SaaS Metrics: What is the best tool for calculating LTV of a customer by marketing channel?

Wow, wow, wow... slow down, you're mixing too much stuff here. LTV in SaaS is a broad value, I don't see many interest to calculate it by channel. What results do you expect? After that, concerning the calculation itself, you can try software like Kissmetrics, but, honestly, LTV is too critical to rely on third party software and it's easy to calculate in Excel. At some point, analytics software always have hiccups (the guy that don't accept cookies for example) and will be biased. A rough spreadsheet from your rough database CSV export is the only judge in the end. That's probably why SaaS people do this that way :-) And one more time, what king of business critical information do you think you will find in LTV by channel?

Like Alain says, I think you're mixing up the terminology here and different metrics. I think you're looking for either a ROI per marketing source or LTV:CAC ratio for AdWords (and other marketing sources). A tracking pixel won't give you any information on the LTV. You need to track other metrics for that. It will only let you flag which leads arrived from AdWords. Actually, you don't even have to use a pixel for that. A nice AdWords and CRM system setup would do great. You can even track how specific adgroups do, and even individual ads! Note that the CAC (customer acquisition cost) is not only the direct marketing costs. You should add any costs related to the sales cycle.

Yes, there are easier and accurate solutions to get the reports you are looking after. We did such an implementation for a customer that also had attribution modeling added to it. A few things need to be taken care of: 1. Make sure you get all the correct revenue data. Refunds, chargebacks, upgrades, downgrades, all need to be included. On long terms, it makes a huge difference. You don't want to rely on JavaScript tracking for this one. Getting the data directly from the payment system is the most reliable way to do it. 2. Make sure you have an unique ID attributed to each user that does not depend on a cookie or external source. An email address can be good, though in B2B businesses, they are changed often. The user ID from your own system should do the job here 3. Own the channel sources tracking. There are open source scripts for this. Relying on other parties will not make it easy to retreive the source of a specific user. The library we use and recommend: https://github.com/innertrends/getSetReff 4. Have all the data in raw format so you can audit it and make sure nothing goes wrong in the calculation. Getting a number and not knowing for sure how it is calculated will just create frustrations and doubts. 5. Add the cost data for each channel. You might want to look into some attribution modelling. Relying only on last click can lead to wrong decisions. All this can be done in a automated setup which, in the end, just offers a page/sheet/dashboard that gives you the numbers.

I haven't necessarily seen an all-in-one tool that tracks every marketing channel you're using. This is usually because there are sooo many marketing channels available (if those that are new) that would almost be impossible and make tool too big to be effective. Here are close all-around tools that may help (some may need a bit of tweaking to be effective): http://Sumall.com http://Ducksboard.com Google Analytics http://Geckoboard.com Here are some tools that track separate marketing channels that integrate with other tools that allow you to pull all the data together: Track your lifecycle marketing ROI - http://Autosend.io Track your email marketing ROI - http://mailchimp.com/ Track your referral program ROI - http://ReferralSnip.com Track your social media ROI - http://Razorsocial.com Clearly, there are other tools out there. But I haven't found an all-in-one tool that covers everything as thoroughly as you'd probably need it to.

Excel is probably the right tool to carry out the actual calculation. There are some great free Excel tools out there, for example on http://forentrepereneurs.com. The key to success is finding a good source of data to calculate metrics from. I personally find that ChartMogul is the best for a SaaS business.

Tons of solution providers have sprung up as of late to tackle this much needed problem for subscription Saas businesses. http://chartmogul.com, http://baremetrics.com, http://firstofficer.io all do the job by connecting directly to your recurring billing platform and sourcing all the necessary variables from there. The demand for this need has finally been met.

Being able to calculate LTV per marketing channel is such a compelling use case when evaluating the performance of your marketing and your ROI. You can very quickly get an idea of which channels are delivering the truly high-value customers.We built a set of features into ChartMogul, with the goal of doing exactly this (and a number of other key use cases). Here’s how it looks: And here’s how you go about getting this:Grab a free trial of ChartMogulThis step’s easy. No credit card ;)→ http://ChartMogul.comPush marketing channel data to ChartMogul’s SegmentationDepending on where the marketing channel data exists, you can push customer metadata to ChartMogul with one of the built-in https://chartmogul.com/integrations/, or through the https://dev.chartmogul.com/docs/introduction-enrichment-api.Add a custom segment, filtering for the Marketing Channel data you’re sendingOnce you’re sending the data to ChartMogul, it’s just a case of creating a segment in ChartMogul to match the metadata you’re sending. Et voilà: If you want some more information about LTV, how best to calculate it and some of the trickier things to consider, we’ve also just launched a cheat sheet dedicated to the topic: You can grab it for free here: https://blog.chartmogul.com/ultimate-saas-ltv-cheat-sheet/

First of all, LTV doesn't differ by the marketing channel. So there's no reason to calculate your LTV by marketing channel.LTV is a projected number, and not something like MRR or churn, which are calculated based on available values.LTV is more of a forecast. You're forecasting that somebody will be with you for so many months, based on a history of data.One school of thought is using this formula for LTV in months:1/Churn RateAnd this formula for LTV in dollars:(ARPU x Gross Margin%)/Customer Churn RateThe problem with this approach is that when the churn varies each month (say a 2% churn this month and a 3% churn the next), your LTV will also vary month-on-month.It is the LTV based on which you expect to be making say $3000 per customer. If you're going to change your forecast to $2000 the next month, how will you decide how much to spend to acquire the customer?The best practice that we have found is to take an average of the last three months of churn to calculate the LTV. Here, there is at least some level of stability when it comes to the projection.Here’s a screenshot of our calculations (with the formulae): If you want to go to a more granular level by segmenting the LTV by channel, you can still do that. We haven't come across any specific tool that can help you in doing the same, though. However, whatever I've explained above can be pulled off with a simple Excel sheet.Hope this helps!

LTV is calculated as the total revenue minus cost over the lifetime of the customer (typically this is the average lifetime of your user base, although not knowing this...2 years is a decent substitute, except mobile gaming which might be much shorter. ) Asking which tool to use is incomplete question. You need to use a system that records every financial transaction for you to sum as well as keeps track of your average cost to serve each user (many times that is found in a spreadsheet, hence why so many LTV calculations happen in Excel.)  Most are not storing the dollar transactions in Google Analytics or KISSmetrics. This is better done in a database where you can more easily save, transfer,  or combine the data at user level. Sum the revenue for a cohort of users with two years lifetime with your service, divide by the number of users in the cohort, subtract the per user cost to serve those users over that time. If you decide to go a KISSmetrics route, just create a calculation where you must hard code the cost to serve. Splitting by first touch attribution channel is an excellent way to see whether you can pay more to acquire users from a particular channel.