Summary: Marketing automation and CRM systems may eventually converge, but for now marketers need help explaining why they need a system of their own.
I hugely enjoyed yesterday’s Boston session of the Silverpop-sponsored B2B Marketing University. (You can catch another session in Atlanta next week and in Seattle on December 1.) I won’t try to recap four hours of insights from Adam Needles from Silverpop, Carlos Hidalgo of Annuitas Group and Joe Moloney of Conselltants (no Web site, it seems), as well as Yours Truly. But there were a couple of topics that caught my fancy:
1. People still don’t understand the difference between marketing automation vs. CRM.
I really thought the distinction was pretty clear by now, but the question came up more than once. My own answer boiled down to a perhaps-not-convincing “trust me, they’re really different”, although I’ve addressed the question in depth in the resources section of the Raab Guide Web site.
Joe Moloney gave a more detailed answer about limits in Salesforce.com in particular, including lack of CAN-SPAM compliance and limits on mass emails. Someone (I think it was end-of-day panelist Meg Heuer of Sirius Decisions) also pointed out that CRM data is often very dirty, which isn't a problem for salespeople working with one record at a time, but making it hard to use for marketing.
The immediate take-away here is that the industry still needs to educate prospective buyers on why marketers need a separate system. Vendors take note.
2. Will Marketing automation and CRM remain separate?
The discussion also segued into whether marketing automation and CRM will merge in the long run. I still suspect they will, driven by the need for ever-closer cooperation between marketing and sales teams in managing prospect relationships. But the other presenters disagreed, largely arguing that the separate groups have distinct needs. (See Who’s Afraid of the Big, Bad Wolf? Is Salesforce.com a threat to vendors of marketing automation solutions? by Market2Lead CMO Kevin Joyce for a good statement of the separatist position.)
Part of the reason I expect convergence to happen is that it’s already taking place. (The past is so much easier to predict than the future.) The movement is coming mostly from the marketing automation side, presumably because there is more money to gain by moving into sales from marketing systems than vice versa:
- marketing automation systems for small businesses (Infusionsoft, Office Autopilot, Net-Results, etc.) typically include a CRM option for clients who don’t want to pay for a separate Salesforce.com or other license.
- firms aimed at larger installations (Marketo, Eloqua, Pardot, Genius.com, Active Conversion) are providing widgets that give sales people direct access to marketing automation information.
3. Technology may impede Software-as-a-Service sales automation vendors from adding marketing automation.
As Joe Moloney was listing the limits that Salesforce.com places on mass access to client data, I recalled that these are in place fundamentally to avoid large analytical queries that could slow down response for all other users of the shared systems. This isn’t an inherent problem with Software-as-a-Service itself: remember, the B2B marketing automation vendors themselves all operate on a SaaS model, and there is a growing number of SaaS business intelligence systems too.
But even though modern database technology allows one system to handle both CRM transactions and analytical marketing queries, this does take an appropriate design. I strongly suspect that existing SaaS CRM vendors like Salesforce.com would need to fundamentally rearchitect their systems to support serious marketing automation processing, especially for clients with millions of contact records. This may impede them from adding marketing automation capabilities, although newer SaaS CRM systems could emerge that are designed from the start to do both.
From this perspective, another reason combined marketing automation/CRM systems are first being offered to small companies may be that it’s easier to provide good performance for both applications when volumes are small.
Yesterday also triggered another set of thoughts regarding the importance of marketing content. But since one of these was the need to keep materials short, I’ll put them into a separate post.
David,
ReplyDeleteI enjoyed your talk at the B2B University in Boston. I'm glad you wrote this post because you saved me from doing so (and clearly you know this subject far better than I do).
Another issue that was raised throughout the afternoon was addressing accounts vs individuals. I imagine that's one more reason that it would be hard to integrate CRM and marketing automation systems. Yet obviously B2B marketers are struggling to reconcile the need/desire to adopt marketing automation (for that 1:1 "dialogue") with the need to manage and address a unit of buyers within a prospect organization. Unfortunately I had to leave before the panel discussion. Did this topic come up? If not, what are your thoughts on it?
