Three months ago became the newest customer of Infusionsoft. I’ve learned a ton about this product during this time and wanted to offer a brain dump of thoughts. These are directed both to potential users of the software as well as Infusionsoft the company.

What it is: CRM + automation fabric

Describing this product concisely is difficult because it’s more of a “fabric” than an application geared towards a specific use case. As far as I know it’s the first system to bring the capabilities of something like Eloqua and Marketo within reach of the small business, only better because it combines CRM. Think of it like a pliable Silly Putty for marketing & CRM automation that allows you to set your marketing material up like one of those “Choose your own Adventure” books from back in the day. It’s tempting to call this system “a nerve center of customer communications for a SMB” but that too would be pigeon-holing it to customer communications when it can be equally valuable for vendors, partners & resellers. I like the “moldable marketing fabric” analogy because it captures the right essence of the flexibility of it. My friend Eric who works there calls it “the OS of small business” and I think that’s an interesting and valid way to put it.

Another way to think of it is via the phrase coined by one of their star customers, Jermaine Griggs. He did a stellar webinar about a week ago that is a must-watch for people getting started with Infusionsoft. He cites the main benefit as enabling the “scaling of personal attention” – basically gaining the ability to render customer experiences you’d expect of a small mom & pop shop only on web scale supporting hundreds of thousands of customers with the same level of intimacy. This guy has a truly incredible story having come up from nothing – he’s worth listening to just for his humility alone but he also has a brilliant mindset of how he approaches marketing and relationship building online.

How this can benefit SMB’s

Again it’s tough to reduce the benefit to a sound byte here because it really depends on how one uses the fabric. What would you do if you could imbue your selling knowledge into your web app so it intelligently did things the way you would do if you were standing there in person talking with the prospective customers? How might that change both their experience as well as your insight into what they want/need? I find it find Dave McClure’s AARRR pirate metrics framework useful when examining online businesses – Infusionsoft can basically help in 4/5 phases (activation, retention, revenue and referral).

We “hired” Infusionsoft for two main reasons: 1) on the back end there’s simply no way for us as two people to stay on top of all the deals handled by our network. We had tried using Google Docs, Longjump, a custom WordPress app and Basecamp and while those each sufficed on the CRM front, none gave us a way to intervene and actively and intelligently babysit transactions. 2) On the front end our leadgen apparatus was very static. We had a single linear funnel for on-ramping new customers. While Mailchimp auto responders are better than nothing, you’re still highly constrained by how adaptive you can make your marketing and it’s not unified with CRM so there’s no straightforward way to pull up contacts and get a clear picture where everyone stands.

What it’s lacking

Infusionsoft has been great at addressing these issues so far but I feel like maybe I’m fawning over this product a bit much; it’s definitely not perfect. The difficulty of setup is its Achilles’ heel right now impeding mainstream usage. It took us roughly two months to implement and we are moderately-technical and highly motivated. While they’ve made strides with their new visual campaign builder, it’s still missing the mark in a couple fundamental ways. There are IMO two separate shortcomings with their current onboarding process:

1. They ask customers to make the leap to wiring up a solution before mapping the current landscape. The campaign builder put some sanity to an extremely unintuitive implementation process that involved jumping amongst many screens to wire up templates, action sets, rules, web forms and follow-up sequences. But it doesn’t solve the intermediary step needed wherein the customer visually maps out their existing processes, identifies the opportunities to fix the funnel and thereby establishes the plan. For those with a background in software development, think of it like parachuting into a programming scenario and trying to leap to writing code to solve a problem before you’ve mapped out the current process and hashed out the user stories for what you’re trying to build.
2. I think even if they were to add this intermediary step, it’s still going to a far stretch to ask the average SMB owner to complete this process on his/her own. There’s real artistry and skill involved in identifying the biggest opportunities and then taking the knowledge of the software and translating the business objectives into implementation. I believe their model needs to shift from a DIY solution for the end consumer to incorporate an agent as the intermediary. These “marketing architects” would have a thorough understanding of both online marketing best practices as well as detailed understanding of Infusionsoft and be bridge builders to close the current gap. In fairness, they do pair you up with a “success coach” employee who helps get you through the initial onboarding process (big shout out to ours, Brett was instrumental in our implementation). But IMO this is a band-aid fix to symptoms which would go away if the consultant ecosystem were made to be a more integral role.

