About Nesta

Nesta is the UK's innovation foundation. An independent charity, we help people and organisations bring great ideas to life. We do this by providing investments and grants and mobilising research, networks and skills.

Nesta Operating Company is a registered charity in England and Wales with company number 7706036 and charity number 1144091. Registered as a charity in Scotland number SC042833. Registered office: 1 Plough Place, London, EC4A 1DE

www.nesta.org.uk © Nesta 2013.

Acknowlegements

We are grateful to the following individuals who have provided us with ideas and thoughts throughout our work in Destination Local to date: Chris Bruno, Dave Harte, Hugh Flouch, Jon Hickman, David Hirschman, Andrew Jackson, Damian Radcliffe, Sion Richards, Judith Townend and Andy Williams.

We would like to thank the Destination Local projects and strategic partners for their help and support during Destination Local, especially during the drafting of this report, as well as our partners at the Technology Strategy Board, and their supported projects. We would also like to commend Innovation Intelligence Group, Kantar Media, Mavens of London and Oliver and Ohlbaum for their contribution to the Programme through their delivery of high quality outputs to tight timescales.

We would like to thank Nesta Fellow Professor Ian Hargreaves and Nesta Associate William Perrin for their thoughts and insights.

Finally we would like to thank our colleagues at Nesta for their support during this phase of Destination Local, especially Hasan Bakhshi and Juan Mateos-Garcia.

1. Executive Summary

Hyperlocal media may be defined as online news or content services pertaining to a town, single postcode or other small, geographically defined community. In this, our interim report, we set out our findings and lessons learnt in the past year from Destination Local, our hyperlocal media programme.

A key component of Destination Local is its practical element. This involved funding the development of mobile-enabled experimental hyperlocal projects around the UK at up to £50,000 each from July 2012.

In addition, we commissioned several pieces of external research including a UK-wide consumer survey to understand the demand for and use of hyperlocal media. We complemented this with a proof-of-concept work which uses consumer search trends to model demand for and use of hyperlocal content.

Given the historic importance of advertising as a means of funding local content, the prevalence of advertising in the business plans of the Destination Local projects, and a lack of robust data, we also commissioned research into local advertising markets, and the potential for advertising to fund hyperlocal media.

The purpose of the interim report is to provide practitioners with some practical recommendations derived from our findings so far. In addition, we also publish some recommendations for public authorities, which can play a significant role in helping to foster civic engagement via hyperlocal media.

This interim report sets out our initial findings from the Destination Local prototype projects. Throughout the report we also draw on evidence from our externally commissioned research and from additional research we have undertaken in-house at Nesta.

Since all of the projects launched between February 2013 and May 2013, it is too soon to carry out a complete evaluation of Destination Local at this time. Much of the data gathered from the prototypes are limited. However, we believe that the evidence gathered to date is representative of issues faced by many hyperlocal media producers and is therefore worthy of publication at an interim stage in order to share lessons for the benefit of the sector.

Understanding how services are being used is important for those seeking to increase the reach and impact of their services

Hyperlocal services may generate relatively little traffic in absolute terms - most of the services we analysed in Destination Local reported around 5,000 or fewer monthly unique visitors. This compares to the millions of visitors associated with some larger media properties. The small volume of traffic to some hyperlocal sites means that traditional industry metrics which include the use of online panels may be less appropriate. Other approaches to measurement which use server data can be used to measure traffic to these sites.

Some services are able to achieve relatively high numbers of users compared to the population of their coverage area, and can also extend their reach through the use of social media, which is a major source of traffic for several of the services. Our analysis of project data suggests that a media-rich service including videos for example, may help to keep visitors on the site for longer. Sites which tended to post multiple content items on a daily basis often had more traffic than those which did not, allowing for a service serving a relatively small population to have a higher number of visitors than some other services serving a much larger population.

Ensuring that websites work well with mobile devices is important given that in general around 30-45 per cent traffic to the project websites in June came from mobile devices (including tablets). For those websites which have been established for several years, our analysis of their traffic suggests that the proportion of mobile visits has increased over time. For example, a mobile version of a service should ensure that users can get access to key content quickly given that mobile users spend less time on average per visit than users accessing the service from another site.

We believe that all hyperlocal services should consider how they collect, analyse and use data, and whether it may also be worth carrying out consumer research to better understand the needs and wants of users. This may be especially relevant for those intending to develop mobile apps given that some projects found it difficult to report on take-up and use of their apps because of technical issues and the granularity of data provided by the relevant app-stores.

Our findings to date suggest that HTML5 mobile-friendly services may be a better option for many hyperlocal media services

The ten projects reported to us that generally mobile apps cost more to develop than HTML5 services, which can run on a broader range of devices than a single 'native' mobile or tablet app. This is especially relevant for hyperlocal services which cover areas with relatively low take-up of smartphones and data plans. Another advantage of using HTML5 is that it does not require approval from platform operators for deployment. Technology has matured to the point that an HTML5 service can do many of the same things as a native app. Despite this, some see apps as being more flexible and more in line with consumer demand – some people expect to find an app when they look among the hundreds of thousands of apps available to download for the major smartphone platforms.

We note that in the US, there is currently much discussion around integrating mobile hyperlocal services with bricks-and-mortar retailing. To date there has been limited integration of Destination Local projects into point-of-sales and payment but willingness to use QR codes, and a lack of distinct legal identity for some hyperlocal services may act as a barrier to this.

Services are continuing to experiment with business models to secure financial sustainability: Starting a printed edition may be one way to do this

At the time of writing, none of the prototype activities by the ten organisations have been able to make a profit on a stand-alone basis.

Advertising revenues were sought by most of the ten services, but selling online advertising to businesses can be difficult, reflecting our previous research which suggests that small local businesses have limited awareness and interest in advertising on hyperlocal websites. In contrast, creating a printed version of the hyperlocal service may help profitability. Projects said that it is easier to sell print advertising and this is consistent with our previous research as well as wider hyperlocal industry trends. A printed edition of a hyperlocal service may also help to market the online service to potential users and drive traffic to the site and/or app. Other projects have tried to experiment with alternatives to traditional online display advertising, such as sponsored content, paid-for notifications, paid-for business listings or video adverts.

While hiring dedicated advertising sales staff may help to grow online advertising revenues, and free up editorial resource, this needs careful consideration given the significant cost involved. Finding the right candidate who understands the local area and can cover their salary cost is important.

