 Water. At times it's scarce, difficult to find and not what it seems. At other times we can be overwhelmed by it. Data to social researchers is like water. Historically data collection processes such as running surveys have been time-consuming and expensive with challenges like low response rates and uncertainties around understanding the representativeness of data. Times have changed. Now we are inundated with data on human behaviour and social networks from myriad sources, transactions, bureaucratic records, computer and smartphone use, geospatial and travel information. If harnessed and queried in the right way, we can start thinking big, presenting boundless, invaluable resources for social researchers. Smart energy meters for example, which the government plans to install in every home by 2020, will provide massive amounts of real-time data on energy usage from 26 million homes relevant to budgeting, fuel poverty and carbon emissions. Where once we feared a data drought, we now fear a flood. The UK data service is investing in a big data solution. As a trusted national digital data service that already curates and provides access to the UK's National Social Science data assets for research, teaching, skills development and policymaking, we are well placed to support research intensive work using big data. State of the art technologies now allow us to organise and curate big data and to channel a data deluge into a digital repository, a charted, bounded and secure data lake. Our data lake is built on the open data platform model, using open source software to provide a system for secure storage, management, discovery and analysis of any type of data, as well as scaling and linkage from smaller sources to big data. The data lake stores data in its pure raw formats. Careful charting of data sources through documenting them allows a thorough understanding of where data have come from. We care about security and about transparency and we prioritise these. We use the FiveSafes framework to manage any disclosive personal data that we are authorised to hold. FiveSafes is an established and trusted protocol induced by research data centres like ourselves. We want researchers to have a meaningful encounter with big data in a way that demystifies the sometimes vague promises of a data panacea. We provide a sandbox and guided training to help researchers experience working with real data sources. And using dashboards with easy to understand analytics, researchers can navigate their journey more easily to reach cleansed, channeled data streams. Instead of fearing a deluge of data, researchers can plot a course to include any data available in the data lake as part of their analysis. A researcher interested in UK fuel poverty could draw on energy readings from smart meters, housing stock information from energy performance certificates, as well as having ready access to relevant survey data. Our aim is to use technology to unlock the value big data brings to social science research. Come and join us on our exploration voyage.