It’s impressive how many people they have using the system in spite of these impediments. It’s a testament to how useful it is that folks want it badly enough to figure out how to bushwhack their way through it all to get it working.

What I would do if I were Infusionsoft

So I hate criticizing things and not offering solution suggestions. Here’s what I would be doing if I were Infusionsoft to remedy these issues:

1. Create an “assembly exchange” They’ve already got what appears to be the equivalent of an AppExchange for vendors to market plugins to users but what I propose is actually making it so I as a user can turn my configured instance into a template I can sell other people. Think the way AWS allows EC2 users to convert a running instance into an AMI that can be made available to others – it opens up a whole new ecosystem. A bicycle store owner in Tampa who has developed a successful marketing framework in Infusonsoft should be able to productize his/her instance and sell it to bicycle store owners elsewhere. This would do some miraculous things, namely a) create a new revenue stream for that Tampa bicycle store owner without cannibalizing his/her local business b) help the new Infusionsoft user expedite her bicycle store marketing implementation and get a proven system unique to her vertical c) reduce customer attrition for Infusionsoft by reducing frustration and failed on boarding d) potentially spawn a whole new class of companies like how WPMU and Thesis grew around WordPress.
2. Disentangle planning from implementation. Right now the “architecture” and “brick laying” are intermingled when setting up a new instance. It would be beneficial to have a separation that allows one to articulate the current business processes, then map out the ideal new flow after Infusionsoft. This document could then be handed off to a commodity implementation resource via oDesk or Elance and the tedious grunt work of implementing stuff could be offloaded. I haven’t formed a strong picture of how this looks but my hunch is they don’t need to reinvent the wheel here – this format already exists either as a specialized sequence diagram in UML or via BPM with an open source tool like Intalio. The sharp consultants are probably using a homegrown tool of their own to achieve this step – I would start by asking them how they’re doing it.
3. Consultant ecosystem of brick layers v. architects I realize there’s already a consultant community for Infusionsoft but as best I can tell there’s no commonly accepted format for mapping out the vision with one “marketing architect” and then taking portable definition to a more commodity “infusionsoft implementor.” In my mind that first piece is what they should give away free: a way for any visitor to come and sketch out their funnel and current vs. proposed marketing interactions. Then simply charge for translating the proposed version into reality. Once the architecture can be partitioned from the implementation it opens it up to be able to commoditize the brick laying aspect and drives the architectural role to be more about formulating valuable constructs for the funnel rather than worrying about nuts & bolts of implementation.
4. Sandboxed dev v. staging v. prod environments I would love to outsource some of the grunt work I’m doing now but I’m also very hesitant to give access to our system over to a stranger in the Philippines. The way we do this with our site now is by cloning our WordPress instance with BackupBuddy and spinning up a duplicate instance on EC2 and giving the contractor access to that system. They can break it all they want and there’s no damage. What’s better, I can spin up 5 instances and compare 5 different contractors apples-to-apples in parallel for two hours of trial work to see who gets the furthest and does the best work. It would be great if there were a way to clone our Infusionsoft environment and run sandboxed versions that could be promoted to staging and ultimately production. If the above separation of “architects vs. brick layers” is to be achieved this will be an important capability.
5. An offline IDE This is perhaps too demanding to ask for this now but I will anyways- it’d be great to have a non-web-based way of building stuff. In the same way that Quickbooks online is cumbersome to accountants who are used to flying through screens on the desktop product, when you’re building a bunch of sequences and emails and whatnot, the latency of the web even on a fast connection is annoying. I totally get why it’s all web-based now (I would do the same) but perhaps there’s opportunity to do an Eclipse-based IDE that would generate a file which could be uploaded to Infusionsoft and provide an easier way for building a system?