The ongoing cost of content is the biggest spend for hyperlocal providers. Technology spend is relatively cheap in comparison. Overall hyperlocal media costs compare favourably to those of traditional media focussed at the town, village and/or postcode level. A volunteer-run hyperlocal website can cost less than £100 a year, though a professional service producing news is likely to be around £30,000 a year or more and may rise more than to £100,000 if several members of staff are employed on a full-time basis.

However, as the projects told us, sustainability is not just about finance. Projects also reported working long hours and finding it a challenge to recruit volunteers – especially younger people.

These challenges to sustainability are not unique to the UK. Since we published Here and Now, several American hyperlocal networks have made cuts to editorial budgets and reduced headcount and one has closed as owners continue to struggle towards profitability.

The Destination Local services generate more than economic value to their shareholders

Our experience to date demonstrates that the Destination Local projects have created value beyond the economic to their shareholders. For example, all provide content to users which may have both social as well as private value, and most of the projects generate content, in addition to or instead of aggregating it. But the relationships that the projects have established with their communities means that the impact of these projects goes beyond providing news and other geographically relevant information. For example:

• Some organisations have used their hyperlocal media activities to engage in direct social and community action, for example, by arranging and co-ordinating responses to public consultations.
• Several projects aim to strengthen local democratic processes, either by providing a direct mechanism to feedback opinion to the local authority, or through their coverage of local politics.
• Projects also supported education and training opportunities. Two organisations work with local colleges, offering students the opportunity to learn about media production and journalism by creating videos for the hyperlocal services. Other projects offer internships or training sessions to young people and adults in the local area.
• By providing a cost-effective means of imparting information to citizens about public services, hyperlocal media organisations may help to make the delivery of public services more efficient.

Lessons and recommendations for practitioners, industry and policymakers

From our programme of practical experimentation and our research, we make the following recommendations based on the lessons we have learnt to date:

1. Hyperlocal media organisations should consider partnerships with larger media organisations: Partnerships between larger more established local media companies and hyperlocals (in particular those operating on a not-for-profit basis) may increase visibility and drive traffic to these services. There may also be opportunities for commercial partnerships to provide hyperlocal content to local radio stations or regional newspapers.
2. Commercial hyperlocal media organisations should consider forming or joining an industry association. By forming or joining an industry association, stand-alone hyperlocal media organisations have the opportunity to co-operate in commissioning further and ongoing research addressing the current gaps in time-series industry data, such as use of hyperlocal services and local online advertising expenditure. Industry associations could also play a role in making the sector and its opportunities more visible to advertisers, investors and policymakers. The organisation needs to aggregate traffic figures from its membership and encourage site owners to understand and be transparent about their audiences. This will attract potential advertisers.
3. We encourage local authorities and other public bodies to engage more fully with the hyperlocal sector. Almost all of the Destination Local projects reported having a relationship with local government, but the quality and level of engagement varied. Local authorities and other public bodies can work with hyperlocal websites to engage in a cost-effective way with residents who may be less likely to engage with the local authorities through other channels.
4. The BBC should link from its websites to a broader range of hyperlocal media organisations. By opening its website to more external links to trusted stand-alone hyperlocal services, as well as to websites operated by newspaper groups, the BBC can address concerns about the lack of 'local' news on bbc.co.uk, without needing to make significant new investments in local staff and offices. Audiences would gain through having access to a wider range of more granular content, while the services themselves would gain through increased traffic to their services. This should be achieved as a matter of priority.
5. We believe that there is a role for industry and funding bodies such as the Technology Strategy Board to work together to facilitate the development of platforms with the potential to nurture and support hyperlocal services in the UK. While our work to date has enabled us to better understand hyperlocal media in the UK, and that there are many examples of hyperlocal media services producing content valued by its users, there appear to be relatively few examples of UK-developed platforms which support and promote hyperlocal content. While there are global platforms which do host hyperlocal content, there is a risk that advertising revenues which flow to them are leaving the UK's content economy, reducing the sums available for investing in hyperlocal news and other socially valuable services.

2. Introduction

Nesta has defined hyperlocal media as online news or content services pertaining to a town, single postcode or other small, geographically defined community.

While there is already recognition that local media can be a valuable and trusted form of news and information, it was less clear how this emerging type of very local content might fit into the rapidly changing landscape of digitised local newspapers or fare alongside the promise of local, digital television. Might hyperlocal media, with an increasing diversity of authentic voices and an ability to directly engage with very small geographical populations, produce new and sustainable forms of public value?

The landscape is potentially very exciting. New online services can be more geographically specific than ever before, and there is an explosion in the adoption of ‘location-aware' devices. Moreover, the economic costs of running hyperlocal services means that practically anyone who wants to can set up and provide valuable information about where they live or work.

With the launch of its Destination Local programme in March 2012, Nesta and its partners set out to understand the potential opportunities for hyperlocal media in the UK. Our goal is for the UK to have the most advanced public understanding of hyperlocal media and how it might thrive. This includes investigating:

• The demand for and the markets in which hyperlocal media services operate.
• Business models that are sustainable and might support the growth of hyperlocal media services.
• Emerging and future trends.
• Potential private/public investment opportunities.
• What policy interventions might support a strong plurality of hyperlocal media.

To do this, we have taken two approaches. Firstly, Destination Local commissioned and published original research where there was a significant knowledge gap. Local media tends to operate in small and fragmented markets and consequently there was almost no aggregated data about the scale or nature of demand for hyperlocal content or for the size of the advertising markets that might sustain it. Secondly, Destination Local funded a number of prototypes that aimed to assess the potential for technological innovation, especially around mobile and location-based services.

This interim report sets out our initial findings from the Destination Local prototype projects. Throughout the report we also draw on evidence from our externally commissioned research and from additional research we have undertaken in-house at Nesta.

The purpose of the report is to provide practitioners with some practical recommendations derived from our findings so far. In addition, we also publish some recommendations for public authorities, which can play a significant role in helping to foster civic engagement via hyperlocal media.

Since all of the projects launched between February 2013 and May 2013, it is too soon to carry out a complete evaluation of the projects at this time. Much of the data gathered from the prototypes are limited. However, we believe that the evidence gathered is representative of issues faced by many hyperlocal media producers and is therefore worthy of publication at an interim stage in order to share lessons for the benefit of the sector.

