National MLS, Whatever That Is

Matt Cohen“Should there be a National MLS?” is a question that is often repeated in the real estate industry—and it never should be asked again in that manner. As my funny Facebook friend Nobu Hata said the other day, whenever that question is asked, a puppy dies. The problem is that the question is vague and can mean three different things, and if we don’t decide which one we are talking about, our conversations will not be productive. Following are three ways to interpret the question:

  1. Should there be one national MLS system (software)?
  2. Should there be one national MLS database?
  3. Should there be one national MLS organization?

Although the first two questions may seem less central than the third, questions about the risks and benefits of a national organization depend, at least in part, on understanding issues relating to possible consolidation of the core software and database.  For that reason, I will address the more detailed questions before examining the more general question.

1.       Should there be one national MLS system (software)?

 There could be one or many local MLS organizations and one or many databases, but one national MLS software package.

Having one national MLS software package is usually compared only with the status quo: having one (or occasionally, two) MLS software packages for the local market. But let us also compare it against a range of possible options, including one where individual subscribers select their own MLS software inside each MLS organization, as well as hybrid approaches.

Having a single local system provides the local MLS organization the buying power needed to make extensive local software customization affordable, splitting the costs over many users over time. Also, while an MLS organization can advocate with a single MLS system vendor for local needs and changes, it may be less feasible to advocate with many MLS system vendors. If the individual users contracted directly with MLS system vendors, it would be more difficult for the MLS to advocate with many MLS system vendors if there weren’t a direct financial and licensing connection as exists between the local MLS organization and the MLS vendor. How would changes in the many systems (and accompanying training) be communicated to subscribers? In addition, as multiple systems change (at once, or over time) somehow the underlying local database changes would need to be managed. That could get expensive for the local MLS organization, especially if many MLS software vendors were constantly innovating. Providing support for mission-critical software—especially as it changes—is a complex process, and while it’s difficult enough to support one MLS software package at the local level, supporting software from multiple vendors could be untenable. Such concerns argue that limiting the core MLS software packages to only one or two is the better approach.

But individual subscriber choice also presents its own advantages. Anyone who has ever helped select (or support) MLS software can tell you that it is difficult for a single MLS software interface to satisfy everyone in one local market. Some users prefer a very simple MLS system, others a more functionally robust system. Some users prefer the CMA of this system, others like the client collaboration tools in that system. Also, since people don’t need local support for popular business applications like Microsoft Word, why would they need it for the MLS software? Finally, since technology may make local customization easier in the future, the “buying power” argument may lose ground to the individual choice option.

However, recently MLSs have begun fielding a single core MLS system but providing an “app store” where individuals can differentiate around that core by purchasing the apps and tools they want—like property reports, CRM, CMA, mobile apps and marketing software—and this new practice may reduce the attractiveness of fielding multiple core MLS systems. This approach could help enable a single national “core” MLS system and a standard national interface to suffice for many users while simultaneously making additional innovative add-ons and interfaces available for agents or brokerages to purchase.  The suppliers of these add-ons could compete for users at the national, state, and local market levels.

Historically, in comparing the advantages of a national system with those of a single local or individually-purchased system, the biggest question has been whether a national system could be customized to handle all of the local requirements and business rules of all the local MLSs. Any vendor that has tried to service 100 or 200 or more MLS organizations can already tell you how difficult it is to balance those customers’ needs and priorities, and serving the needs of over 850 would be far more difficult. With a national system, local customers waiting for a vendor to service the local requirements might experience a lot of frustration..

Finally, would a national MLS system best serve subscribers? Would competition among multiple MLS software companies spur innovation and lower cost, and how would these concerns balance against a national system’s potential efficiency? Currently, the competitive, innovative MLS software market serves local MLS organizations and their subscribers well. Could a system be put in place that would encourage competition for the national MLS system contract? It is unclear that a company would want to spend three to six million dollars to create a competitive MLS system just for the chance to win the business from an incumbent national system, or, if a new system won the business, whether the industry would want to risk a whole-country cutover to an unproven system. But without this competition, would subscribers be well served?

