@Jess_OVO the original post came about after me noticing in MYOVO consumption that my consumption had almost doubled.

It was after that when I checked the readings and saw that jump on the meter itself.

1. According to MYOVO readings, the last officially sent reading was 14/07 = 55304
2. By the time of my checking the meter after noticing the above on 20/07 = 57884
3. According to MYOVO a manual reading on the 27/07 = 55493

Something is obviously askew somewhere as the reading on 27/07 is significantly lower than the reading on my meter on 20/07 and I’m confused by that reading (55493) as I dont recall having submitted it.

The manual reading in MYOVO does not fit with any reading I recorded with photographs for 27/07 date. In fact the readings I have recorded show values of 57987 at the end of that day.

Silly question, am I reading the meter correctly:

1 press 9

2 display kWh

3 record value

4 that value is the consumtion

I have mentioned before that the RED LED’s seem to flash in strange patterns at different times when I check the meter and NOTHING unusual in terms of consumption is going on.

Is it me?

Hi @fatbloke88 - some more great questions to help us get to the bottom of things.

In answer to your first question, how you take a reading will depend on which smart meter you’ve got there. Our Secure S1 meters are read by pressing the number ‘9’ button but obviously this does vary depending on which meter model you’ve got installed. Are you able to take any photos of the meter and post them here? This really helps us see exactly what you’ve got and give the best advice relating to your meter.

In terms of the lights that are flashing - are these labelled WAN and HAN? These lights aren’t based on your consumption but they should flash to show the communication of the meter (both to the IHD and to us). You can find out more about what the flashing meter lights mean on this related topic -

@Jess_OVO@Blastoise186@Transparent@Tim_OVO 

Hi All been a while I know but I thought I’d keep this updated. The current situation is “the meter team are aware” (whatever that means) and my meter is clocking a average of 13.70 kwh pd for the past 40 days. I did try to register an up to date reading on the site and it flagged up a warning that the numebers didn’t look right and the today it just accepted it???

Our gas meter is sending accurate readings  and is immediately adjacent to our electicity meter.

Has anyone seen or heard anything that might why our meter jumped 2500Kwh in just 6 days (that’s why I came here) at all?

Does anyone know what the “Meter team are aware” actually means, what are they doing, how are they doing it, how do I know what they are doing is right and accurate?

We were given an excess usage bill (>£400) at year end and we’re deeply concerned now that this also was inaccurate but dont know how to challenge it. Anyone suggest anything?

Finally our DD is astrononmical (>£200mth) and I can’t seem to get it down which is concerning. “MY OVO” advises that I’m set to be £92appx in credit at year end and I can now reduce by a few £??? Yet I’m currently over £400 in credit!! ( yes I know we havent been through winter yet...)

update after inputting the meter read above it now tells me that I’m underpaying and I’ll be in deficit by £500 at year end and to increase DD to over £300pm!!!!

Thanks

EEEEEEEEEKkkkkkkkkkkk!!!!!

What is going on here?

What I find ‘interesting’ about this situation are the following:

a: that a SMETS1 Smart Meter can ‘overclock’ at all. This suggests a fundamental flaw in its design. It may not necessarily be within the software. It could equally well be the current-detector.

But these meters have been through cyber-security scrutiny to ensure that they can be sited in ‘unguarded’ meter cabinets without threat to the country’s energy supply system. So someone should be keen to analyse this particular meter and ensure that it hasn’t been adversely affected by any external influence. That could be anything from a magnetic field to a software hack.

b: that there’s no software ‘sanity check’ being done by OVO to verify if any unusual usage exists.

This is relatively easy to do. After all, most sudden changes in energy consumption will be due to a particular device being connected. I bought an EV a few months ago which has increased my base-line electricity usage by 2½ times whilst it’s charging. But it obviously doesn’t charge 24hrs/day!

A sensible sanity check on @fatbloke88’s usage should easily be able to show that there must be an ‘energy leakage’ somewhere!

After all, if this was an errant gas meter, such a constant jump in supply would indicate a leak that needs immediate attention.

Any such energy usage anomaly should generate an alert from the Billing System software that gets passed immediately to the Meter Engineering Team!

c: What would be the case if the situation is reversed?

What if there are a hundred meters out there which are sending erroneous data, but the customers haven’t noticed?

Would it make any difference to OVO if the meters were under-clocking and the company was losing money?

d: We discussed the proposed Betelgeuse option nine months ago. Such a method of displaying the usage-data would be more capable of showing up such a faulty meter.

