In February 2009, Richard joined Cint AB, a technology and solutions provider to the research industry, to set up its UK operation and to introduce Cint's Panel Exchange market place (now OpinionHUB) and range of self-service research tools to the market.
A Finite Supply: Finding a Sustainable Model for MR Response
As demand for data and insight grows, how will the MR industry's current supply model be sustained to ensure future availability of online respondents? Richard Thornton, Cint's Global Sales and Operations Director, outlines the challenges facing the industry, and makes recommendations for how this can be addressed.
8th December 2014
The challenge
The MR industry has changed beyond recognition in recent years and, quite frankly, it needed to, although it’s not done yet by any means. Inefficiencies have been notable, with emerging technologies simply making them more prominent and highlighting the outdated nature of certain parts of the industry even further. The market has been flooded with too many players, making it fragmented and processes convoluted. An abundance of steps between buyer and ultimate seller, with an over-reliance on a relatively small pool of the population, has simply added to the situation.
Traditionally, data collection was conducted offline, face to face and later via the telephone (CATI/CAWI); by the late 1990s and early 2000s, online as a methodology started to gain traction; by the mid-2000s, online had become a viable methodology in major markets for a large chunk of (primarily) consumer research.
But along the way, not enough forethought has been given to what the supply chain will look like over the next five to ten years, and what’s required to maintain a steady stream of available and willing respondents, as the entire approach has been geared towards sourcing respondents with the ‘here and now’ as the key driver.
The approach to market research has also evolved, moving from the data collection choice being inherently required to be ‘fit-for-purpose’ in light of the research brief to the domination of either price or speed – often to the detriment of quality. This is why in many markets today, especially where labor costs are reasonably high and internet connection has high penetration, online is the leading method.
However, while all this evolution has been occurring in the industry, there is one age-old challenge that remains: how is the industry going to continue to get people to participate in research and respond to surveys? We already know there isn’t an infinite population of respondents to take surveys, and as technology has evolved along with the internet and trends like social media and mobile have grown, consumers have more ways to spend their time, creating a competitive landscape in grabbing people’s attention and interest. Would you rather take a survey or spend time gaming or on Facebook? Would you be happy providing your opinion in response to a few questions or instead spend your time online gaming?
Keeping a sustainable supply chain of willing participants has also been affected by some other dynamics in the MR market, not least the fact that overall respondent experience has arguably diminished (in some cases it is downright awful). Surveys over time online have become longer not shorter, respondents are exposed to too many poorly designed questionnaires, often not suitable for the device they are being taken on, and incentives have been reduced or cut altogether – or replaced by new currencies that don’t necessarily have the affinity attachment or cash value that they once would have had (as players in the supply chain are forced to strip out cost and find ways to increase their margins). Over-solicitation of panelists is a problem; they receive too many invites, many not relevant or intelligently targeted, are offered fewer payments for their time and encounter a rather ambiguous approach to transparency around what they can expect when they are invited to participate.
In today’s MR model, the quality of product being served to the most important stakeholder in the research process, the respondent, isn’t what it should be. Respondent care has been sidelined in favor of the desire for stacks of data, however good or bad that data might be. The industry seems to be obsessed with cramming as many questions as possible into the same questionnaire, worried that if they don’t, they’ll never get an opportunity to ask those questions again. The irony is that the current approach just ensures that the experience of taking surveys is an unforgettable one, in service of misguided rationale.
Situation in the US... Europe... Asia
Many of these trends originated in the US, where the online data collection industry matured quickest and where there has been a huge appetite for faster, cheaper, bigger data collection - moving so quickly that those who govern MR have been challenged to keep pace with ensuring guidelines and best practices are updated and relevant to business practices of buying and selling data - in an environment driven so aggressively by the buyer it is not surprising to see bad habits creep in as suppliers struggle to keep up with demand - although this is changing, which we will come onto later. There is also a lack of appreciation in some quarters of what it takes to maintain a reliable, robust, good quality supply source. But with pressure on budgets and to the inevitable squeeze all the way down the value chain, it’s perhaps not a surprise that quality suffers.
Building and keeping a sustainable supply source of respondents should be the single most important thing around data collection in MR (data collection that requires a task or self-reported data at least) and it is in the interests of those that need this type of data to keep it alive. If not, the danger is that passively measured and transactional data will start to replace traditional data collection (it already does in some cases, but in most works alongside or on top of it), but even that has its limitations - still today, the typical consumer is somewhat uncomfortable with downloading apps and trusting and accepting tracking meters.
The challenge is change, which is never easy and often resisted. Overhauling what is a fairly inefficient supply chain model in MR is no small feat, but for the longevity of the industry, it needs to happen. Some of the older challenges have been added to with new ones; in the US, an increasing proportion of completed interviews for research surveys are not from double-opted-in (DOI), profiled ‘panelists’, but are from single-opt-in (SOI) and purer ‘river’ sources (profiling and targeting respondents on the fly by website intercept) along with traffic from lead generation and reward-program companies (this last type of supplier is targeting respondents in ‘offerwall’ type environments, moving away from a direct-email-invite model to one where the respondent has the control over the type of offer and the time at which they take a survey).
Looking at sample sources today, and how these will evolve, is where it gets interesting and where the stark differences between markets are evident - namely, what’s playing out in North America vs. the rest of the world. Traditionally, online surveys have been completed by DOI panelists on proprietary access panel assets receiving surveys via a direct email invitation model. There have also been alternative supply models, like those from Cint (a panel exchange platform where panels are built with brands, publishers and the like based around ‘affinity’), however again traditionally a DOI direct email invite model. Geographically, DOI panel assets have ruled in Europe and Asia (and still do for the most part), while in the US the market dynamics have been under a sea of change for quite some time, moving towards alternative supply sources and different methods to reach and target respondents; SOI, river, reward-program traffic. This has been partly driven by a necessity for a broader, larger pool of respondents, by behavioural change in consumers and how they access content and allocate their time towards tasks such as participating in surveys, and is partly due to the fact that for large chunks of research conducted and target groups required for interviews, the economics to invest, build and maintain DOI panel assets are arguably no longer viable as the only solution in the US.