To summarize, it is difficult to argue the benefits of a single system, and easy to raise concerns that are hard to address. The current model of one system per MLS organization works.  But looking forward, if concerns could be addressed, it might be feasible to move toward a more consolidated “core” MLS system, mitigating some of the concerns about competition and innovation by allowing for individual subscriber-purchase of well-integrated ancillary software and user choice of “front-end” software that seamlessly works with the “core”.

2.       Should there be one national MLS database?

It is possible to have multiple MLS systems that have a single national database or “back end” in common, including listing content and even other core types of data found in today’s MLS systems, such as contacts, saved searches, financial worksheet data, and associated documents and other media.

Technically, a national database would consist of more than one physical set of database servers, in multiple locations, and might involve either of two processes: 1) the MLS system’s “front-end” interface uses the national database directly via an API or direct data access, or 2) the MLS system uses a replicated copy of the data or only part of the larger data set as needed by that front-end.

The main benefit of a national database would be that those requiring datasets that cross MLS boundaries could more easily get that data from one place (with local MLS organizational approval for specific local data, of course). Because the hardest aspects of data aggregation are obtaining and managing that permission as well as adequately addressing the related legal agreements and compliance, a unified database technology would not provide that much benefit over what we have now – even less once the RETS data dictionary is adopted. In addition, national aggregators already are addressing the technical aspects of aggregation and distribution.   Syndication companies like ListHub and Point2, and national efforts such as Corelogic’s Data Co-op, RED’s reDataVault, and RPR have each invested several million dollars to aggregate, cleanse, and enhance the local data on a national level, and they can each distribute the data based upon whatever directions and rules the content owners provide.

A national MLS database would certainly provide the benefit of making it easier to generate better real-time nationwide statistics. Would such statistics be much better than those today, based on data that are sent to NAR to generate national statistics that lag the market? Certainly, more accurate national real-time statistics and analytics would have immense value to investors, government organizations, and the lending industry, and this value would represent a substantial revenue stream to the source. This advantage of creating a national database is undeniable.

But, a national MLS database would provide limited benefit to the average real estate practitioner, licensed at the state level and doing business locally.  Although aggregating data in larger regional MLSs and having larger multi-MLS data aggregations would benefit large, multi-office regional brokerages, regional aggregation is a much easier endeavor than attempting to compile a national database. How much effort should the industry expend to create a national MLS database that does not provide much benefit to local practitioners, but mostly to national players?

Also, perhaps most importantly, there are three additional risks posed by a national database. First, it constitutes a greater information security risk: it’s a very big target. Also, if all systems depend on that database to be up and running, it also introduces a single point of failure. Finally, a national database might also constitute an easier legal target for those seeking access to the mother lode of listing data.

To summarize, although creating a national database would offer limited benefits, especially for the local participants who make up the bulk of the stakeholders, a national database would raise a number of concerns. It is unclear that the need for national real-time statistics and analytics, and the desires of those who might legitimately want nationwide listings data for other purposes, would outweigh those concerns. Creating a national database only starts to make sense, as a matter of efficiency, if there is a national MLS organization.

3.       Should there be one national MLS organization?

Most people advocating for national MLS are thinking about the advantages of a single system and database, and having a national MLS organization is seen only as an end to those objectives, since it’s seemingly impossible to get many MLS organizations to agree to a single system and database.

The central argument against the formation of a national MLS organization is that it essentially creates a national monopoly, and monopolies rarely provide the best products and services at the lowest possible cost over the long term.  Also, governance issues would likely make it more difficult for a national MLS organization to service local market needs. Consider the challenges MLSs currently face getting things done at the NAR Multiple Listing Issues and Policies Committee, a group that has only a very limited non-operational scope. Furthermore, like a national database, a national MLS organization also would create a single legal target for those seeking MLS access or recompense for patent or other infringements.

Many local MLS organizations are currently hubs of software and service innovation, both internally and working with small software vendors. They provide an opportunity to experiment and fine-tune offerings on a small stage, out of the national spotlight. Would one large MLS become an impediment to that environment – even putting up financial roadblocks to smaller vendors getting access to subscribers, as some of the larger industry organizations already do?