Betelgeuse presentation extrapolates the energy-usage forward to verify that the account will be in credit at the end of the contract. So an over-clocking meter would have a profound effect on the end of the curve, thereby alerting the customer and OVO that something had suddenly failed at a particular point in time.

It was that topic which resulted in the formation of a Tech Treehouse in which software discussions concerning the Billing System could be fed back directly to the development team. What’s happened with all that feedback?!

Has Betelgeuse entered a black hole?

@fatbloke88 wrote:

EEEEEEEEEKkkkkkkkkkkk!!!!!

What is going on here?

Possibilities include:

1: someone has tapped into your supply and is now running their house on your meter

2: faulty meter

3: cannabis farm

currently waiting to speak to someone on CS I need to get a formal complaint registered about this and make sure I am not billed  for stuff I’m not using

I suggest that a Complaint is always sent by email (hello@ovoenergy.com), rather than left to a voice conversation. If it should get as far as the Office of the Energy Ombudsman (which I doubt!), then you require evidence of when the complaint was started.

That’s a separate issue to my involvement. I’m wanting to see the in-house software changed to prevent such errors going un-noticed. And remember, I’ve been asking for this months before you turned up with an illustration of why it’s needed!

Thanks for the updates you posted today and over the weekend, @fatbloke88. Given the big numbers you’ve outlined, our Support Team will be very understanding of you wanting clarity. As they’ve mentioned this already being assigned to a team, there’s a chance you’ll need to give them some time to get to this, but yes a web chat should be able to confirm this. 

EEEEEEEEEKkkkkkkkkkkk!!!!!

What is going on here?

 

It does look like a spike in usage, @Simon1D - sorry for pointing out the obvious. For the sake of a comparison, what you were hitting for most of July is closer to what I am using now. So this jump that took place on the 27th July, has anything changed since then? You can refer to Jess’s guide at the topic for your options for diagnosing this. 

@Transparent the Direct Debit calculator does this pretty much, I mean, it’s pretty similar wouldn’t you say?:

You can visualize the payment amounts, split out per fuel or combined, you can see how that compares to payment amounts etc over time...

EEEEEEEEEKkkkkkkkkkkk!!!!!

What is going on here?

 

 

It does look like a spike in usage, @Simon1D - sorry for pointing out the obvious.

There’s a lot going on at the moment, but I wonder if @Tim_OVO meant to tag in someone else here, rather than me? However ...

That repeating weekly pattern of small variations in daily usage looks ominously familiar to me. I saw such absolute nonsense for a while on a SMETS1 meter (I don’t live there anymore, so that account is closed and I can’t look up the old data so easily now - Edit: but I found this old post in the Tech Treehouse, a private area, that includes a screenshot ) in the period before it became SMETS1+ and registered usage daily instead of me having to submit manual readings monthly. But at least I never saw a step change like this.

I try to keep negative thoughts to myself, but the flood of problems currently being reported in the forum by Ovo users does make me wonder if I’ll still be with Ovo a couple of months from now.

That’ll be after a little more than 10 years :-|

I found myself re-reading some of the thread following the post to which I just linked. It’s sad, but I wonder if we were within an ace of getting the Ovo technical people to explain themselves, here (apologies for linking to the Tech Treehouse, a closed part of the forum) but the key quote, where Tim is relaying a comment from them, is

With hindsight, I regret not latching on to that offer. I wonder, @Tim_OVO - do you think we might get more info? (if only in the privacy of the Tech Treehouse).

Instead I launched into an explanation of how I thought estimating should be done. Bad move by me. Focus. I need to focus more, and better...

@Tim_OVOwrote:

the Direct Debit calculator does this pretty much, I mean, it’s pretty similar wouldn’t you say?:

Well the data is there just the same. But it produces no response/feedback which prompts OVO to do anything!

UPDATE: apparently at some point, someone somewhere registered ours as a dual rate meter…. bear with me;

That means when I press “9” on the meter I was getting total consumption from both Day & Night registers even though there is no night register… ok so far?

Now when they advise me to press ”6” and look at a range of data there is a reading in there of kwh which is more in line with what is expected

Using this data I get an ave consumpotion of 16 kwh/pd since mid July (I think I’m ok with that)…

Here’s where it gets confusing;

last year Mar20 - Feb21 they tell me the meter reading which were being sent automatically showed a consumtion ave  995 kwh /pm hence the reason Iwe were asked to pay a large end of year settlement value.