Original research

We commissioned three pieces of research from external parties, Kantar Media, Oliver and Ohlbaum and Mavens of London to help address what industry round tables had identified as being gaps in knowledge concerning two critical areas. Firstly the nature of demand for and use of hyperlocal media in the UK; and secondly, the potential for local advertising markets to act as a source of funding for hyperlocal media.

Understanding the demand for and use of hyperlocal media

To better understand the nature of demand for and use of hyperlocal media we commissioned a UK-wide omnibus survey from Kantar Media which we believe provides us with a comprehensive view of the reported use of, and attitudes towards hyperlocal media across the UK.

The summary and Kantar's full research piece can be found at: www.nesta.org.uk/publications/reports/assets/features/uk_demand_for_hyperlocal_media

We complemented this traditional survey approach with a modern technique from Mavens of London, which looks at the online behaviour of consumers through search engine query volumes. We believe that this is the first time that this methodology has been used in the UK to help understand the demand for and use of hyperlocal content.

The summary and Maven of London's full research can be found at: www.nesta.org.uk/areas_of_work/creative_economy/destination_local/assets/features/understanding_the_use_of_hyperlocal_content_through_consumer_search

Understanding local advertising markets

The size and nature of local advertising markets is a key area of interest for hyperlocal providers. This is unsurprising given that the sustainability and growth of both commercial and not-for-profit media services are based upon, to a greater or lesser degree, their ability to generate advertising revenue.

Despite many assertions, there was little hard evidence about the size, or potential size, of hyperlocal advertising markets in the UK. To address this, we commissioned some work from Oliver and Ohlbaum, one of Europe's leading media strategy consultancies. Their research draws on insights from 20 in-depth interviews with people and organisations active in local advertising markets, discusses the industry data currently available on local advertising, and finally presents the analysis of a survey we commissioned of 1,053 small businesses in seven areas around the UK, which looks at their use of advertising and their awareness and potential of hyperlocal media.

The summary and full research piece from Oliver and Ohlbaum can be found at: www.nesta.org.uk/areas_of_work/creative_economy/destination_local/assets/features/local_advertising_markets_and_hyperlocal_media

Our work in this first phase of Destination Local provides a strong foundation for further research

While our work in this area is a valuable starting point, we hope that others are inspired by our research. More work is needed to understand the nature of this nascent, but emerging media sector. There is a strong role for both industry and policymakers to come together to aggregate research on audiences, viable business models and successful content so that the social and economic aspects of hyperlocal media are better understood by potential advertisers and investors.

Our findings from our original research over the past year suggest that the hyperlocal sector as a whole is highly valued by audiences. Increases in reach and consumption are being driven by mobile and smartphone usage. Those who consume hyperlocal media are highly satisfied with the services they use. People are more likely to consume hyperlocal media if they are in the ABC1 demographic and/or at certain times of life, such as when they 'put down roots' due, say, to buying a home or when they are starting families. These findings suggest exciting opportunities for hyperlocal media providers seeking to target specific audiences.

However, our research also highlights key challenges concerning sustainability and growth. Oliver and Ohlbaum's report suggests that advertising spend in this area is likely to be relatively small in comparison to traditional media. It is also highly contested, increasingly by centralised technology platforms or by traditional local media businesses moving online. Other forms of successful business models are yet to be established. Consequently, this emerging sector is very fragile, with entrepreneurialism and a willingness to try a range of business models often being the defining traits of success for small, hyperlocal media providers.

It is these opportunities and challenges that we further investigate from the evidence provided from the prototype projects we have supported.

3. Description of Destination Local Prototype Projects

Nesta issued an open call to fund hyperlocal prototype projects in March 2012. It received 165 applications for consideration. They were diverse, and reflected a wide range of business, content and ownership models. From the many applications, we selected ten projects from England, Scotland, Wales and Northern Ireland. Nesta announced the successful projects in June 2012.

Since then, as well as delivering their new services, the projects have also participated in programme workshops and networking events, and have supplied us with qualitative and quantitative data about their projects on a regular basis. This has been invaluable in helping us to write this interim report and we are grateful to the projects for allowing us to share the lessons learnt more broadly.

In parallel and in partnership with Nesta, the Technology Strategy Board also launched a call for location-based technology projects, with the successful applicants joining the Destination Local workshop events. A full list of these projects is outlined in Appendix 1.

Taken as a whole, the portfolio of ten Nesta projects covers a broad range of ownership models, including two local authority-backed services (LocalSay and Locali), commercially run-for-profit services such as the Kentishtowner, a service owned by a newspaper group (Our Town) and services run by the third sector (Local Edge and Papur Dre). The type of service also varies, including one with its own infrastructure (21VC/Loddon Eye) through to a video-centric hyperlocal TV news service (URTV). Our portfolio included services covering a rural area (MyTown) as well as urban (The City Talking/Solomon).

Each of the ten projects takes advantage of modern mobile-friendly technology to deliver the service to its users. All projects use geotagged content and/or advertising. Some projects use native mobile apps, while others have developed mobile friendly HTML5 services, instead of, or to complement a native app. Some projects have developed, or are intending to combine geolocation, geotagged content, and the compass, camera and accelerometer features on smartphones and tablets to offer users an augmented reality user interface. We believe that our ten projects give us an insight into a wide range of technologies relevant for modern mobile hyperlocal services. While some of the technologies, such as the use of GPS, may also be used by location-based services, hyperlocal media services may use these in different ways.

Note: 'Website' includes HTML5 websites/services which may automatically adopt the layout to a mobile device. Web apps are HTML5 services which may make use of local storage, allowing some functionality when then device is not connected to the Internet. Scottish ward populations relate to 2001 census. Source: Projects, census data.

4. Measuring Hyperlocal

In this section we discuss our initial findings on the important issue of measuring and analysing the consumption of hyperlocal services.

Due to the small number of users for hyperlocal services, standard industry metrics cannot always be used

A perceived advantage of online media is that it is possible to monitor usage and engagement accurately and in real time, in contrast to the survey-based methods used for local radio (RAJAR), newspapers (National Readership Survey - NRS) and local TV (in the US, France and Spain).

However, despite the range of audience measurement systems which are currently in use across a range of platforms, there is still no single-source measurement tool which enables audience use of hyperlocal media to be captured across all platforms.