Other random thoughts

So this is my random catch all for other things I wanted to mention that would be useful:

• Bizspark equivalent the $2-3k initial startup cost is going to be preclusive for most startups even though they stand to benefit dramatically from this software. If Infusionsoft can stomach giving away some super-cheap accounts they should strive to offer a no-support, self-serve, feature-dilluted option for startups and take a longer-arc view harboring people who can’t afford it today but represent prospective future customers. Those folks are also typically more vocal startuppy types so there is likely some marketing benefit to serving them even though they’re free accounts today.
• The gmail core plugin rocks This is missing functionality in the core Infusionsoft app but I’m happy to see that someone nailed the plugin that provides it. This plugin makes it so Infusionsoft will automatically monitor your Gmail and append communications that happen there to their corresponding contacts in IS. The benefit is multiple people can be working how they normally work and a centralized record of communications is retained in Infusionsoft for all to see. I would love to see the option in Gmail Automation Core to do attachments via filebox as well as the ability to do a one-time retroactive pull of past communications from a gmail account.
• Global & shared dashboards There’s no way I can see for me to create a dashboard which gets shared amongst other members of my team. It’d be neat if they made it possible to have tabbed dashboard pages and possible for one to mark his/her page as “shared” so others could use it.
• Past emails should be snapshotted by value rather than by reference There’s a glaring issue IMO right now with how past automated emails are preserved: currently it stores the variable names instead of the actual communication to the client. This needs to be a snapshot of what exactly was sent at that point in time. Think of changing prices, new sales reps, etc. There’s no good reason to store it by reference with variable names- it really needs to be snapshotted.
• Tags and follow-ups listed in search results I would imagine they’ll add this soon but it’d be useful to be able to see sortable search results with the tags appearing in columns separated by tag category. Likewise it’d be nice to scan a bunch of results and see what follow-up sequences they’re in.
• Calculated fields This is the one thing I really miss from Longjump: the ability to have fields on the contact record that are derived from other fields. Consider adding the ability to make a field that is simply a formula builder consisting of other fields on the record.
• Lead origin tracking We’re likely going to do a bit of a tap dance on our site to get what I want in terms of tracking all the data on how customers originated (ie. which ad channel, campaign version, landing page, keywords, etc). This is fundamental and important enough that I believe this belongs in the core product. For now we’re planning to roll a cookie with all this info and then pass it via hidden fields on form submissions.
• Snooze button on follow-up sequences It’d sure be nice to be able to click on the date of a follow-up sequence and just override it for that contact record. We end up doing a hoaky process now whereby we remove the sequence, create a scheduled task to remind ourselves to start that sequence over again at a later date. It’d be much preferable to be able to have a “snooze button” that would let you manually override the date of the follow-up sequence and push the whole thing out X days.
• File-type form field to upload attachments The form builder tool is very good but it’s missing an obvious field type: there’s on ability to do file attachments. This one shouldn’t be terribly difficult to implement but there should be a new type of form field for attachments and they should be sent to the filebox for that contact.
• No split testing I’ve seen the recommendations for how to implement split testing in Infusionsoft and unfortunately it’s pretty ghetto. I understand people are working on IS plugins that will lend this functionality but A/B testing is so integral to online marketing now that I would argue it belongs in the core product. If this is too peripheral to be in the core product today I would suggest making a one-click option for integrating something like Optimizely or Visual Website Optimizer. We’ve used Optimizely for awhile now and it’s excellent.
• No funnel visualization There’s no way I’ve seen to visualize the various funnel interactions and run cohort analysis. Likewise this is pretty core to online marketing and if it’s not something they want to bake into the product itself I’d lobby for making it easy to enable KissMetrics or MixPanel by just entering in your acct ID for those products. I’ve used KissMetrics for about two years and it’s great for determining which section of your funnel demands attention.

Anyways, in summary, Infusionsoft is a really promising product and I’m excited to finally have our v.1 implementation in place. I’ll report back here on what I learn as we go. It’s great to see a local AZ company delivering a service that’s represents so much promise and has customers around the world. I look forward to seeing it evolve. If you’re interested in potentially using Infusionsoft for your business request a demo or check out some of their past client case studies.