I am now confused, because NOTHING in our consumtion use has changed in the last 17 months yet our consumption pattern was well over double the volume that we are now using and since I’ve been keeping an irregular eye on…

Just going to have to keep a wary eye as we go through winter..

Glad to hear our Support Team managed to get to the bottom of things in terms of the difference between readings you were taking manually and those being received automatically from the meter, @fatbloke88 .

If there were previously 2 registers on the meter (even if one is no longer active) there’s a slightly different method for taking readings. It could be that there were storage heaters in the property previously which have since been removed. Find out more about taking a smart meter reading manually here.

In terms of your usage increases over the past year, again without knowing exactly what’s been plugged in and using energy it’s a difficult one to answer. Not sure if it could just be world events which kicked off around March 2020 which might explain things? Now that you’re keeping an active eye on things going forward (and the best place to do this is the usage pages of your online account or OVO app - download for Android or iOS) you’ll be able to spot any inconsistencies much quicker.

Hope you’re able to see some more expected usage figures - as ever pop back here if you do need any further general help. Whilst we’re not able to access your account we may be able to advise of common issues or getting the best out of our online tools. 

Great to see that there’s a solution here. I don’t know why I never even thought of that one! I’ll try to remember this in case it ever affects anyone else in the future.

Admittedly, this is another reason why I’m not a huge fan of the overly complicated interface with the Secure Liberty meters (It’s definitely not the worst one ever though!). While having a full numeric keypad is useful for cases if you’re on PAYG and and you need to manually punch in a UTRN to top-up because it didn’t go through automatically, it serves little purpose for anything else - especially if you’re on Pay Monthly. And besides that, the majority of IHDs also have PPMID (Prepayment Meter Interface Device if memory serves) capability anyway that allows a customer to enter a UTRN code via the IHD and send it to the meter from there. Most of the other manufacturers have just two or three buttons instead which is a lot simpler and Aclara goes the extra mile by utilising the giant displays to clearly indicate exactly which register you’re looking at when taking readings - and does a pretty good job of using plain English to label them - rather than complicated short codes that confuse even me sometimes.

I’m still learning what all the buttons do on Secure Liberty meters and I can’t practice on Raichu because it’s an S2 Aclara. Luckily there’s a few guides around that I can read up on.

You’re far from the first person who has been caught out by those confusing buttons, and I doubt you’ll be the last. 

Hi all, latest info on the ongoing saga; after days of inputting what I’ve been advised are the correct consumption values by pressing “6” on my meter, this is where we are:

Readings submitted daily most days since 01/09

Average usage  13.42kwh/pd

Readings appear to be accepted

If the above statements are correct and I believe they are particularly the readings are being accepted, why is the estimated consumption still so high through the back end of July and all of August - see image;-

This is despite the readings in Sept 21 apparenly being accepted… With no justification that I can see the consumption estimated throughout Late July and Aug are not being amended.

I’m confused.

I’ve experienced a similar issue @fatbloke88 

From what I’ve seen of the way the Billing System works, there is only limited capability to revise the usage readings for previous days where estimates have been applied.

I think the ‘rule’ is this:

When correct readings are accepted, these are applied to the usage data going back in time until a day is reached when an actual reading was obtained.

I can’t confirm this, and I certainly wouldn’t condone it, but that appears to fit the facts such as I’ve seen them.

After that rule has been applied, any further amendments are effected by applying a one-off negative reading to bring the financial system into balance. There are two reasons for this:

a: it satisfies Ofgem regulations for the Billing System providing adequate invoices

b: it annoys the hell out of @Simon1D who is left trying to analyse historical usage containing negative gas usage

:-))

If only Ovo’s “rule” were as rational as you suggest, @Transparent … As you have previously noted yourself, Ovo are apparently no longer capable of making estimates that lead to usages that respect the "sum rule”. (I mean, estimates that ensure that the sum of usages between the two most recent actual readings is correct. None of the negative usages Ovo presented to me achieve that.)

TBH, I’ve long passed beyond annoyance and have reached a zen-like tranquillity, which comes from understanding, for myself, how to make optimal estimates (of readings and usages) in the absence of actual data from the smart meter.