ComScore's Media Metrix (MMX) product uses a combined methodology which tracks all the websites visited in its panels of home and work Internet users, and also combines this with server data from participating publisher websites. The GSMA Mobile Media Metrix product also gives comScore a census view of mobile Internet traffic. Combining these data sources into comScore's MMX Multi-Platform product gives an indication of unduplicated audiences across mobile and PC/laptop access. However, a hybrid panel-server approach may not be so suitable for hyperlocal media given that the audience bases may be too small for the panel-based approach to be robust. We note that UKOM/comScore requires a minimum threshold of visits to a site for it to be counted. This means that it is possible that very small sites may receive regular traffic but remain unmeasured if none of the panel uses the service. This is a major issue for national brands which may otherwise be potential hyperlocal advertisers.

We've therefore concluded that there needs to be a very simple way for hyperlocal services to monitor and demonstrate what audiences they are reaching. Being open and transparent about audience numbers will help increase the confidence of both local and national advertisers to buy media space.

At an industry level, it would be useful to aggregate reach and consumption metrics of hyperlocal services to demonstrate the mass appeal for these types of content across the UK as a whole. This suggests that either existing online industry bodies, such as the Association of Online Publishers (AOP) or the Internet Advertising Bureau UK (IABUK) or an entirely new industry body can play an important role in providing information useful to advertisers.

Not all hyperlocal media providers routinely gather and analyse their audiences

We required each of the Destination Local projects to submit qualitative and quantitative data to us on a monthly basis. Quantitative data requested each month included details such as monthly unique visitors, page views, as well as app downloads for those services which were developing mobile apps. Although the focus of the Destination Local experimentation is on the use of mobile technologies, we felt that it was important to also collect data on broader aspects of the service, including more the more traditional elements such as standard websites. Other data relating to content production and financial data was also requested from the projects on a monthly basis.

Some of the projects, especially those run at very low cost, or those managing high numbers of volunteers, were not used to capturing and then analysing their data in more than the most basic of ways. One of the first things any media service should be focusing on when considering how to provide a better service, is to allocate resources towards capturing and understanding who is using the service. It is especially important for a hyperlocal service to understand where audiences are geographically located and why they are interested in the service (e.g. what percentage of a service's audiences are residents, diaspora, tourists or workers in an area). Analysing this information will dramatically increase audience insight and help improve decisions on specific content choices, advertising opportunities and where to invest in promotion.

At the current time hyperlocal services may need to consider carrying out their own consumer research in addition to using their analytics

In the current absence of a hyperlocal industry body, hyperlocal media providers should also consider carrying out further primary research to gain greater insight into their use than is possible solely with server-side data. This may include, for example, running on-site polls (which enable demographic data about users to be collected¹) or a telephone or face-to-face survey in the service's coverage area, which potentially allows for reach, use by demographic group and use to be assessed.

If developing apps, plan carefully for audience metrics right from the start

The experience of the ten projects to date suggests that the metrics available to app owners vary according to the platform used and how the app has been developed. One of the managers of a Destination Local project reported that they had been frustrated with an early app they developed because they found it difficult to get data from the relevant app store about the number of downloads and usage.

The way apps are designed and coded also impacts the type of data which can be collected about app use. For example, while there is no direct comparison of a page-view in an app, trigger points/actions can be coded in to app an to report this event back to the server and hence to the app developer. One project told us that their use of a rapid application development (RAD) tool for their app had resulted in some incompatibilities with their analytics system, meaning that the analytics system failed to recognise the app. As a temporary workaround, the developers forced the analytics system to treat the app as a website, though this meant that certain key metrics such as app downloads and length of time per session were unable to be collected.

Those services developing apps should therefore plan the gathering of metrics carefully into their production process.

5. Performance

This section of the report highlights key metrics from the performance of the ten prototype projects. Since some of these metrics are commercially sensitive, we have anonymised the charts and discussion which follows.

In general, the number of monthly unique visitors to the Destination Local projects has been relatively static, though a few sites have reported upward trends in unique visitor numbers since the start of the programme.

One project (Website A) reported a significant spike in traffic in one month (to over 70,000 monthly unique visitors) in March 2013, with a feature which attracted comment on social networks, and which became prominent to a UK-wide and international audience. This shows that some content while being hyperlocal, may also draw interest from a broader audience. This same project has seen the number of visitors rise over time from the start of the project.

While there is a general trend between population of coverage area and the number of monthly unique visitors, this is not the case for all services which suggests that penetration varies significantly between the projects. In the chart below, websites A–D have a coverage area whose population exceeds 25,000, while E–G have a reported coverage area of 25,000 people or less. Despite this, in some months, website E outperforms site B, which covers a significantly larger population base. However, four of the websites, which include both urban and rural services, report fewer than or up to 5,000 monthly unique visitors.

It is difficult to provide robust figures for penetration, given that a unique visit does not equate to an individual person, and that visits may be made by people living outside the coverage area (see below). However, dividing the total number of unique visits by the population of the coverage area gives an upper bound to the reach of the service, although in reality reach may be significantly lower. In Q1 2013, these 'penetrations' ranged from 2 per cent up to 45 per cent. The rural services tended to have higher reach measured in this way than the sites serving urban communities. This may reflect greater competition for attention from other media serving urban areas, but also suggests that services need to invest more in advertising to reach urban audiences.

We also collected data on those new services which started up in the period of the project. To preserve their anonymity and to allow comparability, the base of this chart is from the launch date up to three months and includes active users (for apps) and website unique visitors (for websites). We note that to date several projects have held 'soft' launches and are planning a more formal launch and marketing campaign in summer 2013. Despite this, there are differences in performance, and some projects have expressed their concern at what they see as the poor performance of their new services — a few projects reported fewer than 100 monthly unique visitors or users at the end of April 2013, despite the service/app being available to the general public. We believe that one project in particular suffered low performance because it required user registration. This backs up our Kantar research which suggests that any form of obstruction to local content is a potential 'put-off' for audiences.

The low performance of some of the projects also highlights the need to effectively market and promote the services. The 'soft launch' of some of the prototypes has not helped audience numbers.

However, we intend to continue to monitor the performance of these brand new hyperlocal services to assess their success over the longer term.