Panel recruitment is also inherently tied into online advertising costs and the cost per recruit; taking into account attrition and general maintenance of assets. It is also a very crowded market and the panel recruitment ‘farms’ and other previously reliable sources of recruitment now have many options for monetizing their traffic, so they and anyone building an asset of recruiting is in a much more competitive and savvy field today than ever before.
As a result, in the US, we have seen the rapid growth of panelists being delivered to live surveys in more diverse and innovative ways. Cint and others, have developed supply-APIs that allow suppliers to direct respondents into surveys, often into a router first, on the fly and to be very efficient with targeting, quota management, and speed and volume of delivery. These sources work effectively alongside DOI traffic and add to the supply pool available in the market for buyers to benefit from. Many of these new sources are from non-traditional MR companies like MyPoints, Prodege, Points2Shop and many others who have seen an opportunity to disrupt the current supply model and innovate through technology and automation, and they are doing extremely well. While back in Europe, the trend continues to see much wider adoption and utilization of DOI panels, investment in assets and recruitment, and to view river and single-opt-in sources with more skepticism than their US counterparts, perhaps because the need is not as great since many panels in Europe are generally relatively healthy and tend to still deliver.
Over in Asia, there is a mix of everything - DOI panels still dominate, although ‘offerwall’ environments where respondents can pick and choose their surveys in their own time are very popular, and there are a crop of platform players and supply models more akin to the exchange system Cint runs globally, where panels are built with publishers and media owners, and opened up for access to third-party buyers; plus mobile is a viable methodology and widely adopted ahead of PC and laptops in many southeast Asian markets and particular demographics.
Committed to change?
Technology is playing a significant role in the changing landscape of online data collection, and doing things the way they have always been done just doesn’t cut it anymore. On the topic of MR supply chain inefficiency, major buyers are genuinely worried and discussing the future of supply for their businesses; suppliers are concerned and challenged by increasing costs of recruitment, maintenance and downward pricing pressure, while those taking surveys are filling their time and satisfaction in other newer, more exciting ways than taking surveys.
This is where technology comes in. Most organizations reviewing their own business and ways to improve operational margins have already done the obvious things first – cut headcount, centralize procurement, tighten the belts on costs, squeeze their vendors on price, remove discretionary budget items – so beyond that technology is the greatest agent of change. And there are two routes here: buy or build.
Many appear to be building upon their own technology capabilities; this is commonplace amongst the many traditional access panel firms that are busy reinventing themselves as something new, whether that is a platform, an exchange model, or a nuts-and-bolts full-service online research business. So building routing, API and river capabilities along with a more profitable recruitment model is a natural extension of that. Others are buying and licensing technology to improve supply chain management from the likes of Cint and others.
Whatever the route, what’s changing the supply landscape for online data collection is buyer behavior in the US market, along with APIs. . As a result of system integrations, it is now possible toautomatically purchase and manage sample and the supply chain while operating within one’s own system. Some of the larger MR firms are either doing this already with chosen partners or figuring out how to. Many of the leading suppliers have sophisticated technology capabilities and APIs, both demand and supply-side, connecting up to survey-tool software, panel management systems, eye-tracking and emotion tracking software, and the influx of tech start-up firms that need quick, affordable and efficient insight but don’t have audience access.
Routing technology is now used broadly amongst nearly all leading panel suppliers. If it hasn't already, it will soon reach a point where if you don’t have a router and good technology to drive sample optimization and efficiency, you will continuously struggle to be competitive – especially with where pricing is today and where it's heading. Winning will mean having a router plus a multitude of different supply sources, plus volume, with in-built quality that’s market suitable – without these, panel assets and providers will likely be forced to connect up to other existing panel owners, or join an exchange platform for distribution and access to global buyers. It is important however, that technology should be an enabler that works alongside best practice and robust sampling methodology, not to the detriment of it.
Future scenarios
The key contributors to creating a sustainable supply model will be better standardization of adoption and usage of technology within data collection and sampling around routers (the recent ARF Router initiative is a good example of a step in the right direction). As well, the expansion of the pool of survey participants into non-traditional sources, and ensuring that the quality controls and associations circling MR (ESOMAR, CASRO, ARF, AMSRS etc) are close by.
Other positive actions include increasing transparency, reducing the number of steps in the procurement process and management of the supply chain, removal of middlemen and agents that don't add value, and undertaking controlled sample blending and mixing of panel sources to understand skews, biases, and impact on data quality and reliability. Reducing the distance between buyers and sellers will also help.
So change is here and more is coming. Expect further consolidation, more and more non-traditional MR players entering the game on the supply side, and US buying behaviour heading to Europe. In one very possible future scenario, we may see a handful of global platforms acting as exchanges on the supply side for distribution while technology, routing, river, and pricing become barriers to entry and sustainability for smaller, local or regional players (not too dissimilar to the advertising, finance and other industries where exchange models have thrived in a B2B environment), with niche and quality-led players still having an important role to play. This will be enhanced by more collaboration via APIs to create a deeper, broader, better-profiled and more active pool of supply sources and respondents.
But remember, without regular people like you and I willing to take surveys and participate in research, there will be no sustainable future for the industry. So let’s not kill the one thing that makes everything tick and help this industry embrace everything that it has strived to always do: understand consumers. So doesn’t this also include those providing the insight?