Along the way I have learned enough about the process of estimation to have a good idea (so I believe, anyway) of what are the various mistakes and omissions that are probably leading Ovo to come up with such silly estimates. Including that bizarre weekly pattern in the data that @fatbloke88 just posted - I’ve seen that one too.

I believe that negative estimates of usage come about when Ovo’s model of seasonal variations in usage has a large difference, summer to winter, and the customer’s summer usage is low: if Ovo make the mistake of taking their model too seriously, their algorithm “thinks” that recent low usage is “too high”, and compensates for this by estimating a reading which is only possible if the most recent (missing) usage is negative. Whether one says the algorithm is stupid, or concludes that its author was careless, or somehow lacking insight into the problem they’re trying to solve, I don’t know (or, really, care). Perhaps the algorithm has Machine Learning at it’s core but, whatever the details, it boils down to Garbage In, Garbage Out.

If Ovo are interested to do better they will, at some point, be able to download some python scripts from my github area. Scripts that do the right thing, and can be used to demonstrate the robustness of good estimates to the deletion of readings and/or usage from a reference test dataset (my own data - it’s fairly complete, but it’s the work of moments to go in and modify it to remove a lot of intermediate smart readings and see how badly the estimates can go wrong. Or not.)

If Ovo aren’t interested, well, that would say it all. But don’t go looking for it yet, I’m still coding…

BW all

It was the issuing of incorrect bills and inaccurate estimates which landed OVO with a fine of £8.9m from Ofgem at the beginning of 2018.

As a consequence the Board of Directors decided to create their own new Billing System in-house, thereby removing the company’s future from the poor algorithms which beset the 3rd-party software.

So we might think that OVO would be highly motivated to be using Billing software which properly addresses incomplete, incorrect and estimated readings within customer’s usage records.

It therefore seems decidedly odd that the Board have now decided to distance Kaluza, who develop the Billing system code, from OVO Energy who use it to invoice us.

They’ve come full circle to the situation where OVO Energy is reliant on software from a ‘3rd party’ who will themselves not suffer from an Ofgem fine for incorrect and inaccurate billing.

The difference this time is that some of us customers here on the Forum have been busy isolating and reporting the bugs.

That leaves OVO with an option to avoid another fine from Ofgem, and instead to listen to its customers.

 Can I count the ways … in which the “invisible hand” can push organisations into patterns of behaviour that are not even in their own best interests?

No, I probably cannot.

But I know what I support, when given the chance.

Apparently, iirc, this isn’t even a “blame the shareholders” issue, because Ovo is not a publicly listed company.

Well, they can listen to us as customers, or they can browse github (where, incidentally, kaluza already post their own code, or so I assume - I’ve never had need to look at it).

Isn’t free software wonderful?

Hi everyone, 

Tim here your resident waffler and fact checker. Thanks for the update, @fatbloke88.

I want to make sure it’s clear here for you and anyone else that follows that we’re discussing the usage graphs, for a time period before you’ve been submitting actual readings.

The best way to get accurate billing has and always will be to ensure there’s actual meter readings on your energy account. Most of the time that means a normally functioning smart meter sending them automatically. Sometimes that means submitting meter readings manually. 

@Simon1D and @Transparent are referring to the usage sections, and I’ll be passing on some insight from Simon to the teams involved in this area on the online account. So far so good: Anything that might help us to visualise a member’s usage so they feel empowered to make energy reduction changes is a win for everyone. 

Here’s where things take a turn for the worse

This unfortunately is conjecture. See OVO’s statement on this fine here. OVO Group’s mission is to provide clean and affordable energy for all. Building that future ourselves is what we wanted. Kaluza (part of OVO Group) is doing just that. 

Likewise here I refer to my earlier comment about actual meter readings, billing, and the difference between that and your usage graph visualizations. 

This forum puts everything out in the open, for everyone to see. This is transparent, honest, and inclusive. If we all respond to that transparency with a relentless pursuit of accuracy when stating facts (which in itself is one of the house rules), these threads will be better off as a result. 

I apologise if I have provided an inaccurate précis of those historical billing events. The full details of the financial penalty are downloadable from Ofgem. The events date back as far as 2013 and were created when the earlier 3rd-party software was in use.

The issues referred to by Ofgem were no longer evident in accounts that used the new Billing System…. until @fatbloke88 started this topic.

I’ve described how two of us on the Forum have tried to go back through usage figures produced by the new Billing System in order to better understand how the estimated readings are being handled. I’ve found this impossibly complex, but @Simon1D is persisting.

