Labours landslide and the lessons on data

The 2024 general election is a case study in the weird and wonderful world of data

Labours landslide and the lessons on data

The predications were correct. We have a new government who have barnstormed their way into one of the most impressive political turnarounds of all time. Or have they? 

Well, the answer to that depends on what data you use and what test you apply to measure the scale of success. 

In the UK our democracy is based on a first past the post system, where Seats are contested in local areas where each seat equals a Member of Parliament. It makes no odds whether you win by one vote, or 50,000, a win is a win. Once the magic number is reached, 326, hey presto, you have a majority and can claim victory.

Just a reminder, the 3 largest parties’ results were as follows:

Labour: 411
Conservative: 121
Liberal Democrats: 72
Others: 46

Source: https://commonslibrary.parliament.uk/research-briefings/cbp-10009/ 

There are however alternative approaches to democracy. These can take various forms but Proportional Representation for example looks at the share of the vote nationally. 

Looking at this approach the results were as follows:

Labour: 33.7%
Conservative: 23.7%
Liberal Democrats: 12.2%
Others: 30.4%

Source: https://www.statista.com/statistics/717004/general-elections-vote-share-by-party-uk/ 

The challenge is that one set of data is reported in seats won, the other as a % of the vote. If we convert the seats won to an effective %, this makes the differences in approach easier to see. 

Top 3 PartiesElected Seats as %Vote Share %Difference
Labour63.2%33.7%GAIN 29.5%
Conservative18.6%23.7%LOSS 5.1%
Liberal Democrats11%12.2%LOSS 0.2%
Others7%30.4%LOSS 23.4%

One method suggests an extraordinary result that belongs in the pages of history, the other, not so much. Labour secured 63.2% of the seats with a 33.7% share in votes. Smaller parties (Others) achieved 7% of the seats with 30% of the votes. 

One could draw the conclusion the make up of our Parliament in this particular election does not reflect the votes cast. The Liberal Democrats have the closest outcome between the two methods, followed by the Conversatives but when you consider Labour and Other, smaller parties, the result is different. Despite receiving over 30% of the votes cast, only 7% of the seats are held by smaller parties. 

Whatever emotion this result stirs in you, there is a chasm between the two approaches of democracy and the strength of power given. 

Consider for a moment, that sort of difference in decision making without your own Franchise. A data driven approach, say for example, how to evolve a brand pricing strategy, has the potential to come up with different answers depending on how you use what you have. 

As brands embark on data driven strategies to growth, there is always a risk that by reading the data in the wrong way, or not fully appreciating why it appears as it does, management may make the wrong decisions. 

To avoid this, these are the key things you should consider long before you embark on a project:

  1. Definitions – at the earliest opportunity it is imperative to define everything you want to measure in simple terms. What is a Sale? What is an Order? Is it when the money arrives? Is it when a contract is signed? Is it when a service or product is delivered? 
  2. Document – write your definitions and approach to data in such a way that all teams can access it easily to ensure the same approach is being followed. 
  3. Consistency – it is important to measure on a consistent basis to give you the best possible chance of identifying change. Going back to our election example if the first election was First Past the Post and the second was on Proportional Representation, comparing relative performance would be highly misleading
  4. Rubbish In, Rubbish Out – if the data you use to help inform you is not in tip top condition this is going to increase the risk of a misstep, the big offenders are old or incomplete data that will give you a misleading picture 
  5. KISS – keep it simple stupid. Data can be confusing and complex at the best of time and there is a huge risk in over measuring and over analysing. Even the most sophisticated of operators could do well to simplify. Pick a handful of metrics, measure them and test out some adjustments. My top tip is the start with the holy trinity: Revenue in £, Orders, Debt Management – getting these factors understood in your networks is the key to long term sustainable growth
  6. Why – curiosity is your friend, don’t be afraid to ask your teams this one question. If the marketing team declare a particular activity a roaring success, why is that? They might say, sales have gone up or orders have increase but what evidence can you bring together to support that. Understanding why something went well is just as important as understanding what needs to improved upon. 

Transitioning to a data driven strategy is not easy, however if you can embrace these six key lessons I learnt on the way, you will massively increase the chances of success.  

ABOUT THE AUTHOR
Theo Millward
Theo Millward
RELATED ARTICLES