Best,
Stephanie Tilton
Hi Stephanie,
ReplyDeletePleased to meet you in Boston. I agree that account vs individual is one of the toughest issues in integrating marketing automation with CRM.
At the technical level, some (not all) marketing automation systems have account tables, account-level lead scores, and revenue attribution logic that finds activity for all leads within an account. The key question to ask vendors is whether changes to account data are automatically applied to all leads associated with that account, or are only applied to the account data for that one lead.
Most marketing automation systems can do selections based on account attributes, which is great so long as the account data is consistent across all associated leads (per my previous comment). You also want campaign logic that can works at the account level (e.g., top promoting to all leads in accounts that are assigned to sales). That is harder to find, since campaigns usually process one lead at a time.
And while we're at it, shouldn't all this really happen at the opportunity level (leads associated with an opportunity) rather than all leads associated with an account? If not, what happens in a big account with many opportunities?
Another issue that didn't come up in Boston but does apply in the real world is that CRM systems often allow multiple leads with the same email address, because in fact salespeople want it that way (several people at a company might share a departmental email or use a shared email to identify a project team). Neary all marketing automation systems dedupe based on email address, so they can't handle this.
But even if the marketing automation system has the right capabilities, the really big issue is synchronizing with salespeople's data in CRM. Most companies let the salesperson decide which person is associated with which account, but you also want the marketing automation system to at least take its own guess before leads are assigned to sales or if the salesperson hasn't made the determination. Generally the marketing automation systems have rules in the synchronization logic that let salespeople override the marketing automation system's linkages but not vice versa. Yet even this can cause problems.
In addition, when you start assigning credit for revenue, the salespeople often don't link an opportunity to a lead that was generated by marketing. This could be for perfectly valid reasons: the initial lead might have come from an assistant who was doing research, but the salesperson might link the opportunity to the department head.
At a practical level, you want the marketing automation system to be "aware" of account and opportunity level linkages, so it can provide appropriate treatments. This is the key to gaining sales productivity by empowering the marketing automation system to handle routine interactions that the saleperson would otherwise have to do personally. But that will only be possible if the marketing automation system can "see" the data in the same way that the saleperson sees it. To me, this argues strongly for having both work on the same database, even if there is also a separate "marketing" view of the same database that's organized in ways that make more sense from a marketing perspective.
Hi David, thank you for your thoughtful post. It is true that technology limitations might limit the native functionality offered by CRM applications. But this technical hurdle will likely be overcome in the near future by new applications that effectively leverage the power of cloud computing. With storage and processing prices declining and existing APIs getting more robust, it is now possible to deliver powerful analytic applications with seamless integration and robust scalability are just around the corner.
ReplyDeleteHi Manuel,
ReplyDeleteI think the capabilities of cloud-based analytics are well proven. But the issue isn't whether the analytical systems can handle the CRM data; it's whether the CRM systems can make the data available. Today's SaaS-based CRM products (and, for that matter, SaaS-based marketing automation products) limit direct queries against their databases because it could harm performance. For example, I think the Salesforce.com API limits query results to something like 1,000 rows (don't quote me on that number). This is why the CRM - marketing automation synchronization is typically done in batch, not real time. I suspect that the technical changes needed to allow direct analytical queries against the SaaS CRM systems are the same changes needed to allow those systems to execute analytical queries internally. Thus, if you're going to solve the problem at all, you might as well add the analytical (i.e., marketing automation) capabilities to the CRM system itself.
(Background for non-geeks: marketing automation systems do "analytical queries", which scan the entire prospect/customer database to find records that meet specified characteristics. This is how you'd select an email list, for example. The opposite are "transactional queries", which go directly to specific records based on identifiers such as account number or name. Databases built for transaction processing, like CRM systems, typically don't do well with analytical queries.)
Hi David
ReplyDeletePardon my overwhelmingly self-interested perspective on this, but application designers have long since solved the problem of crm cohabitating with marketing.
Rather than being a question of technology (terms like SAAS only obfuscate here), the issue is a business one - the velocity of change.
The business processes of marketing and CRM have been changing so fast it has been difficult for solution providers with genuinely integrated systems to simultaneously evolve both fast enough.