We note that there is a difference between installing and using an app (something which we were careful to note when designing the survey for our consumer research).² Native apps must first be installed from an app store, before being accessed and this may create a barrier to use. Research from the US suggests that on average each smartphone user has 41 mobile apps on their smartphone, but overall use of many of these may be very low.³ Therefore the installed base of an app may not really represent the number of active users. Again, this suggests the importance of effective promotion of the app and regularly refreshed content in order to encourage users to check the app frequently.

Google, Twitter and Facebook can increase reach and be critical sources of traffic for hyperlocal websites

Our work with the projects to date suggests that an effective social media strategy is invaluable for increasing the visibility and reach of hyperlocal media services. The organisations developing the ten prototype services have increased their number of Twitter and Facebook followers throughout the duration of the programme, though the relative importance of these social networks in terms of aiding the reach of the hyperlocal service varies. One service, which developed from a Facebook page, reaches over 50,000 people via its presence on Facebook, which is more than twice the average of the number of unique monthly users the service receives.

An analysis of source traffic for the majority of Destination Local project websites suggests that Google, Facebook and Twitter account for significant proportions of sources of visits, though for some of the websites which launched in Q1 2013 (websites 6–10) direct visits (from people typing the address in to the browser) accounted for a significant majority of traffic. Facebook was an importance source of traffic for websites 1, 2 and 4; these are all websites that generate news stories on a daily basis. In contrast, across the sites, Reddit, Wikipedia and the BBC were marginal sources of traffic to the websites.

The Mavens of London research highlighted that www.bbc.co.uk in particular often appears as one of the highest search engine returns for local searches, despite it not having relevant up-to-date content on the service. This may demonstrate how the 'gravitational pull' of BBC Online's search engine rating attracts away audiences that might otherwise visit hyperlocal services. To address this, we believe that BBC Online should be doing more to drive its own audience traffic to hyperlocal services where editorially relevant.

Our previous research found that comparatively few users made use of following people or organisations on social networking websites (11 per cent of hyperlocal media users) or followed people or organisations on Twitter (7 per cent) to get news or information about the immediate area they are in. However, in late 2012 Facebook was the third highest app penetration on Android handsets in the UK, representing a 72 per cent penetration, ahead of Google Maps, but behind Google Search. Engagement with Facebook and Twitter, given the penetration of their apps is an important way of promoting use of these services on mobiles.

We believe that an emerging lesson from our hyperlocal experiments is that social media can be important in increasing reach of a service and driving traffic to websites.

Recognising this, one of the volunteer-led Destination Local projects recommended that the function of content editor should be separate from that of social media co-ordinator, given that combining the role meant that insufficient time was given to engaging with social media users.

A significant proportion of traffic to hyperlocal websites appears to come from outside the geographic area of coverage

We have noted above that understanding the geographic reach of hyperlocal services should be acknowledged as an important tool for hyperlocal providers.

The performance of IP geolocation has improved over time, but measurement still remains more challenging for mobile devices, and traffic locations of users may differ depending on the exact telephone exchange they are connected to and which ISP they use. To assess use of services by geography this we chose two Destination Local websites as examples and analysed their visits over a three-month period according to the visitor's city, as reported by Google Analytics. Given that some localities are too small to register as individual places, locations for each visit were reported as occurring: 1) In the same county; 2) The same region (or nation for Northern Ireland, Wales or Scotland); 3) A neighbouring region (or devolved nation); 4) Being from the rest of the UK and 5) Being from outside the UK.⁴ Coding in this way allows for a consistent approach to be taken between projects, though we note that some locations (such as Welshpool) are close to regional/nations borders, meaning a significant proportion of local traffic may be reported as coming from another region.

For the two websites analysed, the majority of visits did not come from locations within the same county as the coverage of the service.

In the US, local advertising firm Borrell Associates found that in 2011, looking at a range of local websites, over 30 per cent of site visitors were from outside the local area.⁵

Average session length is highest on a regularly refreshed video-rich service

In addition to looking at measures which reflect the reach of hyperlocal, we also collected data relating to measures of impact, i.e. how users engage with the hyperlocal services. We also look at impact more broadly in Section 8.

The longest average length of visit was for one of the sites which is relatively video-rich, using a 304 minute bulletin format. This suggests that video can be used to hold and retain attention. In general, the average length of a visit from a mobile device was shorter than from a desktop device, reflecting that users tend to consume content differently on mobile than from desktop devices. In the diagram above, websites 1–3 have a coverage area with a population which exceeds 25,000, websites 4–6 are those serving an area with 25,000 people or less.

We also examined the average visit duration for the projects which launched in the period of Destination Local. For some services these varied considerably on a month-to-month basis, and some services reported very high average session lengths compared to those of the more established service (almost 18 minutes in one case).

This data implies that where resources allow it, hyperlocal services should consider the cost/benefits of using audio and video content to enhance their service and thereby increase consumption.

Ensuring that websites work well with mobile devices is important

The use of mobile devices to access the Destination Local services suggests that where websites are made 'mobile friendly', there is an appetite for users to access hyperlocal content via mobile devices, mirroring broader trends in the consumption of online content.

Overall, between 35 per cent and 45 per cent of visits to Destination Local projects have come from mobile devices including smartphones and tablets. Websites 5, 6 and 7 cover rural areas – their mobile penetration does not appear to be lower than that of the services based in rural areas. Interestingly, the services reporting the highest and lowest proportion of visits from mobile devices were mobile friendly HTML5 apps (websites 9 and 10).

While HTML5 services can be accessed by a range of mobile and desktop-based devices, mobile apps are much more restricted in terms of the number of devices through which they can be accessed.

And given that many services have launched since January 2013, and others launched not long before the start of Destination Local, it has not been possible to obtain sufficient time series data to look at the trend over a significant time series. However, for those services for which we have time sufficient series data, there has been a clear upward trend in the proportion of visits for both urban and rural sites in the past two years.

Given these findings, and our consumer research which shows that getting a smartphone or tablet has been a factor behind the increased use of hyperlocal media by those who have increased their use of it over the past two years, we believe that it is imperative that providers ensure that their services can be easily accessed on mobile devices.

6. App and Service Development and Technology

This section considers some of the pros and cons of some specific mobile technologies. Understanding the potential and impact of modern mobile technology for hyperlocal media is a key outcome for our practical work in Destination Local, and how the projects intended to use these technologies formed key criteria for the projects we selected for funding.

