue.

[8] Development is not “over” and that value released – assume they continue to fully exploit all the development environments as they continue to build out their app.

[9] Note that TCV can be seen as an “evil” metric in SaaS and rightfully so when you try to pretend that TCV is ARR (e.g., calling a three-year $100K deal “a $300K deal,” kind of implying the $300K is ARR when it’s not).  In this usage, where you’re trying to express total commitment made to the company to emphasize the importance of the customer, I think it’s fine to talk about TCV – particularly because it also indirectly highlights the built-in upsell yet to come.

[10] Or perhaps some intelligent mix thereof.  In this case, I’d want to weight towards TCV because if they are not successful in year 1, then I fail to collect 5/6th of the deal.  While I’d never tell an investor this was a $6M ARR deal (because it’s not true), I’d happily tell my Customer Success team that this a $6M TCV customer who we better take care of.  (And yes, you should probably give equal care to a $2M ARR customer who buys on one-year contracts – in reality, either way, they’d both end up “Tier 1” and that should be all that matters.)

[11] Or you could of the ASC 606 revenue backlog and/or Remaining Performance Obligation (RPO) – and frankly, I’d have trouble distinguishing between the two at this point.  I think RPO includes deferred revenue whereas ASC 606 revenue backlog doesn’t.

[12] In the event your compensation plan offers a kicker for multi-year contracts.

[13] And while you should factor in the pre-committed upsell in setting the reps targets in years 2 and 3, you shouldn’t go so far as to give them a normal upsell target with the committed upsell atop it.  There is surely middle ground to be had.  My inclination is to give the rep a “normal” comp plan and build in collecting the $1M as a bonus on top — but, not of course at regular new ARR rates.  The alternative is to build (all or some of) it into the quota which will possibly demotivate the rep by raising targets and reducing rates, especially if you just pile $1M on top of a $1M quota.

[14] This ain’t one – e.g., it has $6M of TCV as opposed to $9M.

Leave a comment

Posted in Finance, Metrics, SaaS, Sales Model, Startups

Tagged

Important Subtleties in Calculating Quarterly, Annual, and ATR-based Churn Rates

Posted on June 27, 2018 | 3 comments

This post won’t save your life, or your company.  But it might save you a few precious hours at 2:00 AM if you’re working on your company’s SaaS metrics and can’t foot your quarterly and annual churn rates while preparing a board or investor deck.

The generic issue is a lot of SaaS metrics gurus define metrics in a generic way using “periods” without paying attention to some subtleties that can arise in calculating these metrics for a quarter vs. a year.  The specific issue is, if you do what many people do, that your quarterly and annual churn rates won’t foot — i.e., the sum of your quarterly churn rates won’t equal your annual churn rate.

Here’s an example to show why.

saas churn subtle

If I asked you to calculate the annual churn rate in the above example, virtually everyone would get it correct.  You’d look at the rightmost column, see that 2018 started with 10,000 in ARR, see that there were 1,250 dollars of churn on the year, divide 1,250 by 10,000 and get 12.5%.  Simple, huh?

However, if I hid the last column, and then asked you to calculate quarterly churn rates, you might come up with churn rate 1, thinking churn rate = period churn / starting period ARR.  You might then multiply by 4 to annualize the quarterly rates and make them more meaningful.  Then, if I asked you to add an annual column, you’d sum the quarterly (non-annualized) rates for the annual churn and either average the annualized quarterly rates or simply gray-out the box as I did because it’s redundant [1].

You’d then pause, swear, and double-check the sheet for errors because the sum of your quarterly rates (10.2%) doesn’t equal your annual rate (12.5%).

What’s going on?  The trap is thinking churn rate = period churn / starting period ARR.

That works in a world of one-year contracts when you look at churn on an annual basis (every contract in the starting ARR base of 10,000 faces renewal at some point during the year), but it breaks on a quarterly basis.  Why?  Because starting ARR is increasing every quarter due to new sales that aren’t in the renewal base for the year.  This depresses your churn rates relative to churn rate 2, which defines quarterly churn as churn in the quarter divided by starting-year ARR.  When you use churn rate 2, the sum of the quarterly rates equals the annual rate, so you can mail out that board deck and go back to bed [2].

Available to Renew (ATR-based) Churn Rates

While we’re warmed up, let’s have some more fun.  If you’ve worked in enterprise software for more than a year, you’ll know that the 10,000 dollars of starting ARR is most certainly not distributed evenly across quarters:  enterprise software sales are almost always backloaded, ergo enterprise software renewals follow the same pattern.

So if we want more accurate [3] quarterly churn rates, shouldn’t we do the extra work, figure out how much ARR we have available to renew (ATR) in each quarter, and then measure churn rates on an ATR basis?  Why not!

Let’s first look at an example, that shows available to renew (ATR) split in a realistic, backloaded way across quarters [4].

ATR churn 1

In some sense, ATR churn rates are cleaner because you’re making fewer implicit assumptions:  here’s what was up for renewal and here’s what we got (or lost).  While ATR rates get complicated fast in a world of multi-year deals, for today, we’ll stay in a world of purely one-year contracts.

Even in that world, however, a potential footing issue emerges.  If I calculate annual ATR churn by looking at annual churn vs. starting ARR, I get the correct answer of 12.5%.  However, if I try to average my quarterly rates, I get a different answer of 13.7%, which I put in red because it’s incorrect.

Quiz:  what’s going on?

Hint:  let me show the ATR distributed in a crazy way to demonstrate the problem more clearly.

atr churn 2

The issue is you can’t get the annual rate by averaging the quarterly ATR rates because the ATR is not evenly distributed.  By using the crazy distribution above, you can see this more clearly because the (unweighted) average of the four quarterly rates is 53.6%, pulled way up by the two quarters with 100% churn rates.  The correct way to foot this is to instead use a weighted average, weighting on an ATR basis.  When you do that (supporting calculations in grey), the average then foots to the correct annual number.

# # #

Notes:

[1] The sum of the quarterly rates (A, B, C, D) will always equal the average of the annualized quarterly rates because (4A+4B+4C+4D)/4 = A+B+C+D.

[2] I won’t go so far as to say that churn rate 1 is “incorrect” while churn rate 2 is “correct.”  Churn rate 1 is simple and gives you what you asked for “period churn / starting period ARR.”  (You just need to realize that the your quarterly rates will only sum to your annual rate if you have zero new sales and ergo you should calculate the annual rate off the yearly churn and starting ARR.)  Churn rate 2 is somewhat more complicated.  If you live in a world of purely one-year contracts, I’d recommend churn rate 2.  But in a world of mixed one- and multi-year contracts, then lots of contracts are in starting period ARR aren’t in the renewal base for the year, so why would I exclude only some of them (i.e,. those signed in the year) as opposed to others.

[3] Dividing by the whole ARR base basically assumes that the base renews evenly across quarters.  Showing churn rates based on available-to-renew (ATR) is more accurate but becomes complicated quickly in a world of mixed, multi-year contracts of different duration (where you will need to annualize the rates on multi-year contracts and then blend the average to get a single, meaningful, annualized rate).  In this post, we’ll assume a world of exclusively one-year contracts, which sidesteps that issue.

[4] ATR is normally backloaded because enterprise sales are normally backloaded.  Here the linearity is 15%, 17.5%, 25%, 42.5% or a 32.5/67.5 split across the first vs. second half of the year (which is pretty backloaded even for enterprise software).

[5] The spreadsheet I used is available here if you want to play with it.

3 Comments

Posted in Churn, Metrics, SaaS, Startups

Tagged