Nevertheless, the integrated solutions are catching up quickly to the one-offs, and the advantages of that integration are now overwhelming.
A similar dynamic took place with the integration of order management and accounting systems throughout the 1980s and 1990s.
David, CRM systems can make the data available. The salesforce bulk API is designed to support millions of records. You start with a historical load and then use a change data capture framework to synchronize incrementally as often as near real time (if needed).
ReplyDeleteIt looks like the industry is leaning towards open platforms that can support a large partner ecosystem (i.e. there is an app for that). In this setting it seems better to invest developing a platform that can support multiple solutions instead of building the solution yourself. (even if the effort might be equivalent)
This makes sense as applications become more sophisticated and analytics go beyond data extraction/filtering to predictive modeling and optimization.
Hi David,
ReplyDeleteFirst, thanks for the post. We've always maintained that CRMs will be separate from MA for the foreseeable future. CRMs just do not have the focus on MA that MA requires.
Although we appreciate the mention, I'd like to clarify that we cater to SMBs, which typically means companies with less than 1000 employees, not small businesses per se. As a result, we do not compete against Infusionsoft et al, and moreso against Pardot, Genius. Also, although we have Sales interface into ActiveConversion, it is not a CRM and that we use Salesforce.com integration for those cases.
Thanks Fred (that's Fred Yee, president of Active Conversion). Per his comment, I edited the post to put him in the same category as Pardot, Genius, etc.
ReplyDeleteDavid,
ReplyDeleteThanks for the post. It's an interesting debate, but I think that there are a couple of other strong reasons for the "separatist" position, as you call it. I agree with your point on the technology design - it is a very different design goal to process the style of computing you see in marketing (lots of automation, analytical queries, bulk transactions like email, high traffic from websites, etc). Not impossible, but would require a differently architected system to do both.
However, pricing is another interesting hurdle. Sales-focused CRM systems have a natural value point per seat. More salespeople, more value, and the price changes accordingly. Marketing does not grow in value like that as the marketing teams for even huge organizations are generally quite small. It seems like a minor point, but it is hard to mesh the two pricing structures in a clean way.
Additionally, and perhaps most importantly, is political reasons. As long as sales and marketing are different departments (and you and I both see these merging *eventually*, but not for a while), the idea that they head of marketing will have to ask the head of sales to have his team add a field or change a rule to allow for the next campaign will not be popular.
Keeping the systems tightly linked, but operationally independant, allows marketing to build out its business processes while sales builds out its own. I wrote a bit about the marketing automation/CRM integration stack and also what to think about in terms of technology and business process considerations when looking at the integration if you're interested.
Steve, thanks for your thoughtful comment. Political realities may well prevent rational solutions (see "health care reform"). But I'm enough of a believer in the free market to trust that companies willing to adopt a more effective, integrated solution will eventually have a competitive advantage that forces everyone else to follow. Of course, if an integrated solution really doesn't provide a net advantage, then I'm wrong and separate-but-synchronized systems will continue to exist.
ReplyDeleteMarketing automation and CRm will integrate even more.
ReplyDeleteStill the most important to integrate with the CRM is your website as all customers and leads will visit your website before, during and after their buying decision.
Look into LEADSExplorer providing lead generation, monitoring of nurturing and customer retention. http://bit.ly/2KxbKz
Great discussion! Coming from Chicago, I appreciated the posts that recognize the political elements of this controversy, i.e., marketing vs. sales. I do believe we've seen a trend towards marketing owning more of the customer/prospect relationship, especially in the lead development stage, hence the profusion of marketing automation offerings.
ReplyDeleteOf course, the migration of customer relationship ownership isn't stopping at marketing; with the proliferation of social media (even in the B2B world) ownership is moving towards the customer/prospect and the community, and hence we're starting to see a proliferation of Social CRM tools, whether its extensions from Oracle, SF.com, etc. or new tools like Lithium.
EM (can I call you 'E'?) -- I certainly agree. The whole dimension of social community "ownership" of the relationship, existing outside the company itself, is one I've been meaning to write about but just can't find the time. Somewhere there's a notion that companies really will need to manage relationships with communities rather than individuals. We're just beginning to understand how that works from a marketing perspective.
ReplyDelete