These technologies include the use of geolocation (via the phone's on-board GPS receiver), HTML5 (which allows dynamic, interactive, content-rich websites which adapt to the desktop and mobile devices), and augmented reality, which in the context of a hyperlocal service, may combine use of GPS with on-board compasses and the device's built-in camera to provide an interactive overlay of geographically relevant content on an image of the immediate locality.

We selected a range of projects using different mobile technologies, for example, Wi-Fi infrastructure (21VC), augmented reality (LocalSay) as well as a mixture of projects some of which use HTML5 but no native app (Kentishtowner and OurTown and the MyTown apps) those developing apps solely for smartphones/tablets (Locali and LocalSay) and others which developed both a website and app presence (Hebe Media, Papur Dre and URTV).

Native apps were generally more costly and took longer to develop than the HTML5 services

Experimenting widely with a range of business models meant that Phase 1 of Destination Local included the creation of new mobile-enabled services from scratch, as well as the upgrade and transformation of existing hyperlocal websites.

The development costs for apps tended to be in the region of £15,000 to £35,000, reflecting differences in the level of complexity in the app and back-end systems as well as the way in which the app had been developed. On average, development of the apps took three months, though this was substantially longer in some cases with some of the more technically complex apps or where the development team was small.

The development costs associated with websites and HTML5 services tended to be lower – ranging from under £5,000, to just over £25,000, again reflecting differences in complexity. One particularly complex HTML5 service was considerably more expensive to develop, at around £80,000. Development time for most of these services tended to be in the region of two to three months.

HTML5 runs on a wide range of devices and does not require approval from platform operators but some see apps as being more flexible and more in line with consumer demand

Several projects reported that they had had internal discussions and debates about whether or not to develop mobile apps instead of or in addition to, an HTML5 service. Some projects said that they did not feel that HTML5 was currently capable of providing an experience akin to that of a native app, while others believed that HTML5's performance is adequate for this, and that it is already a workable alternative to apps.

From my perspective the HTML5 web application provides us with the very widest way of reaching mobiles, whereas native apps have to be commissioned for all operating systems, and within that, different versions of the same operating system.” (MyTown Media).

Recognition of an app versus a modern mobile website was something which we had to consider when developing the survey for our consumer research as it may be difficult for some users to distinguish between modern HTML5 services and apps, in particular if the HTML5 service is launched via a short-cut from the phone's screen.

However, some consumers appear to want apps in addition to an HTML5 service, as noted by the OurTown team in their Destination Local blog post. Therefore, as services launch consumers may come to expect mobile apps.

People love apps... Despite there being little practical difference between our 'web app' and a conventional 'native' app, the single most common question we had from users has been: 'so when are you going to have an app?” (OurTown)

One project (Our Town) decided to adopt HTML5 rather than develop an app after their market research showed that while smartphone penetration in their area was relatively high, take-up of data plans was low, meaning that it was important to have a service which works equally well on desktop and notebook computers as on smartphones and tablets.

Several projects had to use or modify plug-ins to achieve their project goals. For example, one of the projects found that the absence of Welsh language support required development and modification of plug-ins to ensure that the entire service could be provided in Welsh. Another project was able to use a plug-in to offer push notification on an HTML5 service – something which historically only mobile apps have been able to offer. One developer told us that although their use of a rapid application development tool helped to speed up development, its frequent upgrade cycle sometimes resulted in incompatibilities with plug-ins, delaying the project by several weeks.

App store approval processes were also a factor behind the development choices of some projects. Some felt that the requirement to submit apps in advance to app store owners for approval might impact their flexibility in rolling out new versions of the service. Another project cited concerns around policies which govern in-app purchases and revenue share.

The debate between proponents of HTML5 and those who champion native apps is not confined to hyperlocal media. The Financial Times migrated from a native app to HTML5 in 2011 and the FT.com Managing Director recently stated “I challenge anyone to tell the difference between our HTML5 app and a native app. There is no drawback to working in HTML5, and there are lots of advantages.”⁶ arguing that the adoption of HTML5 mitigates having to develop and maintain multiple versions of apps. However, the fact that there are now 800,000 apps on Apple's App store, and 600,000 on Google Play⁷ suggest that for many developers, apps remain an important way of packaging content to consumers.

We feel that from the evidence of the ten projects, there are cost and time benefits in optimising a hyperlocal service for mobile by using HTML5 as the default solution.

To date there is limited integration of Destination Local projects into point of sales

The Local Edge apps in Leith and Broughton uses QR codes at the point of sale to issue reward 'stamps' to users who shop at local businesses. These are stored centrally and when stamps are redeemed, the merchant just presses a button on the app running on the customer's smartphone. This system avoids the need for a point of sale terminal, and the use of QR codes rather than NFC means that it can be used without merchant equipment and on non-NFC phones. However, another project decided not to progress with reader offers via QR codes after their trial suggested to them that few people actually scanned the codes.

A point of sale discount voucher scheme is also a part of the OurTown service in Johnstone. This system allows merchants to redeem vouchers by putting a unique code into a website.

One project (Papur Dre) allows for in-app and on website purchases for the digital copy of the printed newspaper. To do that, the service is required to integrate with a payment service (Go Cardless) for the processing of payments. They found that some payment services were unable to offer service to them due to the fact that they are neither a charity nor a company. This is a situation common to many not-for-profit community organisations which may have grown up without a distinct legal entity of their own.

More broadly, there has been interest in the investment community, especially in the US, in developing apps and devices that allow smartphones, tablets and other connected devices to be used by merchants at the point of sale.⁸ These allow for sophisticated POS terminals at a fraction of the price of dedicated devices, and allow small businesses to accept mobile and online payment systems and may also integrate with loyalty and discount schemes.

Given that our research shows that it is small, independent, locally-owned businesses which are those advertising on hyperlocal media, it is important that any integration between advertising and point of sale promotion reflects this in terms of the cost and complexity of point of sales equipment. However, by linking hyperlocal advertising and promotion to the point of sale through the use of check-ins or vouchers, small merchants may be able to gain a better understanding of the customer base, allowing for more effective advertising and marketing in future.⁹ While the integration of location-based or hyperlocal advertising and the point of sale remains nascent for many small businesses, there may be exciting opportunities for hyperlocal content providers to get involved and this may provide an additional way of promoting hyperlocal services to audiences.