On the other hand, a national MLS organization could provide significant benefits.  First, there always seem to be well-funded efforts on the verge of disrupting the industry, and a single MLS organization would be better positioned, in terms of both governance and funding, to take on those challenges. In addition, the arguments against the formation of a national MLS could be addressed. In terms of product competition, for example, the organization could field multiple MLS system options or choices of front-ends, which would generate daily competition among technology providers, which would fight for market share among the subscribers and compete on price and quality. Further, although many large decisions would be made at the national level, there could still be local MLS service centers which would be expected to meet regularly measured standards of service, and potentially compete for subscribers on service and price as well. Although the governance and legal issues may be more difficult to address, those challenges and risks might be managed to some degree, as could the potential risks of a single national MLS organization.

Finally, some might argue that, if the present industry doesn’t create a national MLS (organization, system, and database) then another group might. If a well-funded company wanting first access to MLS data offered service to MLS subscribers nationwide for a large discount over what subscribers currently pay, could that company successfully host a national MLS?  Assuming the value was there, many practitioners would move to the lowest priced system, with little or no regard to who the MLS provider is, and obviously this could be very disruptive to the current structure of MLS.  If structured properly, and care is taken to address the issues raised above, a national MLS organization should be able to deliver more cost-effective MLS service than the current model and reduce the risk of this type of potential disruption.

There’s Middle Ground: MLS Regionalization

Having more than 850 Multiple Listing Service organizations, as we currently do, certainly has inherent problems. Many of the smaller MLS organizations don’t have the resources to provide a strong service or software package to subscribers, to defend against legal challenges, to implement a secure infrastructure, to plan for disaster recovery, or even to hire professional IT staff or management. Having more than one MLS in a property market area causes great inefficiency: practitioners must belong to multiple MLSs, which increases subscriber cost and data aggregation cost and effort. In addition, people need to learn multiple MLS software packages and manage multiple sets of login credentials—but still can’t perform a unified MLS prospecting search to provide search results to their clients.

The answer to these problems is not necessarily to create a national MLS organization, system, or database. A middle ground would be for the smaller MLSs to merge into regional MLSs, ranging in size from multi-state MLSs (for example, a New England MLS, Mid-Atlantic MLS, or Dakota MLS) down to some states which might have two or three (such as Northern California and Southern California). Many of the concerns surrounding a “national” MLS system, database, and organization would still hold to some degree if one substitutes the terms “state-wide” or “regional” for “national,” and many regional systems may still not have quite the buying power or governance advantage of a national model. Nonetheless, a new modern network of fewer MLSs could provide either a viable middle ground between the current model and a national model, or a logical building block toward a national model.

Finally, we might consider a hybrid model: that is, simultaneously with MLS regionalization, to evaluate the formation of a National Association of MLSs. Just as Realtor® associations cooperate to have a large voice via the National Association of Realtors, perhaps MLS operators can evaluate that same model to improve their success.

Conclusions

Some put forth “National MLS” as an answer to perceived threats to our industry, while others see “National MLS” as the threat itself. There is no doubt that this debate will continue to be divisive. However, going forward, the industry should be able to have productive conversations about whether we want a national MLS system, database, and/or organization. Each option has its own disadvantages.   These issues have been raised not with the intention to shut down discussion or hamper progress toward the goal, but rather, as a starting point toward a continuing discussion of each national goal’s value and risks, and how those risks might be mitigated if the goal is deemed worth the effort. Some might point to possible threats on the horizon and say, “Now is the time to figure this out and act toward a national MLS”; and it seems wise that this conversation should move forward with deliberate speed. But, as an industry friend of mine once said, “We’re an industry driven too often by our fears and not enough by our dreams.” Let us consider, with an open mind, the MLS future that would be best for practitioners and their clients, using that as our inspiration in evaluating the potential for consolidating MLS interfaces, databases, and organizations.

Posted in Matt's Real Estate Technology Blog
2 comments on “National MLS, Whatever That Is
  1. Judith lindenau says:

    Well argued, Matt. Thanks for your clarity in addressing these issues….even if I may not agree with your conclusion ( I like a single core program with app customization…this would stabilize and define the data products, making reports much more usable and consistent…..)

    • Matt Cohen says:

      Judith – the point of this wasn’t to have a conclusion, rather to point out pros and cons of various approaches to what is crudely referred to as “national MLS”. Every approach has pros and cons. The industry could go “your way” – or any one or more of these ways – as long as it goes in with eyes open, with plans to mitigate the cons.