The difference now is that @fatbloke88 has described a situation in which erroneous meter readings have indeed impacted actual bills.

Like @Tim_OVO suggested above, I had largely dismissed the possibility of such errors occurring with the new Billing System, written in-house. My assumption was based on meters yielding cumulative readings, thereby cancelling out previous erroneous estimates.

Based on what @fatbloke88 has now told us, my assumption wasn’t well founded. The errors in this case have resulted in actual bills higher that they should be, and also an amount just shy of £500 requested as a settlement for months of underpayment.

We don’t yet know why FatBloke’s meter readings jumped jumped 2500Kwh in just 6 days, nor if this could be accounted for by the tariff being incorrectly configured as a dual-rate meter.

Yes, Simon and I have been referring to visualised usage figures. But FatBloke’s fault-description in this topic is about actual billing and possible faulty metering.

We are still trying to get to the bottom of this.

FWIW, and without any knowledge of what’s really going on in the present case. I am now confident that I have plumbed the depths, sniffed around the Mariana Trench, so to speak. And am now in the process of putting together a working demonstration of how missing data should be handled when it comes to making estimates.

I’ve actually learned a fair bit in the process.

Cheers all

latest update, I’ve sent 2 formal complaint mails to Ovo and both have been responded to in general terms and one with a promise of a 45 day response….

I looked back at the consumption since I rejoined Ovo in April 2020 and have made the following observations and wonder if any of you guys can assist with understanding.

consumption between 01/04/20 - 30/03/21 = 13526.51

The read values omn a daily basis are weird fluctuating from “-254.89, yes minus” to a staggering 2526.1 on a single day!!! All of that within not much more than a week!

When I look more closely at the data I noticed that there is a difference in formatting of the date stamp, (remember this is taken from My OVO). Specifically during the period in question you can see that the date format coding is different on the negative consumption days, see below;

Is this in any way relevant, does it suggest a change of software on OVO’s part for reasons unknown, does it suggest a change of hardware or other change. It seems more than curious or coincidental..Thoughts ?

Next issue, as I review the readings I see consumption values of identical or near identical use across multiple days, this appears to be the case right from the start of the account  e.g.

Although the actual consumption value has altered across periods, the values across continuous days are identical or very close, again see below;

This pattern is repeated across multiple periods with variations in the actual consumption value noted agin see below;

Does this look in any way normal? Previous posts will indicate that our daily usage does vary between 13 - 18kWh/pd. I’m willing to accept a rise through the winter months but again a closer look at the values and patterns of usage below seems strange;

The consistency of consumption values just doesn’t seem correct. Can I ask what are your thoughts on this so far?

When I was advised about about the dual fuel tariff mix up, I was also advised that my meter had been sending readings consistently throughout all of this period, now that I’ve looked at it I’m honestly not convinced. @Blastoise186@Transparent  Is this worth raising to OFGEM at this point or am I missing something?

@Tim_OVO@Jess_OVO I realise you cant comment in detail on a public forum so I’d like to ask that you get your seniors to review your processes in relation to dealing with issues that are not fully resolved at the point of origin. Might be worth considering a single point of contact, a dedicated number (pref direct dial) to the nominated point of contact. may be worth aligning processes and staff to respond within 24 hours to info requests from said point of contact and having the ability to “pull” knowledgeable resources into the conversation and solution at short notice.

TIA

FB88

Thanks for updates @fatbloke88 

Answers in no particular order:

1: You can’t take any issue to Ofgem. They’re the industry regulator.

If you exhaust the complaints procedure with OVO (unlikely), then you have the option to take your complaint to the Office of the Energy Ombudsman. OVO have a very high rate of complaints resolution, and they score highly on the  Ombudsman statistics for that aspect of customer service.

2: Repeating identical values within the Billing System usually indicate that there have been days of missing readings. When the next actual reading occurs, the usage is apportioned equally across the preceding days.

Note that this is not the same concept as an ‘estimate’. This is simply the mechanism used to assign genuine usage.

3: I’m going to defer to @Simon1D for comment on the different date format being used at the beginning of Oct’20. He’s been actively researching the issue of ‘missing data’ and may have observed such a phenomenon before.

It’s a bit too easy to assume that those 9 days where the figure was negative are to correct the obviously incorrect consumption reading of 2526.1 kWh for 30set20. But the maths doesn’t work, as I assume you already realise.