The use of free and open-source software (FOSS) allowed developers to use their existing skill sets and use high-performance tools with minimal outlay

The technical expertise of the projects varied – in several cases organisations which create hyperlocal content had partnered with app developers for the Destination Local projects. As the projects mature, these hyperlocal organisations may take on responsibility for management, maintenance and upgrade of the services themselves. An alternative is to have the app development owning an equity stake in the software platform itself, promoting a long-term relationship between third-party developers and content producers.

We note that some projects were delayed due to problems developing the app, and in one case the project was delayed as the party contracted to develop the app's front end and user interface withdrew, requiring the hyperlocal content producer and the app developer to find another supplier.

In contrast, in another project, the team producing content told us that they felt they had enough expertise to carry out day-to-day maintenance of their platform. Where more in-depth knowledge was needed (for example to add new functionality to their service), they bought in the services of a freelance web developer on an ad hoc basis.

Both PhoneGap and the MIT App Inventor are open-source software and free to download and use without further royalty payments. Free and Open Source Software (FOSS) was used by other projects, for example, Herald and Times Group reported using Linux, Apache, MySQL and PHP (LAMP) in the development of OurTown. Other open-source software used by the projects includes Django (a rapid web application development tool for the Python language) and Nginx (a lightweight and high-performance web server). Interestingly, Django was developed for use at a US local daily newspaper. Django is used for hyperlocal services elsewhere, for example Openblock¹⁰ (an open-source hyperlocal news platform).

FOSS has been adapted by major corporations (including international media groups), and is used by them for the development and hosting of online service. The availability of powerful and tested development tools with large networks of peer technical support has enabled smaller developers, including the not-for-profit sector to access these same high-end tools.

In addition to the practical advantages offered by the FOSS, for some hyperlocal organisations (in particular those in the not for profit sector), there may be an added incentive to use it given a possible sense of shared values with the open-source community.

This suggests that the FOSS movement has played (and may continue to play) an important role in making available cheap and powerful tools and platforms for the development of hyperlocal media services.

Where budgets allow, projects should be taking steps to promote the technical sustainability of their services by ensuring that there is a member of staff with at least a basic level of expertise who is able to carry out day-to-day maintenance.

7. Sustainability

Introduction

Sustainability is a key issue for all organisations, including those who create and distribute hyperlocal content and has been a key concern for practitioners and policymakers.

Sustainability is not just about financial performance and profitability – in fact, for some hyperlocal organisations, sustainability may be more about ensuring that enough volunteers have been recruited and that those volunteers are not being overworked. We recognise that all of the projects are concerned with non-financial outcomes, such as better engagement with citizens, holding authority to account or creating better local communities.

However, for commercial enterprises, financial performance, profitability and cash generation are at the heart of the any assessment of sustainability.

As many of these prototype services have only recently launched, we are unable to provide a fully formed view as to the sustainability of the business models being tested by the projects. However at this stage, our findings suggest that stand-alone, independent, commercial, hyperlocal media services are finding profit making to be challenging.

At the time of writing, none of the prototype activities by the ten organisations have been able to make a profit on a stand-alone basis. This reinforces our original research that suggests that the hyperlocal sector is currently fragile, with no one business model established. Indeed, several of the organisations that Destination Local supported are experimenting with a number of business models to crack the challenge of profitability and the key success factor for the most profitable is the energetic entrepreneurialism of the publishers.

These findings are consistent with the broader literature and discussions about the hyperlocal sector, both in the UK and internationally.

Advertising revenues were sought by most of the ten services, but selling online advertising to businesses can be difficult

The majority of Destination Local projects see advertising as a source of revenue. There is relatively limited use made of dedicated local advertising networks, though some services (such as 21VC) do use Addiply.¹¹ One site owner said that they had tried to use local advertising networks but found it was not worth their while as they still needed to approach local businesses to let them know about their website and of the opportunity to advertise. This experience is consistent with the findings of our survey of small businesses which show that only a minority of the businesses were aware of hyperlocal websites in their area (25 per cent overall), and only a small proportion of these (6 per cent) actually advertised on them.

Many local businesses appear relatively unsophisticated in their approach to online advertising – many are seeking a similar experience to that seen in print

In general, the projects told us that local businesses are relatively unsophisticated in approach to online advertising and are most comfortable with online advertising when it is similar to print advertising. This has influenced the pricing of adverts, which tends to be a fixed amount per advert per month for a given advert size, rather than being priced on cost per click (CPC) or cost per impression (CPM) basis. Several projects reported that their advertisers were concerned at the use of advert rotation by the sites, which cycles adverts and enables sites to carry advertising from a larger number of businesses. Following from print, small advertisers instead expected to see their advert displayed each time the relevant page loaded.

Several Destination Local services have launched business directory type services or small boxed classified adverts – these provide a print-like, low-level entry product which is similar to printed classified adverts. To sell in an economic manner, several projects were seeking to develop their own back-end, self-service options to keep costs down. A challenge was to ensure that these back-end systems were easy to use for small businesses, which often have limited IT and design skills.

Some projects told us that they saw opportunities for targeting users by interest, based on the channels they had signed up to in their service. They believe that this will be appealing to businesses that may be more willing to pay a premium for this type of service.

Print advertising retains its attractiveness for many advertisers – and profitability for hyperlocal services

While Nesta's definition of hyperlocal media is specifically online,¹² and the focus of the Destination Local funding has been around the use of mobile technology, we recognise that printed media continues to play a role as a source of hyperlocal news and information.

During the Destination Local project, two organisations (the Kentishtowner and Hebe's the City Talking) started to publish printed editions of their service. This is an example of a wider trend which has seen other hitherto online-only hyperlocal media organisations launch printed editions.¹³

In addition, several other Destination Local projects have grown out of printed media. For example 21VC/Loddon Eye which has grown out from a parish magazine, Papur Dre (a Welsh language community newspaper) and the Broughton Spurtle, a printed community newspaper for Broughton in Edinburgh.

Several projects told us that it is much easier to sell advertising in the printed edition of their service than on their website and app, so much so that in some cases they have had to turn advertisers away from the printed edition. Interestingly, it is not clear whether these adverts were then also published on the website – suggesting they are specifically seeking a print outlet. One project noted their ability to attract a major brand to advertise was solely dependent on their printed edition – the major brand was not interested in advertising on their website. The experience of many of these projects echoes the findings of our survey of small businesses which demonstrates a higher propensity to advertise at the hyperlocal level in print than online.

In short, at the present time, printed editions appear be more profitable than websites for some services, and thus may act as a subsidy. This does assume that printing and distribution costs are kept low – for example by using newsprint, rather than glossy magazine stock, and by using local shops and cafes and bars as distribution points. The role of print media in promoting sustainability for hyperlocal media should therefore not be ignored.

It should be noted that the experience some of the projects have had with printed publications cannot really be seen to be indicative of the general health of traditional printed newspapers. The printed editions from the Destination Local projects are typically one of two types, neither of which is likely to compete with traditional printed newspapers,

Footnotes

The PDF document does not contain a Table of Contents page. Therefore, the markdown output will begin directly with the first content section.

DESTINATION LOCAL: OUR LESSONS TO DATE

An interim report on Nesta's Hyperlocal Media Programme

even those operating at a small geographic scale. Printed editions of Papur Dre and the Broughton Spurtle can be seen as traditional community newspapers, created by, and for people living in the area, providing a range of community focussed content. In contrast, publications such as City Talking and the Kentishtowner are published monthly and tend to circulate in a wider geographic area and have a particular content focus on fashion, culture and going out rather than on what some have described as ‘hard news’. They also target a particular demographic, younger than that of traditional newspaper readers. Neither type of publication carries the cost of maintaining a traditional newsroom with dozens of journalists associated with traditional newspaper production.

Think carefully about the cost/benefits of hiring dedicated advertising sales staff

Several projects reported recruiting for a member of staff to lead on advertising sales between July 2012 and June 2013, as the owners of some services felt that their time and skills were better suited to content creation, service development and maintenance. The inability to devote more resources to time-consuming face-to-face visits with local businesses was also seen by these projects as a stumbling block to further revenue growth. Recruitment of dedicated advertising sales resource was seen as a major decision by those projects given the substantial increase in fixed costs (an additional several thousand pounds each month) which result from having to pay an additional salary.²³ Several projects reported difficulties in recruiting the right candidate for the role.

The issue of the most cost-effective way of generating advertising revenue is one of the most debated topics in online local media. Lowering cost of sales and transaction costs remain vital goals, not just for hyperlocal providers, but also for the traditional local press. Services such as Addiply give hyperlocal providers one outlet for selling space, but there is little public data available on the success of advertising platforms.

Only one project reported financial donations as a source of revenue

There are limited instances of projects seeking revenues from other sources, including user subscriptions, consultancy donations and grants (beyond the Nesta Destination Local grant).

Only one project (Papur Dre) is seeking to monetise their online service through a subscription model. This is consistent with the findings from our consumer research which suggests that there is only a very limited willingness to pay for hyperlocal content - though the printed edition of the newspaper is paid for rather than being fully advertising-funded, in line with many other papurau bro.

For some organisations which produce hyperlocal content in fulfilment of their broader objectives, it may be difficult to fully ascribe revenues to the hyperlocal service per se. Donations to the organisation may therefore not be a reflection of the willingness to pay for hyperlocal media

Other sources of revenue reported by some projects include grants and public money. For local authority apps, it is possible that these might in future be funded through the councils' communications budget. In this instance, the business case may be more about cost savings in an essential function at the heart of local government, than as a way to raise revenue.

While our consumer survey shows that there is limited willingness to pay for hyperlocal content in the UK, in France J'aime l'info²⁴ provides a service which allows people to donate to online news services. Services registered with the site include regional/departmental services as a well citywide hyperlocal websites such as the Rue 89 for Lyon and Strasbourg. Services covering smaller geographies are also on the site, such as Dixneufinfo.com,¹ a hyperlocal service covering the 19th arrondissement of Paris. However, it appears that not all listed websites have received donations, despite being on the site for a year or more.

The ongoing cost of content is the biggest spend for hyperlocal providers. Technology spend is relatively cheap in comparison

The largest cost item for the year for the majority of services is content creation, which for the majority of projects was generated in-house. Some services told us that they also on occasion used freelance labour to create content for their services. For those services which have been creating content from scratch throughout the year, staff costs are approximately 70-90 per cent of total reported operating costs over the period. This suggests that some businesses keep other overheads low, for example, by working from home, rather than maintaining offices, though some projects were based in city centre offices, resulting in significantly higher expenditure on overheads.

The cost structure of the projects varies considerably, mainly reflecting that commercially run services incur significant staff costs which are avoided by those operating on a volunteer basis. We note that it is unlikely that those not-for-profit models reporting a surplus would be doing so if they had to pay salaries based on the average wage rates. In other words the value of volunteer labour is a significant hidden subsidy/income to several of the Destination Local projects.

Assuming an average wage of £11.07 hour,² the value of volunteer labour varied from around £100 per month (for a single person working an average of two hours a week to create content for the service) through to a volunteer-run project, which reported volunteer working the equivalent of half a full-time employee. Assuming half the median full-time weekly wage of £400 (i.e. £200 per week), this translates into a monthly cost of around £900 a month - several times greater than its monthly surplus. Volunteers were also used by some of the commercial projects.

We estimate that content creation costs accounts for around half of the ongoing operational cost of a commercial hyperlocal news service

Our cost estimates based on the financial returns submitted to us by the projects suggest that around half of the ongoing operational costs associated with a commercial hyperlocal media service relate to the cost of creating content. This reflects the labour intensive native nature of content creation.

Reported ongoing platform costs are generally low for both commercial and not-for-profit services

In contrast, platform and hosting costs are generally low across the prototypes, projects reporting operating expenditure of between approximately £60 and £8,400 on a yearly basis. The higher number includes a service with an HTML5 element.

Those services which had developed apps which required an app server generally reported higher annual hosting costs. However, the actual cost of the hosting element from a third-party provider such as Amazon is low. We believe that for up to around 10,000 users, only a single virtual server is required and these can be rented for significantly less than £1,000 a year.

Hyperlocal media costs compare favourably to those of traditional media focussed at the town, village and/or postcode level

The wide range of annual costs reflects the range of business models and content type which can be delivered via an online service. From our discussions with a range of projects, both from within Destination Local and outside, we believe that it is possible to run a video-rich hyperlocal news service for around a minimum of £30,000 a year. The cost of a service increases significantly as additional staff are employed to increase operational resilience and allow for greater resource to be added for advertising sales.